SlideShare a Scribd company logo
1 of 58
1
Arabic Handwritten ScriptArabic Handwritten Script
Recognition TowardsRecognition Towards
Generalization: A SurveyGeneralization: A SurveyAuthors:Authors:
 Randa I. M. ElanwarRanda I. M. Elanwar
Assistant Researcher, Electronic Research Institute
 Prof. Dr. Mohsen A. A. RashwanProf. Dr. Mohsen A. A. Rashwan
Professor of Digital Signal Processing, Electronic and communication dept, Cairo University
 Prof. Dr. Samia A. A. MashaliProf. Dr. Samia A. A. Mashali
Head of computers and systems dept, Electronic Research Institute
2
Presentation ContentsPresentation Contents
Introduction
Paper Objective
Arabic handwriting recognition problem
Main Challenges
Recent off-line Arabic handwriting recognition systems
Recent on-line Arabic handwriting recognition systems
Summary and Conclusion
3
IntroductionIntroduction
Handwriting recognition can be defined as the task
of transforming text represented in the spatial form
of graphical marks into its symbolic representation
The main components of a recognizer are:
1. Capturing Data & acquisition
2. Preprocessing & segmentation
3. Defining patterns and model selection
4. Feature Extraction
5. Training
6. Classification
4
IntroductionIntroduction
• First the input device captures an image and convert
it to a usable format
• Data is then preprocessed to eliminate noise for
simplification without loosing relevant information
and may also be segmented to smaller data units
5
IntroductionIntroduction
• The information of each data unit is sent to feature
extractor to reduce them by measuring certain
“features” or “properties”
• Patterns (or classes) should be defined and models
should be selected. These models are trained using
the extracted features.
6
IntroductionIntroduction
• The model for a pattern may be a single specific set
of features
• To recognize (or classify) a novel pattern means to
recover the model that generated the pattern based
on the extracted features
7
IntroductionIntroduction
The feature extractor has reduced the data unit to a
point or feature vector X in a 2D feature space (or
observation space)
Classification rule: Classify the input as Class I if its
feature vector falls below the decision boundary shown,
and as Class II otherwise.
8
IntroductionIntroduction
The problem is that designing a very complex
recognizer is unlikely to give good generalization since it
seems to be “tuned” to the particular training samples
The question is how to optimize this tradeoff:
generalization versus simple classifier
9
IntroductionIntroduction
Usually there is an action taken based on the
classification decision. Each action should be assigned a
certain cost.
We design our decision boundary (classification rule)
so that on the average, the Risk will be as small as
possible.
The Risk (R) is the expected value of cost
Minimizing (R) leads to complex boundaries
The question is how to optimize this tradeoff:
generalization versus minimum risk?
10
IntroductionIntroduction
In order to achieve general purpose recognizer
(unbiased) we should have a sufficient number of
training samples (N) for each class in the data set.
A theoretical estimate claims that
N ≅ 100 / P where P ≡ prob. of misclassification
I.e., for P ≈ 0.01, N ≈ 10000 and for P ≈ 0.03, N ≈ 3000
Such large data set (if available) needs large storage
and long processing time (time complexity)
The question is how to optimize this tradeoff:
generalization versus complexity?
11
Paper ObjectivePaper Objective
Our concern in this paper is to:
1. provide a comprehensive review of recent off-line
and on-line trends in Arabic cursive handwriting
recognition (last 10 years publications)
2. clarify the challenges standing against obtaining a
reliable, accurate, simple, general purpose recognizer
based on these trends.
12
Arabic Handwriting Recognition ProblemArabic Handwriting Recognition Problem
Arabic Script Recognition Systems are categorized as:
1. On-line or Off-line
2. Writer Dependent or Writer Independent
3. Open-vocabulary or closed-vocabulary
13
Arabic Handwriting Recognition ProblemArabic Handwriting Recognition Problem
Types of Recognition:
When the input device is a digitizer tablet that
transmits the signal in real time or includes timing
information together with pen position, this is mostly
referred to as on-line or dynamic recognition
14
Arabic Handwriting Recognition ProblemArabic Handwriting Recognition Problem
Types of Recognition:
When the input device is a still camera or a scanner,
which captures the position of digital ink on the page
but not the order in which it was laid down, this is
defined as off-line or image-based OCR
15
Arabic Handwriting Recognition ProblemArabic Handwriting Recognition Problem
Special Characteristics of Arabic Script:
Always written from right to left
Arabic word consists of one or more portions; each
has one or more characters
Many characters differ only by the position and the
number of dots attached
16
Arabic Handwriting Recognition ProblemArabic Handwriting Recognition Problem
Special Characteristics of Arabic Script:
Every character has more than one shape, depending
on its position
Characters overlap
17
Arabic Handwriting Recognition ProblemArabic Handwriting Recognition Problem
Special Characteristics of Arabic Script:
Existence of ligatures
Due to having these special characteristics, Arabic
handwriting recognition systems still need more research
to be established commercially
18
Main ChallengesMain Challenges
Feature Extraction
Noise
Model Selection and Complexity
Segmentation
Context
Evidence Pooling
Costs and Risks
Computational Complexity
Learning and Adaptation
19
Main ChallengesMain Challenges
Feature Extraction:
A good feature set should helps distinguishing a class
from other classes, be invariant to differences and
contains no redundant information
20
Main ChallengesMain Challenges
Feature Extraction:
A good feature set should helps distinguishing a class
from other classes, be invariant to differences and
contains no redundant information
… How to know which features are most
promising ?
… Is there ways to automatically learn which features are
best for a classifier?
21
Main ChallengesMain Challenges
Feature Extraction:
A good feature set should helps distinguishing a class
from other classes, be invariant to differences and
contains no redundant information
… How to know which features are most
promising ?
… Is there ways to automatically learn which features are
best for a classifier?
It should be limited in number for computational ease
and to limit the amount of training data
22
Main ChallengesMain Challenges
Feature Extraction:
A good feature set should helps distinguishing a class
from other classes, be invariant to differences and
contains no redundant information
… How to know which features are most
promising ?
… Is there ways to automatically learn which features are
best for a classifier?
It should be limited in number for computational ease
and to limit the amount of training data
… How many features
to use?
23
Main ChallengesMain Challenges
Noise:
Random error in a pixel value (deformation) due to
signal-independent, signal-dependent and salt &
pepper noise.
Noise cannot always be totally eliminated; but
smoothing is done
24
Main ChallengesMain Challenges
Noise:
Random error in a pixel value (deformation) due to
signal-independent, signal-dependent and salt &
pepper noise.
Noise cannot always be totally eliminated; but
smoothing is done
… Is the deformation in some signal is noise? or natural
varieties in true models?
… How can we use this information to improve
our classifier?
25
Main ChallengesMain Challenges
Modeling Selection and Complexity:
Determining the complexity of the model: not so
simple that it cannot explain the differences between
the categories, yet not so complex as to give poor
classification on novel patterns.
26
Main ChallengesMain Challenges
Modeling Selection and Complexity:
Determining the complexity of the model: not so
simple that it cannot explain the differences between
the categories, yet not so complex as to give poor
classification on novel patterns.
… how to know when to reject a class of models and
try another one?
… Are there principled methods for finding the best
complexity for a classifier?
… Is it a matter of random trial & error not even guided by
expectations of performance?
27
Main ChallengesMain Challenges
Segmentation:
Segmentation subdivides image into its constituent
regions or objects. Segmentation should stop when the
objects of interest in an application have been isolated.
28
Main ChallengesMain Challenges
Segmentation:
Segmentation subdivides image into its constituent
regions or objects. Segmentation should stop when the
objects of interest in an application have been isolated.
… How do we know where one character “ends” and the
next one “begin”?
… Shall we segment the images before they have been categorized or
categorize them
before they have been segmented?
29
Main ChallengesMain Challenges
Context:
The accuracy of automatic handwriting recognition
systems based on purely visual information seems to
have a ceiling
Incorporating Symantec and syntactic knowledge
sources into the automatic recognition of text can offer
potential improvements in performance
… how, precisely, should we incorporate such
information?
30
Main ChallengesMain Challenges
Evidence Pooling:
For high classification performance or for increased
class coverage, different classification tools are
developed either in parallel or sequentially
When having several component classifiers, and
these categorizers agree on a particular pattern, there
is no difficulty. But suppose they disagree !!!
31
Main ChallengesMain Challenges
Evidence Pooling:
For high classification performance or for increased
class coverage, different classification tools are
developed either in parallel or sequentially
When having several component classifiers, and
these categorizers agree on a particular pattern, there
is no difficulty. But suppose they disagree !!!
… How should a “super” classifier pool the evidence from the component
recognizers to achieve the best decision?
… How would the “super” categorizer know when to base a decision on
a minority opinion when required?
32
Main ChallengesMain Challenges
Costs and Risks:
A classifier is generally used to recommend actions,
each action having an associated cost or risk
We often design our classifier to recommend actions
that minimize some total expected cost or risk
33
Main ChallengesMain Challenges
Costs and Risks:
A classifier is generally used to recommend actions,
each action having an associated cost or risk
We often design our classifier to recommend actions
that minimize some total expected cost or risk
… How do we incorporate knowledge about such risks and how will they
affect the classification decision?
… Is there a way to estimate the total risk and thus tell whether our
classifier is acceptable even before we field it?
34
Main ChallengesMain Challenges
Computational Complexity:
Although we might achieve error-free recognition, the
time & storage requirements would be quite prohibitive
Some pattern recognition problems can be solved
using algorithms that are highly impractical.
35
Main ChallengesMain Challenges
Computational Complexity:
Although we might achieve error-free recognition, the
time & storage requirements would be quite prohibitive
Some pattern recognition problems can be solved
using algorithms that are highly impractical.
… What is the tradeoff between computational ease
and performance?
… How can we optimize an excellent recognizer within the
engineering constraints ?
36
Main ChallengesMain Challenges
Learning and Adaptation:
Any method that incorporates information from training
