1. Plant Disease Detection
(A Case Study on Mango)
Dr. Pallav Rawal
Manisha Balani (19ESKEC080)
Kanika Singhal (19ESKEC063)
Kratik Khandelwal (19ESKEC071)
2. Image Acquisition
Image acquisition refers to the process of capturing or obtaining images using
various devices, such as cameras or scanners.
● Importance: Image acquisition is a crucial step in fields like
photography, medical imaging, computer vision, and more.
● Key Elements:
a. Image Sensors: These are electronic devices that convert light
into digital signals to capture images.
b. Optics: The lenses and components that focus light onto the
image sensor, affecting image quality.
c. Image Capture Devices: Examples include cameras, scanners, or
specialized imaging devices for specific applications.
3. Image Pre-Processing
Image pre-processing refers to a set of operations or techniques applied to an image before it is
analyzed or used in further applications. It involves transforming the raw image data into a format
that is more suitable for subsequent processing tasks such as image recognition, computer vision, or
machine learning algorithms.
The purpose of image pre-processing is to enhance the quality of the image, reduce noise or
artifacts, and extract relevant features or information for better analysis or interpretation. It typically
involves a series of steps, which may include:
● Image resizing: Adjusting the size of the image to a desired resolution or aspect ratio.
● Image cropping: Selecting a specific region of interest (ROI) within the image and discarding
● Color space conversion: Converting the image from one color space to another, such as RGB
to grayscale or RGB to HSV.
● Noise reduction: Removing or reducing unwanted noise or artifacts present in the image,
which can be caused by factors such as sensor limitations or image acquisition conditions.
Is input Image
Convert to Grayscale
Crop detected region
Resize cropped or detected face
Output cropped Image
5. Image Segmentation
● Image segmentation is a method of dividing a digital image into subgroups called
image segments, reducing the complexity of the image and enabling further
processing or analysis of each image segment.
● Technically, segmentation is the assignment of labels to pixels to identify objects,
people, or other important elements in the image.
● It is used for many practical applications including medical image analysis,
computer vision for autonomous vehicles, face recognition and detection, video
surveillance, and satellite image analysis.
6. Image Segmentation Techniques
1. Edge-Based Segmentation
Edge-based segmentation is a popular image processing technique that
identifies the edges of various objects in a given image. It helps locate
features of associated objects in the image using the information from the
edges. Edge detection helps strip images of redundant information,
reducing their size and facilitating analysis.
7. Thresholding is the simplest image segmentation method, dividing pixels based on
their intensity relative to a given value or threshold. It is suitable for segmenting
objects with higher intensity than other objects or backgrounds.
2. Threshold-Based Segmentation
9. 4. Cluster-Based Segmentation
Clustering algorithms are unsupervised classification algorithms that
help identify hidden information in images. They augment human
vision by isolating clusters, shadings, and structures. The algorithm
divides images into clusters of pixels with similar characteristics,
separating data elements and grouping similar elements into
10. Feature Extraction
The reduction of the data helps to build the model with less machine effort and also
increases the speed of learning and generalization steps in the machine learning process
What is feature extraction
●Dimensionality reduction process
●Helps to get the best feature from
those big data sets by selecting and
combining variables into features
●Effectively reducing the amount of
● Useful when you have a large data
●Helps to reduce the amount of
redundant data from the data set.
12. Measure of intensity
contrast between a pixel
and neighbor over entire
Measure spatial closeness
of distribution of co-
Measure of uniformity
where is maximum when
image is constant
of element of Co-
Gray Level Co-occurrence Matrix (GLCM) Method
13. ● Neural network, also known as an artificial neural network (ANN), is a computational model
inspired by the structure and function of biological neural networks.
● It consists of interconnected nodes called neurons, organized in layers: an input layer, one or
more hidden layers, and an output layer.
● Neurons receive inputs, apply a mathematical operation (activation function) to the inputs,
and produce an output.
● Each neuron is associated with weights that determine the strength of the connections
● Neural networks learn from data through a process called training, where the weights are
adjusted iteratively to minimize the difference between predicted and actual outputs.
● Backpropagation is a popular algorithm used to update the weights by propagating the error
backward through the network.
● Neural networks can be used for various tasks, including classification, regression, pattern
recognition, and time series prediction.
