SlideShare a Scribd company logo
1 of 3
Download to read offline
International Journal of Science and Research (IJSR)
ISSN (Online): 2319-7064
Index Copernicus Value (2013): 6.14 | Impact Factor (2013): 4.438
Volume 4 Issue 1, January 2015
www.ijsr.net
Licensed Under Creative Commons Attribution CC BY
Immoral Scene Censoring System
Sam V Varghese1
, Tibin Thomas2
, Alpha Vijayan3
1
Mahatma Gandhi University, Mar Baselios Christian College of Engineering and Technology,
Kuttikkanam, Idukki Dist, 685 531, Kerala, India
2
Mahatma Gandhi University, Mar Baselios Christian College of Engineering and Technology,
Kuttikkanam, Idukki Dist, 685 531, Kerala, India
3
HOD, Mahatma Gandhi University, Mar Baselios Christian College of Engineering and Technology,
Kuttikkanam, Idukki Dist, 685 531, Kerala, India
Abstract: Harmful contents are rising in television day by day and this motivates the essence of more research in fast and reliable
obscene and immoral material filtering. Adultery content image recognition is an important component in each filtering system. We use
Skin detection process of finding skin colored pixels and regions in each frame of a video. This process is typically used as a
preprocessing step to find regions that potentially have human faces and limbs in images. Many computer vision approaches have been
developed for skin detection. Immoral imagery or as they put it pornography has been there for a long time and with the advent of web
it has only just grown. Adultery is considered as obscene material whose intention is to provoke sexual arousal. So, it should be
managed and certainly shouldn't be popping where it is not meant to. So we propose to design a system that takes a frame and detects if
it is a pornographic image or not. We intend to be using SVM (Support Vector Machines) to exploit primitive information from
different images and use the knowledge to determine whether given image is pornographic or not.
Keywords: Adultery detection, Adultery censoring media player, Skin detection, Computer vision
1. Introduction
creasing network bandwidth and storage capacity, video
streaming, and web services like YouTube allow us to
generate and share more digital video content than ever
before ( example, 20 hours of video are uploaded to
YouTube at each minute). Correspondingly, new methods
must be developed to cope with such huge loads of video
data. One challenge in this context is the automatic detection
of adult video content. Applications of this project include
the recursion of personal video digest (e.g., for child
protection) or digital forensics, where law enforcers need to
detect the illegal pornographic content. The image domain,
such technology has already been developed and is also
applied in practice. Regarding pornographic video material,
the straight forward approach would be to extract
representative key frames and apply image classification
techniques on them. In this paper, we demonstrate that better
results can be achieved by enriching this standard approach
with motion information. We compare key frame-based
methods (skin detection and bag-of-visual-words) with two
motion analysis approaches (periodicity detection and motion
histograms).Our evaluation is performed on real-world adult
web videos and intensive content from the web portal
YouTube. Results show that a significant improvement can
be achieved by combining both information sources (image
content and motion) in a late fusion step.
Most work regarding the detection of pornographic material
has been done for the image domain. Forsyth et al[1]
proposed to detect skin regions in an image and match them
with human bodies by applying geometric grouping rules.
Wang et al. presented a system that achieves a speedup by a
fast altering of icons and graphs. Successive steps include the
detection of skin areas and nearest-neighbor classification.
Popular data scientists " Jones and Rehg" focused on the
detection of human skin by constructing RGB color
histograms from a large database of skin and non-skin pixels,
which allows to estimate the skin probability of a pixel based
on its color. For adult image classification simple features of
the detected skin areas are fed to a neural network classier
(we will use similar approach in our evaluation). Rowley et
al. used Jones' skin color histograms in a system installed in
Google's Safe search.
A manually dined color range is used for deciding whether a
pixel belongs to a skin area. The area's shape is then
described by normalized central moments and matched to
samples in a database. The other modalities have been
employed for adult video content classification. Repeal
combined skin color estimation with the detection of periodic
patterns in a video's audio signal (The method could similarly
be applied to the motion signal). For periodicity detection,
the surface of the lines through local maxima and minima in
the signal's auto correlation function is computed.
2. Literature Review
This paper focuses on how to detect immoral scenes in video
by checking each video frame. Here we compare much
adultery scene detection method. But we mainly focus in skin
detection methods. D.A. Forsyth, [1] proposed an automatic
system for telling whether there are immoral scenes present
in an image. The system converts skin-like pixels using
combined color and texture properties. These are then fed to
a specialized grouper, then attempts to detect a human figure
using geometric constraints on human structure.
This method offers an alternate view of object recognition,
thus an object model is an organized collection of grouping
hints obtained from a combination of constraints on color and
texture and constraints on geometric properties such as the
Paper ID: SUB15786 2117
International Journal of Science and Research (IJSR)
ISSN (Online): 2319-7064
Index Copernicus Value (2013): 6.14 | Impact Factor (2013): 4.438
Volume 4 Issue 1, January 2015
www.ijsr.net
Licensed Under Creative Commons Attribution CC BY
structure of individual parts and the relationships between
parts. This system is not as accurate as some recent object
recognition algorithms, but it is performing a much more