samples in the design of a classifier employs learning
If the models were extremely complicated, the classifier
would have complex decision boundaries
To overcome this, more training samples are needed to
obtain a better estimate of the true underlying features
In case of limited training samples, we should incorporate
knowledge of the problem domain. The production
representation is the “best” representation for classification.
37
Main ChallengesMain Challenges
Learning and Adaptation:
Any method that incorporates information from training
samples in the design of a classifier employs learning
If the models were extremely complicated, the classifier
would have complex decision boundaries
To overcome this, more training samples are needed to
obtain a better estimate of the true underlying features
In case of limited training samples, we should incorporate
knowledge of the problem domain. The production
representation is the “best” representation for classification.
… How much training samples are needed for good generalization?
… How can we insure that the learning algorithm favors “simple”
solutions rather than complicated ones?
38
Recent off-line Arabic handwriting recognitionRecent off-line Arabic handwriting recognition
systemssystems
Example: Pechwitz et al research [17]
proposed a recognition system based on a semi-continuous 1-D
HMM using the IFN/ENIT database of handwritten Tunisian
town/village names.
Preprocessing:
1. Extracting image contour and Performing a noise reduction filtering.
2. Skeletonization and normalization are performed.
3. Baseline estimation and word length normalization are performed.
39
Recent off-line Arabic handwriting recognitionRecent off-line Arabic handwriting recognition
systemssystems
Example: Pechwitz et al research [17]
Feature Extraction:
1. A rectangular window is shifted from right to left across the
normalized gray level script image .
2. A Loeve-Karhunen Transformation is performed on the gray values
of each frame to reduce the number of features.
Modeling:
1. A HMM-model is generated for each character shape (all possible
positions) up to 160 different HMM-models.
2. Semi Continuous HMMs are used with 7 states per character.
40
Recent off-line Arabic handwriting recognitionRecent off-line Arabic handwriting recognition
systemssystems
Example: Pechwitz et al research [17]
Database:
1. This database is split into four sets A, B, C & D.
2. The 4 sets contain 26,459 images of segmented Tunisian town
names (115,585 PAWs) handwritten by 411 unique writers.
3. 946 unique word labels, and 762 unique PAW labels.
4. For each image the ground truth information is available.
Lexicon:
The character shape HMM-models are combined to valid word
models using a tree structured lexicon with all 946 different
41
Recent off-line Arabic handwriting recognitionRecent off-line Arabic handwriting recognition
systemssystems
Example: Pechwitz et al research [17]
Recognition:
The standard Viterbi Algorithm is used together with the lexicon.
The authors applied the recognition algorithm to the database
twice, once using the baseline coming from GT (ground truth) and
once using baseline they estimated.
Results:
Recognition rates 82 – 89% are obtained using baseline estimation
Recognition rates 89 – 95% are obtained using GT baseline
42
Recent off-line Arabic handwriting recognitionRecent off-line Arabic handwriting recognition
systemssystems
Example: Pechwitz et al research [17]
Challenges:
1. Working on available database skips the limited training samples challenge
43
Recent off-line Arabic handwriting recognitionRecent off-line Arabic handwriting recognition
systemssystems
Example: Pechwitz et al research [17]
Challenges:
1. Working on available database skips the limited training samples challenge
2. It is not easy to generalize this classifier for open vocabulary applications
because it works on a limited lexicon of words (segmentation-free
recognizer) otherwise context will be a must.
44
Recent off-line Arabic handwriting recognitionRecent off-line Arabic handwriting recognition
systemssystems
Example: Pechwitz et al research [17]
Challenges:
1. Working on available database skips the limited training samples challenge
2. It is not easy to generalize this classifier for open vocabulary applications
because it works on a limited lexicon of words (segmentation-free
recognizer) otherwise context will be a must.
3. Generating the same HMM structure for all characters and ligatures i.e.,
modeling selection & complexity .. we think it would be much better to vary
the model structure according to each character requirement (‫ض‬ shouldn’t
have the same model as ‫ة‬ for example).
45
Recent off-line Arabic handwriting recognitionRecent off-line Arabic handwriting recognition
systemssystems
Example: Pechwitz et al research [17]
Challenges:
1. Working on available database skips the limited training samples challenge
2. It is not easy to generalize this classifier for open vocabulary applications
because it works on a limited lexicon of words (segmentation-free
recognizer) otherwise context will be a must.
3. Generating the same HMM structure for all characters and ligatures i.e.,
modeling selection & complexity .. we think it would be much better to vary
the model structure according to each character requirement (‫ض‬ shouldn’t
have the same model as ‫ة‬ for example).
4. Feature Extraction: The idea of normalizing the word width to use a sliding
window feature extractor is pretty good except for the great dependency on
46
Recent on-line Arabic handwriting recognitionRecent on-line Arabic handwriting recognition
systemssystems
Example: Biadsy et al research [24]
Preprocessing:
1. Geometrical processing phase to minimize handwriting variations.
2. A low-pass filter is used to reduce noise and remove imperfections
caused by acquisition devices.
3. The writing-speed is normalized by re-sampling the consequent
point sequences.
Feature Extraction:
Mainly angles (with x-axis) and loop-presence
47
Recent on-line Arabic handwriting recognitionRecent on-line Arabic handwriting recognition
systemssystems
Example: Biadsy et al research [24]
Modeling:
1. The recognition framework uses discrete Left-to-right HMMs to
represent each Arabic letter shape (isolated, initial, medial, and
final).
2. The number of states for each letter shape model is based on the
geometric complexity of the letter shape. It varies from 5 to 11
states.
For example: 11 states are assigned to isolated ‫,ش‬ and 5 states to
isolated ‫.أ‬
48
Recent on-line Arabic handwriting recognitionRecent on-line Arabic handwriting recognition
systemssystems
Example: Biadsy et al research [24]
Lexicon:
1. The Arabic dictionary D is subdivided into a set of sub-dictionaries {D1, D2,
…, Dn} based on the number of word parts in each word.
2. Letter-shape models are embedded in a network that represents a word-
part dictionary. The segmentation of word parts into letter-shapes and their
recognition are performed simultaneously in an integrated process.
D = {D = {‫انسان‬ ،‫التحدى‬ ،‫ثقافة‬ ،‫جامعة‬ ،‫رواية‬ ،‫فادى‬ ،‫محمد‬ ،‫محمود‬ ،‫معلم‬ ،‫هل‬ ،‫وسام‬‫انسان‬ ،‫التحدى‬ ،‫ثقافة‬ ،‫جامعة‬ ،‫رواية‬ ،‫فادى‬ ،‫محمد‬ ،‫محمود‬ ،‫معلم‬ ،‫هل‬ ،‫وسام‬}}
Sub-dictionaries of DSub-dictionaries of D Word-Part Dictionary for D3Word-Part Dictionary for D3
D1 = {D1 = {‫محمد‬ ،‫معلم‬ ،‫هل‬‫محمد‬ ،‫معلم‬ ،‫هل‬}}
D2 = {D2 = {‫ثقافة‬ ،‫جامعة‬ ،‫محمود‬‫ثقافة‬ ،‫جامعة‬ ،‫محمود‬}}
D3 = {D3 = {‫انسان‬ ،‫التحدى‬ ،‫فادى‬ ،‫وسام‬‫انسان‬ ،‫التحدى‬ ،‫فادى‬ ،‫وسام‬}}
D4 = {D4 = {‫رواية‬‫رواية‬}}
WPD3,1 = {WPD3,1 = {‫ا‬ ،‫فا‬ ،‫و‬‫ا‬ ،‫فا‬ ،‫و‬}}
WPD3,2 = {WPD3,2 = {‫نسا‬ ،‫لتحد‬ ،‫د‬ ،‫سا‬‫نسا‬ ،‫لتحد‬ ،‫د‬ ،‫سا‬}}
WPD3,3 = {WPD3,3 = {‫ن‬ ،‫ى‬ ،‫م‬‫ن‬ ،‫ى‬ ،‫م‬}}
49
Recent on-line Arabic handwriting recognitionRecent on-line Arabic handwriting recognition
systemssystems
Example: Biadsy et al research [24]
Database:
1. 4 trainers are asked to write 800 selected words each.
2. For testing, 10 testers (the 4 trainers, in addition to 6 new volunteers) are
asked to write 280 words not in the training data (2,358 words in total).
3. 5 different dictionary sizes (5K, 10K, 20K, 30K, and 40K words) selected
from different Arabic websites are used. The 280 test words are present in
all dictionary sizes.
Recognition:
Writer dependent (WD) and writer independent (WI) experiments are done
and average word recognition rates 88 – 96% are obtained. The
50
Recent on-line Arabic handwriting recognitionRecent on-line Arabic handwriting recognition
systemssystems
Example: Biadsy et al research [24]
Challenges:
1. Feature Extraction: The features they use are not enough to lead to
satisfying classification of general unconstrained handwritings.
Thus they are in a great need to work under limited vocabulary.
The word parts must be present in the dictionary or the will not be
recognized.
51
Recent on-line Arabic handwriting recognitionRecent on-line Arabic handwriting recognition
systemssystems
Example: Biadsy et al research [24]
Challenges:
1. Feature Extraction: The features they use are not enough to lead to
satisfying classification of general unconstrained handwritings.
Thus they are in a great need to work under limited vocabulary.
The word parts must be present in the dictionary or the will not be
recognized.
2. Database they use looks unnatural. Volunteers are asked to follow
restrict methodology of writing which affects their individual writing
style. Besides, the system handles limited handwriting varieties
due to the small number of volunteers who wrote the database.
52
Summary and ConclusionSummary and Conclusion
Foreign recognizers have found their way to the
markets as commercial products since years while
Arabic recognizers still need more time.
53
Summary and ConclusionSummary and Conclusion
Foreign recognizers have found their way to the
markets as commercial products since years while
Arabic recognizers still need more time.
in the case of Arabic handwritten words many
researchers use a specific, more or less small data set
of their own ∴ it is impossible to compare different
results which would be important to improve existent
methods
54
Summary and ConclusionSummary and Conclusion
Foreign recognizers have found their way to the
markets as commercial products since years while
Arabic recognizers still need more time.
in the case of Arabic handwritten words many
researchers use a specific, more or less small data set
of their own ∴ it is impossible to compare different
results which would be important to improve existent
methods
The complexity of the problem is greatly increased by
noise and by the infinite variability of handwritings
55
Summary and ConclusionSummary and Conclusion
Cursive script requires the segmentation of words in
characters or parts of characters, i.e. graphemes, and
then the detection of individual features.
56
Summary and ConclusionSummary and Conclusion
Cursive script requires the segmentation of words in
characters or parts of characters, i.e. graphemes, and
then the detection of individual features.
Generally, the holistic approach can be used if the
size of the vocabulary is small (such as the recognition
of the legal amount in cheques)
57
Summary and ConclusionSummary and Conclusion
Cursive script requires the segmentation of words in
characters or parts of characters, i.e. graphemes, and
then the detection of individual features.
Generally, the holistic approach can be used if the
size of the vocabulary is small (such as the recognition
of the legal amount in cheques)
The character-based approach is the preferred
method for recognition applications that are
unconstrained or involve large-size vocabularies to
insure good generalization together with reasonable
complexity
58
ThankThank
YouYou