14. Types Of Neural Network
● ANN stands for Artificial Neural
● It is a computational model inspired by
the biological neural networks in the
● ANN consists of interconnected nodes
or "neurons" that process and transmit
● Neurons receive inputs, perform
computations, and produce outputs
using activation functions.
● CNN stands for Convolutional Neural
● It is a specialized type of artificial neural
network designed for image processing and
computer vision tasks.
● CNNs use convolutional layers to extract
features from input images by applying
● Pooling layers are used to reduce the
spatial dimensions of the features and
capture important information.
● KNN (k-nearest neighbors) is a simple algorithm used for classification and regression tasks.
● It determines the class of a new data point based on the majority vote of its k nearest neighbors in
the feature space.
● KNN is a non-parametric algorithm that is easy to understand and implement, but its performance
can be sensitive to the choice of k and it can be computationally expensive for large datasets.
15. Difference Between ANN And CNN
Type of Data Tabular Data, Text Data Image Data
Performance ANN is considered to be less powerful than
CNN is considered to be more powerful than
Application Facial recognition and Computer
Facial recognition, text digitization
and Natural language processing.
Parameter Sharing No Yes
Fixed Length input Yes Yes
Main advantages Having fault tolerance, Ability to
work with incomplete knowledge.
High accuracy in image recognition
problems, Weight sharing.
Disadvantages Hardware dependence, Unexplained
behavior of the network.
Large training data needed, don’t
encode the position and orientation of
16. K-Nearest Neighbors (K-NN)
● K-NN is a simple yet effective machine learning algorithm used
for classification and regression tasks.
● Key Concept:
● Instance-Based Learning: K-NN is an instance-based
learning algorithm where predictions are made based on
the similarity between new data points and training data.
● Nearest Neighbors: K-NN considers the k nearest
neighbors to the new data point to make predictions.
● Advantages of K-NN:
● Simplicity: K-NN is easy to understand and implement,
making it a popular choice for beginners.
● Versatility: It can be applied to both classification and
17. Flowchart for K-Nearest Neighbour(KNN)
Find the K nearest neighbors (D)
to the Test data
Set Maximum label class of
K to Test data
Read value of K,
Type distance(D) & test data
18. K-Means Clustering
● K-Means Clustering is a popular unsupervised machine learning algorithm used for grouping similar data
points into clusters.
● Key Concept:
● Data Clustering: K-Means aims to partition the data into k distinct clusters, where each data point
belongs to the cluster with the nearest mean or centroid.
● Advantages of K-Means Clustering:
● Simplicity: K-Means is easy to understand and implement, making it widely used for clustering tasks.
● Scalability: It can handle large datasets efficiently.
● Versatility: K-Means can be applied to a wide range of data types and is suitable for various clustering
19. K-Mean Clustering in Matlab
No. of Clusters K
Grouping based on
No object moved group?
20. Support Vector Machine(SVM)
Support Vector Machines (SVMs) are utilized in plant disease
detection, an important application in agriculture. SVMs offer an
effective approach for classifying and identifying plant diseases based
on their symptoms or images. By learning from labeled data, SVMs
can accurately distinguish healthy plants from those affected by
diseases, aiding in early detection and targeted interventions. Support
Vector Machines (SVMs) are utilized in plant disease detection, an
important application in agriculture. SVMs offer an effective
approach for classifying and identifying plant diseases based on their
symptoms or images. By learning from labeled data, SVMs can
accurately distinguish healthy plants from those affected by diseases,
aiding in early detection and targeted interventions. SVMs' ability to
handle non-linear data and their robustness make them valuable tools
in plant disease detection.
22. Decision Tree Classification Algorithm
Decision Tree is a Supervised learning technique
It is a tree-structured classifier, where internal nodes represent the features
of a dataset, branches represent the decision rules and each leaf node
represents the outcome.
It is a graphical representation for getting all the possible solutions to a
problem/decision based on given conditions.
A decision tree simply asks a question, and based on the answer (Yes/No),
it further split the tree into subtrees.
24. Naive Bayes Classifier
● supervised learning algorithm
● is based on Bayes theorem
● used for solving classification
● It is a probabilistic classifier, which
means it predicts on the basis of
the probability of an object.
Naïve Bayes Classifier Algorithm
used to determine the probability of a
hypothesis with prior knowledge. It
depends on the conditional probability
P(A|B)=(P(B|A)*P(A) )/ P(B)