abstract task (find a nude rather than find an object matching
this CAD model).
Seyed Mostafa, Hossein [2] in his method, describes the
Fourier descriptors and signature of boundary of skin region
is used as shape descriptor features. Fourier descriptors
describe the shape in terms of its spatial frequency content.
Fourier descriptors are the boundary based descriptors in 2D
space using the Fourier methods. Next, important features for
pornography which are more distinctive and less correlated
are found based on Self Organizing Feature Maps (SOFM)
and correlation analysis.
Jiann-Shu Leea, Yung-Ming Kuob [3] As can be seen in
above methods, none of them consider the inference coming
from special lighting and color altering. There exist a large
number of naked pictures taken under special lighting.
Usually, warm lighting is applied to make skin tone look
more attractive, while human skin color deviates from the
normal case at the same time. Dealing with the special
lighting effect in the naked images is a difficult task.A
feasible solution for the problem is to adapt the adopted skin
Chroma distribution to the lighting of the input image. Based
on this concept, a new naked image detection system is
proposed. In these paper, they develop a learning-based
chromatic distribution-matching scheme that consists of the
online sampler.
V.A.Oliveira,A.Cohini,[4] This work describes an
implementation for skin detection which relies on the H
channel to characterize the skin colors range. Here Open Cv
library is used for image Processing. Proposed system's
program initially converts RGB images to HSV one. H
channel is used to characterize the colors range for skin
detection’s color space.It is more related to human color.
Skin color in channel H is characterized by values between 0
and 50, in the channel S from 0.23 to 0.6 for Asian and
Caucasian ethnics. In this work they used images from
different people, from different places of the world. Their
detection range is better as compared with others.
Martin Stuetz, Martina Lindorfer, [5] In this paper, (svm)
they present an adaptable solution for detecting nudity or
pornography in color images. They combine a novel skin
detection approach with machine learning techniques to
alleviate manual image screening. Not all images have the
same quality. Some are over or underexposed skin may
contain compression artifacts or are simply of a small size
and therefore not rich in detail. In order to use a pixel-based
skin detection mechanism, the important is to equalize input
pictures as good as possible. Naturally, a lot of possibilities
to equalize images exist.
For the purpose of skin detection converting pixel onto RGB
color cordinate alone is not effiecent.Main skin clor space
use can hsv[4], YcbCr[6] and combination these.but better is
hsv color space.by our literature review we analysised that
for better skin detection huge saturation view (hsv) is
effiecent.
3. Proposed System
In our proposed system we develop a application Which use
the technology described by Martin Stuetz, Martina
Lindorfer, [5].Main technique in our system is skin detection
approach. Block diagram of prosed application is show in
below(fig).First we anlaysis video in each frame and check
whether any immoral scene in the frame.To retrive video
frames we use openCv and for skin detetion and nudility test
skin sheriff method [5] is used.if a particular cotinous range
of skin is identified(exposed) and which matches whith
human structure .we impose some restriction in displaying
the video.Also we add a parental facility in application.we
use this feature as a filtering option in an video player.
Figure:Proposed System
4. Future Work
Shape classification take large amount o f time or even
minutes per image, thus classifications are not currently
applicable to such a large number of images[7]. In a next
step, we will devise a method to estimate the age of
individuals on contraband images and categorize them in
adults and underage persons.we can also this techiniq ue as
tool for forensics to automatically screen files.
5. Conclusion
In this project, we have addressed the automatic detection of
pornographic content in videos, a problem with practical
applications in content altering and digital forensics. Our key
Paper ID: SUB15786 2118
International Journal of Science and Research (IJSR)
ISSN (Online): 2319-7064
Index Copernicus Value (2013): 6.14 | Impact Factor (2013): 4.438
Volume 4 Issue 1, January 2015
www.ijsr.net
Licensed Under Creative Commons Attribution CC BY
contribution lies in the fact that we evaluate both image
features and motion information as discriminative clues for
the detection of pornographic material. To the best of our
knowledge, the study presented in this paper is the rest one
that compares both feature modalities. Particularly, we show
that sign cant improvements can be achieved by a
combination of image features and motion in a simple late
fusion step.
References
[1] D.A Forsyth, Jinging Fu “Automatic Detection of Human
Body “ 2013
[2] Seyed Mostafa, Hossein Rahmani, Reza Mohsen
Ebrahimi, Amer Namazi “Skin Novel Scheme for
Intelligent Recognition of Pornography”, 2012.
[3] Jiann-Shu Leea,Yung-Ming Kuob,Pau-Choo Chungb, E-
Liang Chen” Naked image detection based on adaptive
skin color”, 2011.
[4] A.Oliveira, A. Cohini “Skin Detection using HSV color
space”, 2011
[5] Christian Platzer, Martin Stuetz, Martina Lindorfer “Skin
Sheriff: A Machine Learning Solution for Detecting
Explicit Images,” 2013.
[6] Jorge Alberto Marcial Basilio, Gualberto Aguilar Torres,
Gabriel Explicit Image Detection using YCbCr Space
Color Model as Skin Detection”, 2013.
[7] Christian Platzer, Martin Stuetz,Martina Lindorfer Large
Scale Image-Based Adult Content Filtering‖ , 2010
Author Profile
Sam V Varghese is currently pursuing B.Tech Degree in Computer
Science and Engineering in Mahatma Gandhi University Kottayam,
India,
Tibin Thomas is currently pursuing B.Tech Degree in Computer
Science and Engineering in Mahatma Gandhi University Kottayam,
India.
Alpha Vijayan is currently working as HOD in Computer Science
and Engineering department in MBCCET, Peermede, Idukki,
Kerala, India
Paper ID: SUB15786 2119