More Related Content

What's hot

Introduction to CNN
Introduction to CNNIntroduction to CNN
Introduction to CNNShuai Zhang
 
introduction to deep Learning with full detail
introduction to deep Learning with full detailintroduction to deep Learning with full detail
introduction to deep Learning with full detailsonykhan3
 
OCR Presentation (Optical Character Recognition)
OCR Presentation (Optical Character Recognition)OCR Presentation (Optical Character Recognition)
OCR Presentation (Optical Character Recognition)Neeraj Neupane
 
Recurrent Neural Networks for Text Analysis
Recurrent Neural Networks for Text AnalysisRecurrent Neural Networks for Text Analysis
Recurrent Neural Networks for Text Analysisodsc
 
Introduction of Deep Learning
Introduction of Deep LearningIntroduction of Deep Learning
Introduction of Deep LearningMyungjin Lee
 
Convolutional Neural Network (CNN) - image recognition
Convolutional Neural Network (CNN)  - image recognitionConvolutional Neural Network (CNN)  - image recognition
Convolutional Neural Network (CNN) - image recognitionYUNG-KUEI CHEN
 
Machine Learning - Convolutional Neural Network
Machine Learning - Convolutional Neural NetworkMachine Learning - Convolutional Neural Network
Machine Learning - Convolutional Neural NetworkRichard Kuo
 
Handwritten Digit Recognition(Convolutional Neural Network) PPT
Handwritten Digit Recognition(Convolutional Neural Network) PPTHandwritten Digit Recognition(Convolutional Neural Network) PPT
Handwritten Digit Recognition(Convolutional Neural Network) PPTRishabhTyagi48
 
Optical Character Recognition (OCR) based Retrieval
Optical Character Recognition (OCR) based RetrievalOptical Character Recognition (OCR) based Retrieval
Optical Character Recognition (OCR) based RetrievalBiniam Asnake
 
OCR (Optical Character Recognition)
OCR (Optical Character Recognition) OCR (Optical Character Recognition)
OCR (Optical Character Recognition) IstiaqueBinIslam
 
Deep Learning - RNN and CNN
Deep Learning - RNN and CNNDeep Learning - RNN and CNN
Deep Learning - RNN and CNNPradnya Saval
 
Deep learning - A Visual Introduction
Deep learning - A Visual IntroductionDeep learning - A Visual Introduction
Deep learning - A Visual IntroductionLukas Masuch
 
[251] implementing deep learning using cu dnn
[251] implementing deep learning using cu dnn[251] implementing deep learning using cu dnn
[251] implementing deep learning using cu dnnNAVER D2
 
Deep Learning Introduction Lecture
Deep Learning Introduction LectureDeep Learning Introduction Lecture
Deep Learning Introduction Lectureshivam chaurasia
 

What's hot (20)

cnn ppt.pptx
cnn ppt.pptxcnn ppt.pptx
cnn ppt.pptx
 
Introduction to CNN
Introduction to CNNIntroduction to CNN
Introduction to CNN
 
introduction to deep Learning with full detail
introduction to deep Learning with full detailintroduction to deep Learning with full detail
introduction to deep Learning with full detail
 
Deep learning
Deep learning Deep learning
Deep learning
 
OCR Presentation (Optical Character Recognition)
OCR Presentation (Optical Character Recognition)OCR Presentation (Optical Character Recognition)
OCR Presentation (Optical Character Recognition)
 
Recurrent Neural Networks for Text Analysis
Recurrent Neural Networks for Text AnalysisRecurrent Neural Networks for Text Analysis
Recurrent Neural Networks for Text Analysis
 
Introduction of Deep Learning
Introduction of Deep LearningIntroduction of Deep Learning
Introduction of Deep Learning
 
Convolutional Neural Network (CNN) - image recognition
Convolutional Neural Network (CNN)  - image recognitionConvolutional Neural Network (CNN)  - image recognition
Convolutional Neural Network (CNN) - image recognition
 
Machine Learning - Convolutional Neural Network
Machine Learning - Convolutional Neural NetworkMachine Learning - Convolutional Neural Network
Machine Learning - Convolutional Neural Network
 
Deep Learning
Deep Learning Deep Learning
Deep Learning
 
Handwritten Digit Recognition(Convolutional Neural Network) PPT
Handwritten Digit Recognition(Convolutional Neural Network) PPTHandwritten Digit Recognition(Convolutional Neural Network) PPT
Handwritten Digit Recognition(Convolutional Neural Network) PPT
 
Neural networks
Neural networksNeural networks
Neural networks
 
Optical Character Recognition (OCR) based Retrieval
Optical Character Recognition (OCR) based RetrievalOptical Character Recognition (OCR) based Retrieval
Optical Character Recognition (OCR) based Retrieval
 
OCR (Optical Character Recognition)
OCR (Optical Character Recognition) OCR (Optical Character Recognition)
OCR (Optical Character Recognition)
 
Deep Learning - RNN and CNN
Deep Learning - RNN and CNNDeep Learning - RNN and CNN
Deep Learning - RNN and CNN
 
Deep learning - A Visual Introduction
Deep learning - A Visual IntroductionDeep learning - A Visual Introduction
Deep learning - A Visual Introduction
 
[251] implementing deep learning using cu dnn
[251] implementing deep learning using cu dnn[251] implementing deep learning using cu dnn
[251] implementing deep learning using cu dnn
 
Deep Learning Introduction Lecture
Deep Learning Introduction LectureDeep Learning Introduction Lecture
Deep Learning Introduction Lecture
 
Text Detection and Recognition
Text Detection and RecognitionText Detection and Recognition
Text Detection and Recognition
 
Neural networks introduction
Neural networks introductionNeural networks introduction
Neural networks introduction
 

Viewers also liked

Off-line English Character Recognition: A Comparative Survey
Off-line English Character Recognition: A Comparative SurveyOff-line English Character Recognition: A Comparative Survey
Off-line English Character Recognition: A Comparative Surveyidescitation
 
Optical character recognition of handwritten Arabic using hidden Markov models
Optical character recognition of handwritten Arabic using hidden Markov modelsOptical character recognition of handwritten Arabic using hidden Markov models
Optical character recognition of handwritten Arabic using hidden Markov modelsMuhannad Aulama
 
A Semi-Automatic Annotation Tool For Arabic Online Handwritten Text
A Semi-Automatic Annotation Tool For Arabic Online Handwritten TextA Semi-Automatic Annotation Tool For Arabic Online Handwritten Text
A Semi-Automatic Annotation Tool For Arabic Online Handwritten TextRanda Elanwar
 
A simple text ine segmentation method for handwritten documents1
A simple text ine segmentation method for handwritten documents1A simple text ine segmentation method for handwritten documents1
A simple text ine segmentation method for handwritten documents1Prasad Babu
 
Word segmentation method for handwritten documents based on structured learning
Word segmentation method for handwritten documents based on structured learningWord segmentation method for handwritten documents based on structured learning
Word segmentation method for handwritten documents based on structured learningI3E Technologies
 
Holistic Approach for Arabic Word Recognition
Holistic Approach for Arabic Word RecognitionHolistic Approach for Arabic Word Recognition
Holistic Approach for Arabic Word RecognitionEditor IJCATR
 
ICFHR 2014 Competition on Handwritten KeyWord Spotting (H-KWS 2014)
ICFHR 2014 Competition on Handwritten KeyWord Spotting (H-KWS 2014)ICFHR 2014 Competition on Handwritten KeyWord Spotting (H-KWS 2014)
ICFHR 2014 Competition on Handwritten KeyWord Spotting (H-KWS 2014)Konstantinos Zagoris
 
Segmentation - based Historical Handwritten Word Spotting using document-spec...
Segmentation - based Historical Handwritten Word Spotting using document-spec...Segmentation - based Historical Handwritten Word Spotting using document-spec...
Segmentation - based Historical Handwritten Word Spotting using document-spec...Konstantinos Zagoris
 
Online handwritten script recognition
Online handwritten script recognitionOnline handwritten script recognition
Online handwritten script recognitionDhiraj Singh
 
Performance of Statistics Based Line Segmentation System for Unconstrained H...
Performance of Statistics Based Line Segmentation  System for Unconstrained H...Performance of Statistics Based Line Segmentation  System for Unconstrained H...
Performance of Statistics Based Line Segmentation System for Unconstrained H...AM Publications
 
Scalability in Model Checking through Relational Databases
Scalability in Model Checking through Relational DatabasesScalability in Model Checking through Relational Databases
Scalability in Model Checking through Relational DatabasesCSCJournals
 
Stages of image processing
Stages of image processingStages of image processing
Stages of image processingAmal Mp
 
Usage of Shape From Focus Method For 3D Shape Recovery And Identification of ...
Usage of Shape From Focus Method For 3D Shape Recovery And Identification of ...Usage of Shape From Focus Method For 3D Shape Recovery And Identification of ...
Usage of Shape From Focus Method For 3D Shape Recovery And Identification of ...CSCJournals
 
Feature Analysis for Affect Recognition Supporting Task Sequencing in Adaptiv...
Feature Analysis for Affect Recognition Supporting Task Sequencing in Adaptiv...Feature Analysis for Affect Recognition Supporting Task Sequencing in Adaptiv...
Feature Analysis for Affect Recognition Supporting Task Sequencing in Adaptiv...janningr
 
Image Enhancement by Image Fusion for Crime Investigation
Image Enhancement by Image Fusion for Crime InvestigationImage Enhancement by Image Fusion for Crime Investigation
Image Enhancement by Image Fusion for Crime InvestigationCSCJournals
 
Fourth Dimension Level 1 By Dr.Moiz Hussain
Fourth Dimension Level 1  By Dr.Moiz HussainFourth Dimension Level 1  By Dr.Moiz Hussain
Fourth Dimension Level 1 By Dr.Moiz HussainEhtesham Mirxa
 
Improving the Accuracy of Object Based Supervised Image Classification using ...
Improving the Accuracy of Object Based Supervised Image Classification using ...Improving the Accuracy of Object Based Supervised Image Classification using ...
Improving the Accuracy of Object Based Supervised Image Classification using ...CSCJournals
 
Video Key-Frame Extraction using Unsupervised Clustering and Mutual Comparison
Video Key-Frame Extraction using Unsupervised Clustering and Mutual ComparisonVideo Key-Frame Extraction using Unsupervised Clustering and Mutual Comparison
Video Key-Frame Extraction using Unsupervised Clustering and Mutual ComparisonCSCJournals
 

Viewers also liked (20)

Off-line English Character Recognition: A Comparative Survey
Off-line English Character Recognition: A Comparative SurveyOff-line English Character Recognition: A Comparative Survey
Off-line English Character Recognition: A Comparative Survey
 
Optical character recognition of handwritten Arabic using hidden Markov models
Optical character recognition of handwritten Arabic using hidden Markov modelsOptical character recognition of handwritten Arabic using hidden Markov models
Optical character recognition of handwritten Arabic using hidden Markov models
 
Spotting Customers in Trouble
Spotting Customers in TroubleSpotting Customers in Trouble
Spotting Customers in Trouble
 
A Semi-Automatic Annotation Tool For Arabic Online Handwritten Text
A Semi-Automatic Annotation Tool For Arabic Online Handwritten TextA Semi-Automatic Annotation Tool For Arabic Online Handwritten Text
A Semi-Automatic Annotation Tool For Arabic Online Handwritten Text
 
A simple text ine segmentation method for handwritten documents1
A simple text ine segmentation method for handwritten documents1A simple text ine segmentation method for handwritten documents1
A simple text ine segmentation method for handwritten documents1
 
Word segmentation method for handwritten documents based on structured learning
Word segmentation method for handwritten documents based on structured learningWord segmentation method for handwritten documents based on structured learning
Word segmentation method for handwritten documents based on structured learning
 
Holistic Approach for Arabic Word Recognition
Holistic Approach for Arabic Word RecognitionHolistic Approach for Arabic Word Recognition
Holistic Approach for Arabic Word Recognition
 
ICFHR 2014 Competition on Handwritten KeyWord Spotting (H-KWS 2014)
ICFHR 2014 Competition on Handwritten KeyWord Spotting (H-KWS 2014)ICFHR 2014 Competition on Handwritten KeyWord Spotting (H-KWS 2014)
ICFHR 2014 Competition on Handwritten KeyWord Spotting (H-KWS 2014)
 
Segmentation - based Historical Handwritten Word Spotting using document-spec...
Segmentation - based Historical Handwritten Word Spotting using document-spec...Segmentation - based Historical Handwritten Word Spotting using document-spec...
Segmentation - based Historical Handwritten Word Spotting using document-spec...
 
Online handwritten script recognition
Online handwritten script recognitionOnline handwritten script recognition
Online handwritten script recognition
 
Performance of Statistics Based Line Segmentation System for Unconstrained H...
Performance of Statistics Based Line Segmentation  System for Unconstrained H...Performance of Statistics Based Line Segmentation  System for Unconstrained H...
Performance of Statistics Based Line Segmentation System for Unconstrained H...
 
Scalability in Model Checking through Relational Databases
Scalability in Model Checking through Relational DatabasesScalability in Model Checking through Relational Databases
Scalability in Model Checking through Relational Databases
 
Stages of image processing
Stages of image processingStages of image processing
Stages of image processing
 
Usage of Shape From Focus Method For 3D Shape Recovery And Identification of ...
Usage of Shape From Focus Method For 3D Shape Recovery And Identification of ...Usage of Shape From Focus Method For 3D Shape Recovery And Identification of ...
Usage of Shape From Focus Method For 3D Shape Recovery And Identification of ...
 
Feature Analysis for Affect Recognition Supporting Task Sequencing in Adaptiv...
Feature Analysis for Affect Recognition Supporting Task Sequencing in Adaptiv...Feature Analysis for Affect Recognition Supporting Task Sequencing in Adaptiv...
Feature Analysis for Affect Recognition Supporting Task Sequencing in Adaptiv...
 
Image Enhancement by Image Fusion for Crime Investigation
Image Enhancement by Image Fusion for Crime InvestigationImage Enhancement by Image Fusion for Crime Investigation
Image Enhancement by Image Fusion for Crime Investigation
 
Fourth Dimension Level 1 By Dr.Moiz Hussain
Fourth Dimension Level 1  By Dr.Moiz HussainFourth Dimension Level 1  By Dr.Moiz Hussain
Fourth Dimension Level 1 By Dr.Moiz Hussain
 
Improving the Accuracy of Object Based Supervised Image Classification using ...
Improving the Accuracy of Object Based Supervised Image Classification using ...Improving the Accuracy of Object Based Supervised Image Classification using ...
Improving the Accuracy of Object Based Supervised Image Classification using ...
 
Building Identification in Satellite Images Using ANFIS Classifier
Building Identification in Satellite Images Using ANFIS ClassifierBuilding Identification in Satellite Images Using ANFIS Classifier
Building Identification in Satellite Images Using ANFIS Classifier
 
Video Key-Frame Extraction using Unsupervised Clustering and Mutual Comparison
Video Key-Frame Extraction using Unsupervised Clustering and Mutual ComparisonVideo Key-Frame Extraction using Unsupervised Clustering and Mutual Comparison
Video Key-Frame Extraction using Unsupervised Clustering and Mutual Comparison
 

Similar to Arabic Handwritten Script Recognition Towards Generalization: A Survey

AI and ML Skills for the Testing World Tutorial
AI and ML Skills for the Testing World TutorialAI and ML Skills for the Testing World Tutorial
AI and ML Skills for the Testing World TutorialTariq King
 
2013-1 Machine Learning Lecture 01 - Pattern Recognition
2013-1 Machine Learning Lecture 01 - Pattern Recognition2013-1 Machine Learning Lecture 01 - Pattern Recognition
2013-1 Machine Learning Lecture 01 - Pattern RecognitionDongseo University
 
Meetup 29042015
Meetup 29042015Meetup 29042015
Meetup 29042015lbishal
 
IRJET - Cognitive based Emotion Analysis of a Child Reading a Book
IRJET -  	  Cognitive based Emotion Analysis of a Child Reading a BookIRJET -  	  Cognitive based Emotion Analysis of a Child Reading a Book
IRJET - Cognitive based Emotion Analysis of a Child Reading a BookIRJET Journal
 
AI driven classification framework for advanced Test Automation
AI driven classification framework for advanced Test AutomationAI driven classification framework for advanced Test Automation
AI driven classification framework for advanced Test AutomationSTePINForum
 
Imtiaz khan data_science_analytics
Imtiaz khan data_science_analyticsImtiaz khan data_science_analytics
Imtiaz khan data_science_analyticsimtiaz khan
 
Artificial Intelligence & QA
Artificial Intelligence & QAArtificial Intelligence & QA
Artificial Intelligence & QAMalihaAshraf
 
Top 50 Accenture Interview Questions and Answers
Top 50 Accenture Interview Questions and AnswersTop 50 Accenture Interview Questions and Answers
Top 50 Accenture Interview Questions and AnswersSimplilearn
 
What are the Unique Challenges and Opportunities in Systems for ML?
What are the Unique Challenges and Opportunities in Systems for ML?What are the Unique Challenges and Opportunities in Systems for ML?
What are the Unique Challenges and Opportunities in Systems for ML?Matei Zaharia
 
Fixing the program my computer learned: End-user debugging of machine-learned...
Fixing the program my computer learned: End-user debugging of machine-learned...Fixing the program my computer learned: End-user debugging of machine-learned...
Fixing the program my computer learned: End-user debugging of machine-learned...City University London
 
Chapter 10 Testing and Quality Assurance1Unders.docx
Chapter 10 Testing and Quality Assurance1Unders.docxChapter 10 Testing and Quality Assurance1Unders.docx
Chapter 10 Testing and Quality Assurance1Unders.docxketurahhazelhurst
 
Face recognition technology
Face recognition technologyFace recognition technology
Face recognition technologyranjit banshpal
 
Empirical evaluation in 2020: how big, how beautiful?
Empirical evaluation in 2020: how big, how beautiful?Empirical evaluation in 2020: how big, how beautiful?
Empirical evaluation in 2020: how big, how beautiful?Massimiliano Di Penta
 
Machine Learning Interview Questions
Machine Learning Interview QuestionsMachine Learning Interview Questions
Machine Learning Interview QuestionsRock Interview
 
Strata London - Deep Learning 05-2015
Strata London - Deep Learning 05-2015Strata London - Deep Learning 05-2015
Strata London - Deep Learning 05-2015Turi, Inc.
 
Nss power point_machine_learning
Nss power point_machine_learningNss power point_machine_learning
Nss power point_machine_learningGauravsd2014
 
Frequently asked tcs technical interview questions and answers
Frequently asked tcs technical interview questions and answersFrequently asked tcs technical interview questions and answers
Frequently asked tcs technical interview questions and answersnishajj
 
UNIT 3.1 INTRODUCTON TO IDA.ppt
UNIT 3.1 INTRODUCTON TO IDA.pptUNIT 3.1 INTRODUCTON TO IDA.ppt
UNIT 3.1 INTRODUCTON TO IDA.pptManjuAppukuttan2
 

Similar to Arabic Handwritten Script Recognition Towards Generalization: A Survey (20)

AI and ML Skills for the Testing World Tutorial
AI and ML Skills for the Testing World TutorialAI and ML Skills for the Testing World Tutorial
AI and ML Skills for the Testing World Tutorial
 
2013-1 Machine Learning Lecture 01 - Pattern Recognition
2013-1 Machine Learning Lecture 01 - Pattern Recognition2013-1 Machine Learning Lecture 01 - Pattern Recognition
2013-1 Machine Learning Lecture 01 - Pattern Recognition
 
Meetup 29042015
Meetup 29042015Meetup 29042015
Meetup 29042015
 
IRJET - Cognitive based Emotion Analysis of a Child Reading a Book
IRJET -  	  Cognitive based Emotion Analysis of a Child Reading a BookIRJET -  	  Cognitive based Emotion Analysis of a Child Reading a Book
IRJET - Cognitive based Emotion Analysis of a Child Reading a Book
 
Infarec
InfarecInfarec
Infarec
 
AI driven classification framework for advanced Test Automation
AI driven classification framework for advanced Test AutomationAI driven classification framework for advanced Test Automation
AI driven classification framework for advanced Test Automation
 
Imtiaz khan data_science_analytics
Imtiaz khan data_science_analyticsImtiaz khan data_science_analytics
Imtiaz khan data_science_analytics
 
Artificial Intelligence & QA
Artificial Intelligence & QAArtificial Intelligence & QA
Artificial Intelligence & QA
 
Top 50 Accenture Interview Questions and Answers
Top 50 Accenture Interview Questions and AnswersTop 50 Accenture Interview Questions and Answers
Top 50 Accenture Interview Questions and Answers
 
What are the Unique Challenges and Opportunities in Systems for ML?
What are the Unique Challenges and Opportunities in Systems for ML?What are the Unique Challenges and Opportunities in Systems for ML?
What are the Unique Challenges and Opportunities in Systems for ML?
 