More Related Content

What's hot

17 icos2019(2)
17 icos2019(2)17 icos2019(2)
17 icos2019(2)arafah78
 
HUMAN MOTION DETECTION AND TRACKING FOR VIDEO SURVEILLANCE
HUMAN MOTION  DETECTION AND TRACKING FOR VIDEO SURVEILLANCEHUMAN MOTION  DETECTION AND TRACKING FOR VIDEO SURVEILLANCE
HUMAN MOTION DETECTION AND TRACKING FOR VIDEO SURVEILLANCENEHA THADEUS
 
MULTI SCALE ICA BASED IRIS RECOGNITION USING BSIF AND HOG
MULTI SCALE ICA BASED IRIS RECOGNITION USING BSIF AND HOG MULTI SCALE ICA BASED IRIS RECOGNITION USING BSIF AND HOG
MULTI SCALE ICA BASED IRIS RECOGNITION USING BSIF AND HOG sipij
 
IRJET- Comparative Analysis of Video Processing Object Detection
IRJET- Comparative Analysis of Video Processing Object DetectionIRJET- Comparative Analysis of Video Processing Object Detection
IRJET- Comparative Analysis of Video Processing Object DetectionIRJET Journal
 
META-HEURISTICS BASED ARF OPTIMIZATION FOR IMAGE RETRIEVAL
META-HEURISTICS BASED ARF OPTIMIZATION FOR IMAGE RETRIEVALMETA-HEURISTICS BASED ARF OPTIMIZATION FOR IMAGE RETRIEVAL
META-HEURISTICS BASED ARF OPTIMIZATION FOR IMAGE RETRIEVALIJCSEIT Journal
 