Python Project.pptx
Python Project.pptxPython Project.pptx
Python Project.pptx
 
Fixing the program my computer learned: End-user debugging of machine-learned...
Fixing the program my computer learned: End-user debugging of machine-learned...Fixing the program my computer learned: End-user debugging of machine-learned...
Fixing the program my computer learned: End-user debugging of machine-learned...
 
Chapter 10 Testing and Quality Assurance1Unders.docx
Chapter 10 Testing and Quality Assurance1Unders.docxChapter 10 Testing and Quality Assurance1Unders.docx
Chapter 10 Testing and Quality Assurance1Unders.docx
 
Face recognition technology
Face recognition technologyFace recognition technology
Face recognition technology
 
Empirical evaluation in 2020: how big, how beautiful?
Empirical evaluation in 2020: how big, how beautiful?Empirical evaluation in 2020: how big, how beautiful?
Empirical evaluation in 2020: how big, how beautiful?
 
Machine Learning Interview Questions
Machine Learning Interview QuestionsMachine Learning Interview Questions
Machine Learning Interview Questions
 
Strata London - Deep Learning 05-2015
Strata London - Deep Learning 05-2015Strata London - Deep Learning 05-2015
Strata London - Deep Learning 05-2015
 
Nss power point_machine_learning
Nss power point_machine_learningNss power point_machine_learning
Nss power point_machine_learning
 
Frequently asked tcs technical interview questions and answers
Frequently asked tcs technical interview questions and answersFrequently asked tcs technical interview questions and answers
Frequently asked tcs technical interview questions and answers
 
UNIT 3.1 INTRODUCTON TO IDA.ppt
UNIT 3.1 INTRODUCTON TO IDA.pptUNIT 3.1 INTRODUCTON TO IDA.ppt
UNIT 3.1 INTRODUCTON TO IDA.ppt
 

More from Randa Elanwar

الجزء السادس ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء السادس ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةالجزء السادس ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء السادس ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةRanda Elanwar
 
الجزء الخامس ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الخامس ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةالجزء الخامس ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الخامس ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةRanda Elanwar
 
الجزء الرابع ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الرابع ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةالجزء الرابع ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الرابع ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةRanda Elanwar
 
الجزء الثالث ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الثالث ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةالجزء الثالث ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الثالث ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةRanda Elanwar
 
الجزء الثاني ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الثاني ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةالجزء الثاني ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الثاني ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةRanda Elanwar
 
الجزء الأول ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الأول ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةالجزء الأول ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الأول ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةRanda Elanwar
 
تدريب مدونة علماء مصر على الكتابة الفنية (الترجمة والتلخيص )_Pdf5of5
تدريب مدونة علماء مصر على الكتابة الفنية (الترجمة والتلخيص    )_Pdf5of5تدريب مدونة علماء مصر على الكتابة الفنية (الترجمة والتلخيص    )_Pdf5of5
تدريب مدونة علماء مصر على الكتابة الفنية (الترجمة والتلخيص )_Pdf5of5Randa Elanwar
 
تدريب مدونة علماء مصر على الكتابة الفنية (القصة القصيرة والخاطرة والأخطاء ال...
تدريب مدونة علماء مصر على الكتابة الفنية (القصة القصيرة والخاطرة  والأخطاء ال...تدريب مدونة علماء مصر على الكتابة الفنية (القصة القصيرة والخاطرة  والأخطاء ال...
تدريب مدونة علماء مصر على الكتابة الفنية (القصة القصيرة والخاطرة والأخطاء ال...Randa Elanwar
 
تدريب مدونة علماء مصر على الكتابة الفنية (مقالات الموارد )_Pdf3of5
تدريب مدونة علماء مصر على الكتابة الفنية (مقالات الموارد   )_Pdf3of5تدريب مدونة علماء مصر على الكتابة الفنية (مقالات الموارد   )_Pdf3of5
تدريب مدونة علماء مصر على الكتابة الفنية (مقالات الموارد )_Pdf3of5Randa Elanwar
 
تدريب مدونة علماء مصر على الكتابة الفنية (المقالات الإخبارية )_Pdf2of5
تدريب مدونة علماء مصر على الكتابة الفنية (المقالات الإخبارية  )_Pdf2of5تدريب مدونة علماء مصر على الكتابة الفنية (المقالات الإخبارية  )_Pdf2of5
تدريب مدونة علماء مصر على الكتابة الفنية (المقالات الإخبارية )_Pdf2of5Randa Elanwar
 
تدريب مدونة علماء مصر على الكتابة الفنية (المقالات المبنية على البحث )_Pdf1of5
تدريب مدونة علماء مصر على الكتابة الفنية (المقالات المبنية على البحث )_Pdf1of5تدريب مدونة علماء مصر على الكتابة الفنية (المقالات المبنية على البحث )_Pdf1of5
تدريب مدونة علماء مصر على الكتابة الفنية (المقالات المبنية على البحث )_Pdf1of5Randa Elanwar
 
تعريف بمدونة علماء مصر ومحاور التدريب على الكتابة للمدونين
تعريف بمدونة علماء مصر ومحاور التدريب على الكتابة للمدونينتعريف بمدونة علماء مصر ومحاور التدريب على الكتابة للمدونين
تعريف بمدونة علماء مصر ومحاور التدريب على الكتابة للمدونينRanda Elanwar
 
Entrepreneurship_who_is_your_customer_(arabic)_7of7
Entrepreneurship_who_is_your_customer_(arabic)_7of7Entrepreneurship_who_is_your_customer_(arabic)_7of7
Entrepreneurship_who_is_your_customer_(arabic)_7of7Randa Elanwar
 
Entrepreneurship_who_is_your_customer_(arabic)_5of7
Entrepreneurship_who_is_your_customer_(arabic)_5of7Entrepreneurship_who_is_your_customer_(arabic)_5of7
Entrepreneurship_who_is_your_customer_(arabic)_5of7Randa Elanwar
 
Entrepreneurship_who_is_your_customer_(arabic)_4of7
Entrepreneurship_who_is_your_customer_(arabic)_4of7Entrepreneurship_who_is_your_customer_(arabic)_4of7
Entrepreneurship_who_is_your_customer_(arabic)_4of7Randa Elanwar
 
Entrepreneurship_who_is_your_customer_(arabic)_2of7
Entrepreneurship_who_is_your_customer_(arabic)_2of7Entrepreneurship_who_is_your_customer_(arabic)_2of7
Entrepreneurship_who_is_your_customer_(arabic)_2of7Randa Elanwar
 
يوميات طالب بدرجة مشرف (Part 19 of 20)
يوميات طالب بدرجة مشرف (Part 19 of 20)يوميات طالب بدرجة مشرف (Part 19 of 20)
يوميات طالب بدرجة مشرف (Part 19 of 20)Randa Elanwar
 
يوميات طالب بدرجة مشرف (Part 18 of 20)
يوميات طالب بدرجة مشرف (Part 18 of 20)يوميات طالب بدرجة مشرف (Part 18 of 20)
يوميات طالب بدرجة مشرف (Part 18 of 20)Randa Elanwar
 
يوميات طالب بدرجة مشرف (Part 17 of 20)
يوميات طالب بدرجة مشرف (Part 17 of 20)يوميات طالب بدرجة مشرف (Part 17 of 20)
يوميات طالب بدرجة مشرف (Part 17 of 20)Randa Elanwar
 
يوميات طالب بدرجة مشرف (Part 16 of 20)
يوميات طالب بدرجة مشرف (Part 16 of 20)يوميات طالب بدرجة مشرف (Part 16 of 20)
يوميات طالب بدرجة مشرف (Part 16 of 20)Randa Elanwar
 

More from Randa Elanwar (20)

الجزء السادس ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء السادس ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةالجزء السادس ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء السادس ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
 
الجزء الخامس ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الخامس ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةالجزء الخامس ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الخامس ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
 
الجزء الرابع ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الرابع ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةالجزء الرابع ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الرابع ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
 
الجزء الثالث ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الثالث ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةالجزء الثالث ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الثالث ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
 
الجزء الثاني ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الثاني ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةالجزء الثاني ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الثاني ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
 
الجزء الأول ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الأول ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوةالجزء الأول ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
الجزء الأول ماذا ستقدم لعميلك ريادة الأعمال خطوة بخطوة
 
تدريب مدونة علماء مصر على الكتابة الفنية (الترجمة والتلخيص )_Pdf5of5
تدريب مدونة علماء مصر على الكتابة الفنية (الترجمة والتلخيص    )_Pdf5of5تدريب مدونة علماء مصر على الكتابة الفنية (الترجمة والتلخيص    )_Pdf5of5
تدريب مدونة علماء مصر على الكتابة الفنية (الترجمة والتلخيص )_Pdf5of5
 
تدريب مدونة علماء مصر على الكتابة الفنية (القصة القصيرة والخاطرة والأخطاء ال...
تدريب مدونة علماء مصر على الكتابة الفنية (القصة القصيرة والخاطرة  والأخطاء ال...تدريب مدونة علماء مصر على الكتابة الفنية (القصة القصيرة والخاطرة  والأخطاء ال...
تدريب مدونة علماء مصر على الكتابة الفنية (القصة القصيرة والخاطرة والأخطاء ال...
 
تدريب مدونة علماء مصر على الكتابة الفنية (مقالات الموارد )_Pdf3of5
تدريب مدونة علماء مصر على الكتابة الفنية (مقالات الموارد   )_Pdf3of5تدريب مدونة علماء مصر على الكتابة الفنية (مقالات الموارد   )_Pdf3of5
تدريب مدونة علماء مصر على الكتابة الفنية (مقالات الموارد )_Pdf3of5
 
تدريب مدونة علماء مصر على الكتابة الفنية (المقالات الإخبارية )_Pdf2of5
تدريب مدونة علماء مصر على الكتابة الفنية (المقالات الإخبارية  )_Pdf2of5تدريب مدونة علماء مصر على الكتابة الفنية (المقالات الإخبارية  )_Pdf2of5
تدريب مدونة علماء مصر على الكتابة الفنية (المقالات الإخبارية )_Pdf2of5
 
تدريب مدونة علماء مصر على الكتابة الفنية (المقالات المبنية على البحث )_Pdf1of5
تدريب مدونة علماء مصر على الكتابة الفنية (المقالات المبنية على البحث )_Pdf1of5تدريب مدونة علماء مصر على الكتابة الفنية (المقالات المبنية على البحث )_Pdf1of5
تدريب مدونة علماء مصر على الكتابة الفنية (المقالات المبنية على البحث )_Pdf1of5
 
تعريف بمدونة علماء مصر ومحاور التدريب على الكتابة للمدونين
تعريف بمدونة علماء مصر ومحاور التدريب على الكتابة للمدونينتعريف بمدونة علماء مصر ومحاور التدريب على الكتابة للمدونين
تعريف بمدونة علماء مصر ومحاور التدريب على الكتابة للمدونين
 
Entrepreneurship_who_is_your_customer_(arabic)_7of7
Entrepreneurship_who_is_your_customer_(arabic)_7of7Entrepreneurship_who_is_your_customer_(arabic)_7of7
Entrepreneurship_who_is_your_customer_(arabic)_7of7
 
Entrepreneurship_who_is_your_customer_(arabic)_5of7
Entrepreneurship_who_is_your_customer_(arabic)_5of7Entrepreneurship_who_is_your_customer_(arabic)_5of7
Entrepreneurship_who_is_your_customer_(arabic)_5of7
 
Entrepreneurship_who_is_your_customer_(arabic)_4of7
Entrepreneurship_who_is_your_customer_(arabic)_4of7Entrepreneurship_who_is_your_customer_(arabic)_4of7
Entrepreneurship_who_is_your_customer_(arabic)_4of7
 
Entrepreneurship_who_is_your_customer_(arabic)_2of7
Entrepreneurship_who_is_your_customer_(arabic)_2of7Entrepreneurship_who_is_your_customer_(arabic)_2of7
Entrepreneurship_who_is_your_customer_(arabic)_2of7
 
يوميات طالب بدرجة مشرف (Part 19 of 20)
يوميات طالب بدرجة مشرف (Part 19 of 20)يوميات طالب بدرجة مشرف (Part 19 of 20)
يوميات طالب بدرجة مشرف (Part 19 of 20)
 
يوميات طالب بدرجة مشرف (Part 18 of 20)
يوميات طالب بدرجة مشرف (Part 18 of 20)يوميات طالب بدرجة مشرف (Part 18 of 20)
يوميات طالب بدرجة مشرف (Part 18 of 20)
 
يوميات طالب بدرجة مشرف (Part 17 of 20)
يوميات طالب بدرجة مشرف (Part 17 of 20)يوميات طالب بدرجة مشرف (Part 17 of 20)
يوميات طالب بدرجة مشرف (Part 17 of 20)
 
يوميات طالب بدرجة مشرف (Part 16 of 20)
يوميات طالب بدرجة مشرف (Part 16 of 20)يوميات طالب بدرجة مشرف (Part 16 of 20)
يوميات طالب بدرجة مشرف (Part 16 of 20)
 

Recently uploaded

Factory Acceptance Test( FAT).pptx .
Factory Acceptance Test( FAT).pptx       .Factory Acceptance Test( FAT).pptx       .
Factory Acceptance Test( FAT).pptx .Poonam Aher Patil
 
module for grade 9 for distance learning
module for grade 9 for distance learningmodule for grade 9 for distance learning
module for grade 9 for distance learninglevieagacer
 
THE ROLE OF BIOTECHNOLOGY IN THE ECONOMIC UPLIFT.pptx
THE ROLE OF BIOTECHNOLOGY IN THE ECONOMIC UPLIFT.pptxTHE ROLE OF BIOTECHNOLOGY IN THE ECONOMIC UPLIFT.pptx
THE ROLE OF BIOTECHNOLOGY IN THE ECONOMIC UPLIFT.pptxANSARKHAN96
 
Reboulia: features, anatomy, morphology etc.
Reboulia: features, anatomy, morphology etc.Reboulia: features, anatomy, morphology etc.
Reboulia: features, anatomy, morphology etc.Cherry
 
Porella : features, morphology, anatomy, reproduction etc.
Porella : features, morphology, anatomy, reproduction etc.Porella : features, morphology, anatomy, reproduction etc.
Porella : features, morphology, anatomy, reproduction etc.Cherry
 
Selaginella: features, morphology ,anatomy and reproduction.
Selaginella: features, morphology ,anatomy and reproduction.Selaginella: features, morphology ,anatomy and reproduction.
Selaginella: features, morphology ,anatomy and reproduction.Cherry
 
LUNULARIA -features, morphology, anatomy ,reproduction etc.
LUNULARIA -features, morphology, anatomy ,reproduction etc.LUNULARIA -features, morphology, anatomy ,reproduction etc.
LUNULARIA -features, morphology, anatomy ,reproduction etc.Cherry
 
Dr. E. Muralinath_ Blood indices_clinical aspects
Dr. E. Muralinath_ Blood indices_clinical  aspectsDr. E. Muralinath_ Blood indices_clinical  aspects
Dr. E. Muralinath_ Blood indices_clinical aspectsmuralinath2
 
Genome sequencing,shotgun sequencing.pptx
Genome sequencing,shotgun sequencing.pptxGenome sequencing,shotgun sequencing.pptx
Genome sequencing,shotgun sequencing.pptxCherry
 
Phenolics: types, biosynthesis and functions.
Phenolics: types, biosynthesis and functions.Phenolics: types, biosynthesis and functions.
Phenolics: types, biosynthesis and functions.Cherry
 
FS P2 COMBO MSTA LAST PUSH past exam papers.
FS P2 COMBO MSTA LAST PUSH past exam papers.FS P2 COMBO MSTA LAST PUSH past exam papers.
FS P2 COMBO MSTA LAST PUSH past exam papers.takadzanijustinmaime
 
Site specific recombination and transposition.........pdf
Site specific recombination and transposition.........pdfSite specific recombination and transposition.........pdf
Site specific recombination and transposition.........pdfCherry
 
Cot curve, melting temperature, unique and repetitive DNA
Cot curve, melting temperature, unique and repetitive DNACot curve, melting temperature, unique and repetitive DNA
Cot curve, melting temperature, unique and repetitive DNACherry
 
FAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and SpectrometryFAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and SpectrometryAlex Henderson
 
The Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptxThe Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptxseri bangash
 
Cyanide resistant respiration pathway.pptx
Cyanide resistant respiration pathway.pptxCyanide resistant respiration pathway.pptx
Cyanide resistant respiration pathway.pptxCherry
 
Plasmid: types, structure and functions.
Plasmid: types, structure and functions.Plasmid: types, structure and functions.
Plasmid: types, structure and functions.Cherry
 
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 bAsymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 bSérgio Sacani
 
Human genetics..........................pptx
Human genetics..........................pptxHuman genetics..........................pptx
Human genetics..........................pptxCherry
 
development of diagnostic enzyme assay to detect leuser virus
development of diagnostic enzyme assay to detect leuser virusdevelopment of diagnostic enzyme assay to detect leuser virus
development of diagnostic enzyme assay to detect leuser virusNazaninKarimi6
 

Recently uploaded (20)

Factory Acceptance Test( FAT).pptx .
Factory Acceptance Test( FAT).pptx       .Factory Acceptance Test( FAT).pptx       .
Factory Acceptance Test( FAT).pptx .
 
module for grade 9 for distance learning
module for grade 9 for distance learningmodule for grade 9 for distance learning
module for grade 9 for distance learning
 
THE ROLE OF BIOTECHNOLOGY IN THE ECONOMIC UPLIFT.pptx
THE ROLE OF BIOTECHNOLOGY IN THE ECONOMIC UPLIFT.pptxTHE ROLE OF BIOTECHNOLOGY IN THE ECONOMIC UPLIFT.pptx
THE ROLE OF BIOTECHNOLOGY IN THE ECONOMIC UPLIFT.pptx
 
Reboulia: features, anatomy, morphology etc.
Reboulia: features, anatomy, morphology etc.Reboulia: features, anatomy, morphology etc.
Reboulia: features, anatomy, morphology etc.
 
Porella : features, morphology, anatomy, reproduction etc.
Porella : features, morphology, anatomy, reproduction etc.Porella : features, morphology, anatomy, reproduction etc.
Porella : features, morphology, anatomy, reproduction etc.
 
Selaginella: features, morphology ,anatomy and reproduction.
Selaginella: features, morphology ,anatomy and reproduction.Selaginella: features, morphology ,anatomy and reproduction.
Selaginella: features, morphology ,anatomy and reproduction.
 
LUNULARIA -features, morphology, anatomy ,reproduction etc.
LUNULARIA -features, morphology, anatomy ,reproduction etc.LUNULARIA -features, morphology, anatomy ,reproduction etc.
LUNULARIA -features, morphology, anatomy ,reproduction etc.
 
Dr. E. Muralinath_ Blood indices_clinical aspects
Dr. E. Muralinath_ Blood indices_clinical  aspectsDr. E. Muralinath_ Blood indices_clinical  aspects
Dr. E. Muralinath_ Blood indices_clinical aspects
 
Genome sequencing,shotgun sequencing.pptx
Genome sequencing,shotgun sequencing.pptxGenome sequencing,shotgun sequencing.pptx
Genome sequencing,shotgun sequencing.pptx
 
Phenolics: types, biosynthesis and functions.
Phenolics: types, biosynthesis and functions.Phenolics: types, biosynthesis and functions.
Phenolics: types, biosynthesis and functions.
 
FS P2 COMBO MSTA LAST PUSH past exam papers.
FS P2 COMBO MSTA LAST PUSH past exam papers.FS P2 COMBO MSTA LAST PUSH past exam papers.
FS P2 COMBO MSTA LAST PUSH past exam papers.
 
Site specific recombination and transposition.........pdf
Site specific recombination and transposition.........pdfSite specific recombination and transposition.........pdf
Site specific recombination and transposition.........pdf
 
Cot curve, melting temperature, unique and repetitive DNA
Cot curve, melting temperature, unique and repetitive DNACot curve, melting temperature, unique and repetitive DNA
Cot curve, melting temperature, unique and repetitive DNA
 
FAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and SpectrometryFAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
 
The Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptxThe Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptx
 
Cyanide resistant respiration pathway.pptx
Cyanide resistant respiration pathway.pptxCyanide resistant respiration pathway.pptx
Cyanide resistant respiration pathway.pptx
 
Plasmid: types, structure and functions.
Plasmid: types, structure and functions.Plasmid: types, structure and functions.
Plasmid: types, structure and functions.
 