HSV Brightness Factor Matching for Gesture Recognition System
HSV Brightness Factor Matching for Gesture Recognition SystemHSV Brightness Factor Matching for Gesture Recognition System
HSV Brightness Factor Matching for Gesture Recognition SystemCSCJournals
 
System for Fingerprint Image Analysis
System for Fingerprint Image AnalysisSystem for Fingerprint Image Analysis
System for Fingerprint Image Analysisvivatechijri
 
50Combining Color Spaces for Human Skin Detection in Color Images using Skin ...
50Combining Color Spaces for Human Skin Detection in Color Images using Skin ...50Combining Color Spaces for Human Skin Detection in Color Images using Skin ...
50Combining Color Spaces for Human Skin Detection in Color Images using Skin ...idescitation
 
Robust Motion Detection and Tracking of Moving Objects using HOG Feature and ...
Robust Motion Detection and Tracking of Moving Objects using HOG Feature and ...Robust Motion Detection and Tracking of Moving Objects using HOG Feature and ...
Robust Motion Detection and Tracking of Moving Objects using HOG Feature and ...CSCJournals
 
Icete content-based filtering with applications on tv viewing data
Icete   content-based filtering with applications on tv viewing dataIcete   content-based filtering with applications on tv viewing data
Icete content-based filtering with applications on tv viewing dataElaine Cecília Gatto
 
IRJET- Application of MCNN in Object Detection
IRJET-  	  Application of MCNN in Object DetectionIRJET-  	  Application of MCNN in Object Detection
IRJET- Application of MCNN in Object DetectionIRJET Journal
 
Threshold based filtering technique for efficient moving object detection and...
Threshold based filtering technique for efficient moving object detection and...Threshold based filtering technique for efficient moving object detection and...
Threshold based filtering technique for efficient moving object detection and...eSAT Journals
 
Extraction of spots in dna microarrays using genetic algorithm
Extraction of spots in dna microarrays using genetic algorithmExtraction of spots in dna microarrays using genetic algorithm
Extraction of spots in dna microarrays using genetic algorithmsipij
 
EXPLOITING REFERENCE IMAGES IN EXPOSING GEOMETRICAL DISTORTIONS
EXPLOITING REFERENCE IMAGES IN EXPOSING GEOMETRICAL DISTORTIONSEXPLOITING REFERENCE IMAGES IN EXPOSING GEOMETRICAL DISTORTIONS
EXPLOITING REFERENCE IMAGES IN EXPOSING GEOMETRICAL DISTORTIONSijma
 
WAVELET PACKET BASED IRIS TEXTURE ANALYSIS FOR PERSON AUTHENTICATION
WAVELET PACKET BASED IRIS TEXTURE ANALYSIS FOR PERSON AUTHENTICATIONWAVELET PACKET BASED IRIS TEXTURE ANALYSIS FOR PERSON AUTHENTICATION
WAVELET PACKET BASED IRIS TEXTURE ANALYSIS FOR PERSON AUTHENTICATIONsipij
 
OBJECT DETECTION AND RECOGNITION: A SURVEY
OBJECT DETECTION AND RECOGNITION: A SURVEYOBJECT DETECTION AND RECOGNITION: A SURVEY
OBJECT DETECTION AND RECOGNITION: A SURVEYJournal For Research
 
A Survey on Approaches for Object Tracking
A Survey on Approaches for Object TrackingA Survey on Approaches for Object Tracking
A Survey on Approaches for Object Trackingjournal ijrtem
 

What's hot (19)

17 icos2019(2)
17 icos2019(2)17 icos2019(2)
17 icos2019(2)
 
HUMAN MOTION DETECTION AND TRACKING FOR VIDEO SURVEILLANCE
HUMAN MOTION  DETECTION AND TRACKING FOR VIDEO SURVEILLANCEHUMAN MOTION  DETECTION AND TRACKING FOR VIDEO SURVEILLANCE
HUMAN MOTION DETECTION AND TRACKING FOR VIDEO SURVEILLANCE
 