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 bAsymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
 
Human genetics..........................pptx
Human genetics..........................pptxHuman genetics..........................pptx
Human genetics..........................pptx
 
development of diagnostic enzyme assay to detect leuser virus
development of diagnostic enzyme assay to detect leuser virusdevelopment of diagnostic enzyme assay to detect leuser virus
development of diagnostic enzyme assay to detect leuser virus
 

Arabic Handwritten Script Recognition Towards Generalization: A Survey

  • 1. 1 Arabic Handwritten ScriptArabic Handwritten Script Recognition TowardsRecognition Towards Generalization: A SurveyGeneralization: A SurveyAuthors:Authors:  Randa I. M. ElanwarRanda I. M. Elanwar Assistant Researcher, Electronic Research Institute  Prof. Dr. Mohsen A. A. RashwanProf. Dr. Mohsen A. A. Rashwan Professor of Digital Signal Processing, Electronic and communication dept, Cairo University  Prof. Dr. Samia A. A. MashaliProf. Dr. Samia A. A. Mashali Head of computers and systems dept, Electronic Research Institute
  • 2. 2 Presentation ContentsPresentation Contents Introduction Paper Objective Arabic handwriting recognition problem Main Challenges Recent off-line Arabic handwriting recognition systems Recent on-line Arabic handwriting recognition systems Summary and Conclusion
  • 3. 3 IntroductionIntroduction Handwriting recognition can be defined as the task of transforming text represented in the spatial form of graphical marks into its symbolic representation The main components of a recognizer are: 1. Capturing Data & acquisition 2. Preprocessing & segmentation 3. Defining patterns and model selection 4. Feature Extraction 5. Training 6. Classification
  • 4. 4 IntroductionIntroduction • First the input device captures an image and convert it to a usable format • Data is then preprocessed to eliminate noise for simplification without loosing relevant information and may also be segmented to smaller data units
  • 5. 5 IntroductionIntroduction • The information of each data unit is sent to feature extractor to reduce them by measuring certain “features” or “properties” • Patterns (or classes) should be defined and models should be selected. These models are trained using the extracted features.
  • 6. 6 IntroductionIntroduction • The model for a pattern may be a single specific set of features • To recognize (or classify) a novel pattern means to recover the model that generated the pattern based on the extracted features
  • 7. 7 IntroductionIntroduction The feature extractor has reduced the data unit to a point or feature vector X in a 2D feature space (or observation space) Classification rule: Classify the input as Class I if its feature vector falls below the decision boundary shown, and as Class II otherwise.
  • 8. 8 IntroductionIntroduction The problem is that designing a very complex recognizer is unlikely to give good generalization since it seems to be “tuned” to the particular training samples The question is how to optimize this tradeoff: generalization versus simple classifier
  • 9. 9 IntroductionIntroduction Usually there is an action taken based on the classification decision. Each action should be assigned a certain cost. We design our decision boundary (classification rule) so that on the average, the Risk will be as small as possible. The Risk (R) is the expected value of cost Minimizing (R) leads to complex boundaries The question is how to optimize this tradeoff: generalization versus minimum risk?
  • 10. 10 IntroductionIntroduction In order to achieve general purpose recognizer (unbiased) we should have a sufficient number of training samples (N) for each class in the data set. A theoretical estimate claims that N ≅ 100 / P where P ≡ prob. of misclassification I.e., for P ≈ 0.01, N ≈ 10000 and for P ≈ 0.03, N ≈ 3000 Such large data set (if available) needs large storage and long processing time (time complexity) The question is how to optimize this tradeoff: generalization versus complexity?
  • 11. 11 Paper ObjectivePaper Objective Our concern in this paper is to: 1. provide a comprehensive review of recent off-line and on-line trends in Arabic cursive handwriting recognition (last 10 years publications) 2. clarify the challenges standing against obtaining a reliable, accurate, simple, general purpose recognizer based on these trends.
  • 12. 12 Arabic Handwriting Recognition ProblemArabic Handwriting Recognition Problem Arabic Script Recognition Systems are categorized as: 1. On-line or Off-line 2. Writer Dependent or Writer Independent 3. Open-vocabulary or closed-vocabulary
  • 13. 13 Arabic Handwriting Recognition ProblemArabic Handwriting Recognition Problem Types of Recognition: When the input device is a digitizer tablet that transmits the signal in real time or includes timing information together with pen position, this is mostly referred to as on-line or dynamic recognition
  • 14. 14 Arabic Handwriting Recognition ProblemArabic Handwriting Recognition Problem Types of Recognition: When the input device is a still camera or a scanner, which captures the position of digital ink on the page but not the order in which it was laid down, this is defined as off-line or image-based OCR
  • 15. 15 Arabic Handwriting Recognition ProblemArabic Handwriting Recognition Problem Special Characteristics of Arabic Script: Always written from right to left Arabic word consists of one or more portions; each has one or more characters Many characters differ only by the position and the number of dots attached
  • 16. 16 Arabic Handwriting Recognition ProblemArabic Handwriting Recognition Problem Special Characteristics of Arabic Script: Every character has more than one shape, depending on its position Characters overlap
  • 17. 17 Arabic Handwriting Recognition ProblemArabic Handwriting Recognition Problem Special Characteristics of Arabic Script: Existence of ligatures Due to having these special characteristics, Arabic handwriting recognition systems still need more research to be established commercially
  • 18. 18 Main ChallengesMain Challenges Feature Extraction Noise Model Selection and Complexity Segmentation Context Evidence Pooling Costs and Risks Computational Complexity Learning and Adaptation
  • 19. 19 Main ChallengesMain Challenges Feature Extraction: A good feature set should helps distinguishing a class from other classes, be invariant to differences and contains no redundant information
  • 20. 20 Main ChallengesMain Challenges Feature Extraction: A good feature set should helps distinguishing a class from other classes, be invariant to differences and contains no redundant information … How to know which features are most promising ? … Is there ways to automatically learn which features are best for a classifier?
  • 21. 21 Main ChallengesMain Challenges Feature Extraction: A good feature set should helps distinguishing a class from other classes, be invariant to differences and contains no redundant information … How to know which features are most promising ? … Is there ways to automatically learn which features are best for a classifier? It should be limited in number for computational ease and to limit the amount of training data
  • 22. 22 Main ChallengesMain Challenges Feature Extraction: A good feature set should helps distinguishing a class from other classes, be invariant to differences and contains no redundant information … How to know which features are most promising ? … Is there ways to automatically learn which features are best for a classifier? It should be limited in number for computational ease and to limit the amount of training data … How many features to use?
  • 23. 23 Main ChallengesMain Challenges Noise: Random error in a pixel value (deformation) due to signal-independent, signal-dependent and salt & pepper noise. Noise cannot always be totally eliminated; but smoothing is done
  • 24. 24 Main ChallengesMain Challenges Noise: Random error in a pixel value (deformation) due to signal-independent, signal-dependent and salt & pepper noise. Noise cannot always be totally eliminated; but smoothing is done … Is the deformation in some signal is noise? or natural varieties in true models? … How can we use this information to improve our classifier?
  • 25. 25 Main ChallengesMain Challenges Modeling Selection and Complexity: Determining the complexity of the model: not so simple that it cannot explain the differences between the categories, yet not so complex as to give poor classification on novel patterns.
  • 26. 26 Main ChallengesMain Challenges Modeling Selection and Complexity: Determining the complexity of the model: not so simple that it cannot explain the differences between the categories, yet not so complex as to give poor classification on novel patterns. … how to know when to reject a class of models and try another one? … Are there principled methods for finding the best complexity for a classifier? … Is it a matter of random trial & error not even guided by expectations of performance?
  • 27. 27 Main ChallengesMain Challenges Segmentation: Segmentation subdivides image into its constituent regions or objects. Segmentation should stop when the objects of interest in an application have been isolated.
  • 28. 28 Main ChallengesMain Challenges Segmentation: Segmentation subdivides image into its constituent regions or objects. Segmentation should stop when the objects of interest in an application have been isolated. … How do we know where one character “ends” and the next one “begin”? … Shall we segment the images before they have been categorized or categorize them before they have been segmented?
  • 29. 29 Main ChallengesMain Challenges Context: The accuracy of automatic handwriting recognition systems based on purely visual information seems to have a ceiling Incorporating Symantec and syntactic knowledge sources into the automatic recognition of text can offer potential improvements in performance … how, precisely, should we incorporate such information?
  • 30. 30 Main ChallengesMain Challenges Evidence Pooling: For high classification performance or for increased class coverage, different classification tools are developed either in parallel or sequentially When having several component classifiers, and these categorizers agree on a particular pattern, there is no difficulty. But suppose they disagree !!!
  • 31. 31 Main ChallengesMain Challenges Evidence Pooling: For high classification performance or for increased class coverage, different classification tools are developed either in parallel or sequentially When having several component classifiers, and these categorizers agree on a particular pattern, there is no difficulty. But suppose they disagree !!! … How should a “super” classifier pool the evidence from the component recognizers to achieve the best decision? … How would the “super” categorizer know when to base a decision on a minority opinion when required?
  • 32. 32 Main ChallengesMain Challenges Costs and Risks: A classifier is generally used to recommend actions, each action having an associated cost or risk We often design our classifier to recommend actions that minimize some total expected cost or risk
  • 33. 33 Main ChallengesMain Challenges Costs and Risks: A classifier is generally used to recommend actions, each action having an associated cost or risk We often design our classifier to recommend actions that minimize some total expected cost or risk … How do we incorporate knowledge about such risks and how will they affect the classification decision? … Is there a way to estimate the total risk and thus tell whether our classifier is acceptable even before we field it?
  • 34. 34 Main ChallengesMain Challenges Computational Complexity: Although we might achieve error-free recognition, the time & storage requirements would be quite prohibitive Some pattern recognition problems can be solved using algorithms that are highly impractical.