MULTI SCALE ICA BASED IRIS RECOGNITION USING BSIF AND HOG
MULTI SCALE ICA BASED IRIS RECOGNITION USING BSIF AND HOG MULTI SCALE ICA BASED IRIS RECOGNITION USING BSIF AND HOG
MULTI SCALE ICA BASED IRIS RECOGNITION USING BSIF AND HOG
 
IRJET- Comparative Analysis of Video Processing Object Detection
IRJET- Comparative Analysis of Video Processing Object DetectionIRJET- Comparative Analysis of Video Processing Object Detection
IRJET- Comparative Analysis of Video Processing Object Detection
 
META-HEURISTICS BASED ARF OPTIMIZATION FOR IMAGE RETRIEVAL
META-HEURISTICS BASED ARF OPTIMIZATION FOR IMAGE RETRIEVALMETA-HEURISTICS BASED ARF OPTIMIZATION FOR IMAGE RETRIEVAL
META-HEURISTICS BASED ARF OPTIMIZATION FOR IMAGE RETRIEVAL
 
HSV Brightness Factor Matching for Gesture Recognition System
HSV Brightness Factor Matching for Gesture Recognition SystemHSV Brightness Factor Matching for Gesture Recognition System
HSV Brightness Factor Matching for Gesture Recognition System
 
System for Fingerprint Image Analysis
System for Fingerprint Image AnalysisSystem for Fingerprint Image Analysis
System for Fingerprint Image Analysis
 
F010433136
F010433136F010433136
F010433136
 
50Combining Color Spaces for Human Skin Detection in Color Images using Skin ...
50Combining Color Spaces for Human Skin Detection in Color Images using Skin ...50Combining Color Spaces for Human Skin Detection in Color Images using Skin ...
50Combining Color Spaces for Human Skin Detection in Color Images using Skin ...
 
Robust Motion Detection and Tracking of Moving Objects using HOG Feature and ...
Robust Motion Detection and Tracking of Moving Objects using HOG Feature and ...Robust Motion Detection and Tracking of Moving Objects using HOG Feature and ...
Robust Motion Detection and Tracking of Moving Objects using HOG Feature and ...
 
Icete content-based filtering with applications on tv viewing data
Icete   content-based filtering with applications on tv viewing dataIcete   content-based filtering with applications on tv viewing data
Icete content-based filtering with applications on tv viewing data
 
IRJET- Application of MCNN in Object Detection
IRJET-  	  Application of MCNN in Object DetectionIRJET-  	  Application of MCNN in Object Detection
IRJET- Application of MCNN in Object Detection
 
Threshold based filtering technique for efficient moving object detection and...
Threshold based filtering technique for efficient moving object detection and...Threshold based filtering technique for efficient moving object detection and...
Threshold based filtering technique for efficient moving object detection and...
 
Extraction of spots in dna microarrays using genetic algorithm
Extraction of spots in dna microarrays using genetic algorithmExtraction of spots in dna microarrays using genetic algorithm
Extraction of spots in dna microarrays using genetic algorithm
 
EXPLOITING REFERENCE IMAGES IN EXPOSING GEOMETRICAL DISTORTIONS
EXPLOITING REFERENCE IMAGES IN EXPOSING GEOMETRICAL DISTORTIONSEXPLOITING REFERENCE IMAGES IN EXPOSING GEOMETRICAL DISTORTIONS
EXPLOITING REFERENCE IMAGES IN EXPOSING GEOMETRICAL DISTORTIONS
 
N010226872
N010226872N010226872
N010226872
 
WAVELET PACKET BASED IRIS TEXTURE ANALYSIS FOR PERSON AUTHENTICATION
WAVELET PACKET BASED IRIS TEXTURE ANALYSIS FOR PERSON AUTHENTICATIONWAVELET PACKET BASED IRIS TEXTURE ANALYSIS FOR PERSON AUTHENTICATION
WAVELET PACKET BASED IRIS TEXTURE ANALYSIS FOR PERSON AUTHENTICATION
 
OBJECT DETECTION AND RECOGNITION: A SURVEY
OBJECT DETECTION AND RECOGNITION: A SURVEYOBJECT DETECTION AND RECOGNITION: A SURVEY
OBJECT DETECTION AND RECOGNITION: A SURVEY
 
A Survey on Approaches for Object Tracking
A Survey on Approaches for Object TrackingA Survey on Approaches for Object Tracking
A Survey on Approaches for Object Tracking
 

Viewers also liked

2012 11(123) motoreview
2012 11(123) motoreview2012 11(123) motoreview
2012 11(123) motoreviewgsxr77
 
2004 04(20)april motoreview
2004 04(20)april motoreview2004 04(20)april motoreview
2004 04(20)april motoreviewgsxr77
 
2008 02(66)february motoreview
2008 02(66)february motoreview2008 02(66)february motoreview
2008 02(66)february motoreviewgsxr77
 
Fordrazborka.zu8.ru focus 2_1.6
Fordrazborka.zu8.ru focus 2_1.6Fordrazborka.zu8.ru focus 2_1.6
Fordrazborka.zu8.ru focus 2_1.6gsxr77
 
2008 05(69)may motoreview
2008 05(69)may motoreview2008 05(69)may motoreview
2008 05(69)may motoreviewgsxr77
 
2007 01(53)january motoreview
2007 01(53)january motoreview2007 01(53)january motoreview
2007 01(53)january motoreviewgsxr77
 
immoral scene sensoring
immoral scene sensoringimmoral scene sensoring
immoral scene sensoringSam varghese
 
2011 05(105) may_motoreview
2011 05(105) may_motoreview2011 05(105) may_motoreview
2011 05(105) may_motoreviewgsxr77
 
Fordrazborka.zu8.ru focus 2
Fordrazborka.zu8.ru focus 2 Fordrazborka.zu8.ru focus 2
Fordrazborka.zu8.ru focus 2 gsxr77
 
2012 03(115) motoreview
2012 03(115) motoreview2012 03(115) motoreview
2012 03(115) motoreviewgsxr77
 
2008 04(68)april motoreview
2008 04(68)april motoreview2008 04(68)april motoreview
2008 04(68)april motoreviewgsxr77
 
Jaringankomputer lsp telematika
Jaringankomputer lsp telematikaJaringankomputer lsp telematika
Jaringankomputer lsp telematikaArif Nurnanto
 
2010 07(95) july_motoreview
2010 07(95) july_motoreview2010 07(95) july_motoreview
2010 07(95) july_motoreviewgsxr77
 
Integration of educational philosophy to dep ed mission vision
Integration of educational philosophy to dep ed mission visionIntegration of educational philosophy to dep ed mission vision
Integration of educational philosophy to dep ed mission visionteska08
 

Viewers also liked (16)

2012 11(123) motoreview
2012 11(123) motoreview2012 11(123) motoreview
2012 11(123) motoreview
 
2004 04(20)april motoreview
2004 04(20)april motoreview2004 04(20)april motoreview
2004 04(20)april motoreview
 
2008 02(66)february motoreview
2008 02(66)february motoreview2008 02(66)february motoreview
2008 02(66)february motoreview
 
Fordrazborka.zu8.ru focus 2_1.6
Fordrazborka.zu8.ru focus 2_1.6Fordrazborka.zu8.ru focus 2_1.6
Fordrazborka.zu8.ru focus 2_1.6
 
2008 05(69)may motoreview
2008 05(69)may motoreview2008 05(69)may motoreview
2008 05(69)may motoreview
 
2007 01(53)january motoreview
2007 01(53)january motoreview2007 01(53)january motoreview
2007 01(53)january motoreview
 
áLbum lector de 4eso 2013 2104
áLbum lector de 4eso 2013 2104áLbum lector de 4eso 2013 2104
áLbum lector de 4eso 2013 2104
 
immoral scene sensoring
immoral scene sensoringimmoral scene sensoring
immoral scene sensoring
 
2011 05(105) may_motoreview
2011 05(105) may_motoreview2011 05(105) may_motoreview
2011 05(105) may_motoreview
 
Fordrazborka.zu8.ru focus 2
Fordrazborka.zu8.ru focus 2 Fordrazborka.zu8.ru focus 2
Fordrazborka.zu8.ru focus 2
 