  • 35. 35 Main ChallengesMain Challenges Computational Complexity: Although we might achieve error-free recognition, the time & storage requirements would be quite prohibitive Some pattern recognition problems can be solved using algorithms that are highly impractical. … What is the tradeoff between computational ease and performance? … How can we optimize an excellent recognizer within the engineering constraints ?
  • 36. 36 Main ChallengesMain Challenges Learning and Adaptation: Any method that incorporates information from training samples in the design of a classifier employs learning If the models were extremely complicated, the classifier would have complex decision boundaries To overcome this, more training samples are needed to obtain a better estimate of the true underlying features In case of limited training samples, we should incorporate knowledge of the problem domain. The production representation is the “best” representation for classification.
  • 37. 37 Main ChallengesMain Challenges Learning and Adaptation: Any method that incorporates information from training samples in the design of a classifier employs learning If the models were extremely complicated, the classifier would have complex decision boundaries To overcome this, more training samples are needed to obtain a better estimate of the true underlying features In case of limited training samples, we should incorporate knowledge of the problem domain. The production representation is the “best” representation for classification. … How much training samples are needed for good generalization? … How can we insure that the learning algorithm favors “simple” solutions rather than complicated ones?
  • 38. 38 Recent off-line Arabic handwriting recognitionRecent off-line Arabic handwriting recognition systemssystems Example: Pechwitz et al research [17] proposed a recognition system based on a semi-continuous 1-D HMM using the IFN/ENIT database of handwritten Tunisian town/village names. Preprocessing: 1. Extracting image contour and Performing a noise reduction filtering. 2. Skeletonization and normalization are performed. 3. Baseline estimation and word length normalization are performed.
  • 39. 39 Recent off-line Arabic handwriting recognitionRecent off-line Arabic handwriting recognition systemssystems Example: Pechwitz et al research [17] Feature Extraction: 1. A rectangular window is shifted from right to left across the normalized gray level script image . 2. A Loeve-Karhunen Transformation is performed on the gray values of each frame to reduce the number of features. Modeling: 1. A HMM-model is generated for each character shape (all possible positions) up to 160 different HMM-models. 2. Semi Continuous HMMs are used with 7 states per character.
  • 40. 40 Recent off-line Arabic handwriting recognitionRecent off-line Arabic handwriting recognition systemssystems Example: Pechwitz et al research [17] Database: 1. This database is split into four sets A, B, C & D. 2. The 4 sets contain 26,459 images of segmented Tunisian town names (115,585 PAWs) handwritten by 411 unique writers. 3. 946 unique word labels, and 762 unique PAW labels. 4. For each image the ground truth information is available. Lexicon: The character shape HMM-models are combined to valid word models using a tree structured lexicon with all 946 different
  • 41. 41 Recent off-line Arabic handwriting recognitionRecent off-line Arabic handwriting recognition systemssystems Example: Pechwitz et al research [17] Recognition: The standard Viterbi Algorithm is used together with the lexicon. The authors applied the recognition algorithm to the database twice, once using the baseline coming from GT (ground truth) and once using baseline they estimated. Results: Recognition rates 82 – 89% are obtained using baseline estimation Recognition rates 89 – 95% are obtained using GT baseline
  • 42. 42 Recent off-line Arabic handwriting recognitionRecent off-line Arabic handwriting recognition systemssystems Example: Pechwitz et al research [17] Challenges: 1. Working on available database skips the limited training samples challenge
  • 43. 43 Recent off-line Arabic handwriting recognitionRecent off-line Arabic handwriting recognition systemssystems Example: Pechwitz et al research [17] Challenges: 1. Working on available database skips the limited training samples challenge 2. It is not easy to generalize this classifier for open vocabulary applications because it works on a limited lexicon of words (segmentation-free recognizer) otherwise context will be a must.
  • 44. 44 Recent off-line Arabic handwriting recognitionRecent off-line Arabic handwriting recognition systemssystems Example: Pechwitz et al research [17] Challenges: 1. Working on available database skips the limited training samples challenge 2. It is not easy to generalize this classifier for open vocabulary applications because it works on a limited lexicon of words (segmentation-free recognizer) otherwise context will be a must. 3. Generating the same HMM structure for all characters and ligatures i.e., modeling selection & complexity .. we think it would be much better to vary the model structure according to each character requirement (‫ض‬ shouldn’t have the same model as ‫ة‬ for example).
  • 45. 45 Recent off-line Arabic handwriting recognitionRecent off-line Arabic handwriting recognition systemssystems Example: Pechwitz et al research [17] Challenges: 1. Working on available database skips the limited training samples challenge 2. It is not easy to generalize this classifier for open vocabulary applications because it works on a limited lexicon of words (segmentation-free recognizer) otherwise context will be a must. 3. Generating the same HMM structure for all characters and ligatures i.e., modeling selection & complexity .. we think it would be much better to vary the model structure according to each character requirement (‫ض‬ shouldn’t have the same model as ‫ة‬ for example). 4. Feature Extraction: The idea of normalizing the word width to use a sliding window feature extractor is pretty good except for the great dependency on
  • 46. 46 Recent on-line Arabic handwriting recognitionRecent on-line Arabic handwriting recognition systemssystems Example: Biadsy et al research [24] Preprocessing: 1. Geometrical processing phase to minimize handwriting variations. 2. A low-pass filter is used to reduce noise and remove imperfections caused by acquisition devices. 3. The writing-speed is normalized by re-sampling the consequent point sequences. Feature Extraction: Mainly angles (with x-axis) and loop-presence
  • 47. 47 Recent on-line Arabic handwriting recognitionRecent on-line Arabic handwriting recognition systemssystems Example: Biadsy et al research [24] Modeling: 1. The recognition framework uses discrete Left-to-right HMMs to represent each Arabic letter shape (isolated, initial, medial, and final). 2. The number of states for each letter shape model is based on the geometric complexity of the letter shape. It varies from 5 to 11 states. For example: 11 states are assigned to isolated ‫,ش‬ and 5 states to isolated ‫.أ‬
  • 48. 48 Recent on-line Arabic handwriting recognitionRecent on-line Arabic handwriting recognition systemssystems Example: Biadsy et al research [24] Lexicon: 1. The Arabic dictionary D is subdivided into a set of sub-dictionaries {D1, D2, …, Dn} based on the number of word parts in each word. 2. Letter-shape models are embedded in a network that represents a word- part dictionary. The segmentation of word parts into letter-shapes and their recognition are performed simultaneously in an integrated process. D = {D = {‫انسان‬ ،‫التحدى‬ ،‫ثقافة‬ ،‫جامعة‬ ،‫رواية‬ ،‫فادى‬ ،‫محمد‬ ،‫محمود‬ ،‫معلم‬ ،‫هل‬ ،‫وسام‬‫انسان‬ ،‫التحدى‬ ،‫ثقافة‬ ،‫جامعة‬ ،‫رواية‬ ،‫فادى‬ ،‫محمد‬ ،‫محمود‬ ،‫معلم‬ ،‫هل‬ ،‫وسام‬}} Sub-dictionaries of DSub-dictionaries of D Word-Part Dictionary for D3Word-Part Dictionary for D3 D1 = {D1 = {‫محمد‬ ،‫معلم‬ ،‫هل‬‫محمد‬ ،‫معلم‬ ،‫هل‬}} D2 = {D2 = {‫ثقافة‬ ،‫جامعة‬ ،‫محمود‬‫ثقافة‬ ،‫جامعة‬ ،‫محمود‬}} D3 = {D3 = {‫انسان‬ ،‫التحدى‬ ،‫فادى‬ ،‫وسام‬‫انسان‬ ،‫التحدى‬ ،‫فادى‬ ،‫وسام‬}} D4 = {D4 = {‫رواية‬‫رواية‬}} WPD3,1 = {WPD3,1 = {‫ا‬ ،‫فا‬ ،‫و‬‫ا‬ ،‫فا‬ ،‫و‬}} WPD3,2 = {WPD3,2 = {‫نسا‬ ،‫لتحد‬ ،‫د‬ ،‫سا‬‫نسا‬ ،‫لتحد‬ ،‫د‬ ،‫سا‬}} WPD3,3 = {WPD3,3 = {‫ن‬ ،‫ى‬ ،‫م‬‫ن‬ ،‫ى‬ ،‫م‬}}
  • 49. 49 Recent on-line Arabic handwriting recognitionRecent on-line Arabic handwriting recognition systemssystems Example: Biadsy et al research [24] Database: 1. 4 trainers are asked to write 800 selected words each. 2. For testing, 10 testers (the 4 trainers, in addition to 6 new volunteers) are asked to write 280 words not in the training data (2,358 words in total). 3. 5 different dictionary sizes (5K, 10K, 20K, 30K, and 40K words) selected from different Arabic websites are used. The 280 test words are present in all dictionary sizes. Recognition: Writer dependent (WD) and writer independent (WI) experiments are done and average word recognition rates 88 – 96% are obtained. The
  • 50. 50 Recent on-line Arabic handwriting recognitionRecent on-line Arabic handwriting recognition systemssystems Example: Biadsy et al research [24] Challenges: 1. Feature Extraction: The features they use are not enough to lead to satisfying classification of general unconstrained handwritings. Thus they are in a great need to work under limited vocabulary. The word parts must be present in the dictionary or the will not be recognized.
  • 51. 51 Recent on-line Arabic handwriting recognitionRecent on-line Arabic handwriting recognition systemssystems Example: Biadsy et al research [24] Challenges: 1. Feature Extraction: The features they use are not enough to lead to satisfying classification of general unconstrained handwritings. Thus they are in a great need to work under limited vocabulary. The word parts must be present in the dictionary or the will not be recognized. 2. Database they use looks unnatural. Volunteers are asked to follow restrict methodology of writing which affects their individual writing style. Besides, the system handles limited handwriting varieties due to the small number of volunteers who wrote the database.
  • 52. 52 Summary and ConclusionSummary and Conclusion Foreign recognizers have found their way to the markets as commercial products since years while Arabic recognizers still need more time.
  • 53. 53 Summary and ConclusionSummary and Conclusion Foreign recognizers have found their way to the markets as commercial products since years while Arabic recognizers still need more time. in the case of Arabic handwritten words many researchers use a specific, more or less small data set of their own ∴ it is impossible to compare different results which would be important to improve existent methods
  • 54. 54 Summary and ConclusionSummary and Conclusion Foreign recognizers have found their way to the markets as commercial products since years while Arabic recognizers still need more time. in the case of Arabic handwritten words many researchers use a specific, more or less small data set of their own ∴ it is impossible to compare different results which would be important to improve existent methods The complexity of the problem is greatly increased by noise and by the infinite variability of handwritings
  • 55. 55 Summary and ConclusionSummary and Conclusion Cursive script requires the segmentation of words in characters or parts of characters, i.e. graphemes, and then the detection of individual features.
  • 56. 56 Summary and ConclusionSummary and Conclusion Cursive script requires the segmentation of words in characters or parts of characters, i.e. graphemes, and then the detection of individual features. Generally, the holistic approach can be used if the size of the vocabulary is small (such as the recognition of the legal amount in cheques)
  • 57. 57 Summary and ConclusionSummary and Conclusion Cursive script requires the segmentation of words in characters or parts of characters, i.e. graphemes, and then the detection of individual features. Generally, the holistic approach can be used if the size of the vocabulary is small (such as the recognition of the legal amount in cheques) The character-based approach is the preferred method for recognition applications that are unconstrained or involve large-size vocabularies to insure good generalization together with reasonable complexity