2012 03(115) motoreview
2012 03(115) motoreview2012 03(115) motoreview
2012 03(115) motoreview
 
2008 04(68)april motoreview
2008 04(68)april motoreview2008 04(68)april motoreview
2008 04(68)april motoreview
 
Jaringankomputer lsp telematika
Jaringankomputer lsp telematikaJaringankomputer lsp telematika
Jaringankomputer lsp telematika
 
2010 07(95) july_motoreview
2010 07(95) july_motoreview2010 07(95) july_motoreview
2010 07(95) july_motoreview
 
Integration of educational philosophy to dep ed mission vision
Integration of educational philosophy to dep ed mission visionIntegration of educational philosophy to dep ed mission vision
Integration of educational philosophy to dep ed mission vision
 
Ibm truenorth
Ibm truenorthIbm truenorth
Ibm truenorth
 

Similar to SUB15786

Paper id 25201494
Paper id 25201494Paper id 25201494
Paper id 25201494IJRAT
 
A Review on Matching For Sketch Technique
A Review on Matching For Sketch TechniqueA Review on Matching For Sketch Technique
A Review on Matching For Sketch TechniqueIOSR Journals
 
Paper id 24201475
Paper id 24201475Paper id 24201475
Paper id 24201475IJRAT
 
A novel enhanced algorithm for efficient human tracking
A novel enhanced algorithm for efficient human trackingA novel enhanced algorithm for efficient human tracking
A novel enhanced algorithm for efficient human trackingIJICTJOURNAL
 
Top Cited Articles in Signal & Image Processing 2021-2022
Top Cited Articles in Signal & Image Processing 2021-2022Top Cited Articles in Signal & Image Processing 2021-2022
Top Cited Articles in Signal & Image Processing 2021-2022sipij
 
Skin Color Detection Using Region-Based Approach
Skin Color Detection Using Region-Based ApproachSkin Color Detection Using Region-Based Approach
Skin Color Detection Using Region-Based ApproachCSCJournals
 
Announcing the Final Examination of Mr. Paul Smith for the ...
Announcing the Final Examination of Mr. Paul Smith for the ...Announcing the Final Examination of Mr. Paul Smith for the ...
Announcing the Final Examination of Mr. Paul Smith for the ...butest
 
IRJET - Explicit Content Detection using Faster R-CNN and SSD Mobilenet V2
 IRJET - Explicit Content Detection using Faster R-CNN and SSD Mobilenet V2 IRJET - Explicit Content Detection using Faster R-CNN and SSD Mobilenet V2
IRJET - Explicit Content Detection using Faster R-CNN and SSD Mobilenet V2IRJET Journal
 
A Comparison of People Counting Techniques via Video Scene Analysis
A Comparison of People Counting Techniques viaVideo Scene AnalysisA Comparison of People Counting Techniques viaVideo Scene Analysis
A Comparison of People Counting Techniques via Video Scene AnalysisPoo Kuan Hoong
 
Proposed Multi-object Tracking Algorithm Using Sobel Edge Detection operator
Proposed Multi-object Tracking Algorithm Using Sobel Edge Detection operatorProposed Multi-object Tracking Algorithm Using Sobel Edge Detection operator
Proposed Multi-object Tracking Algorithm Using Sobel Edge Detection operatorQUESTJOURNAL
 
Elderly Assistance- Deep Learning Theme detection
Elderly Assistance- Deep Learning Theme detectionElderly Assistance- Deep Learning Theme detection
Elderly Assistance- Deep Learning Theme detectionTanvi Mittal
 
Automatic motion tracking system for analysis of insect behavior
Automatic motion tracking system for analysis of insect behaviorAutomatic motion tracking system for analysis of insect behavior
Automatic motion tracking system for analysis of insect behaviorWookjin Choi
 
A Framework for Human Action Detection via Extraction of Multimodal Features
A Framework for Human Action Detection via Extraction of Multimodal FeaturesA Framework for Human Action Detection via Extraction of Multimodal Features
A Framework for Human Action Detection via Extraction of Multimodal FeaturesCSCJournals
 
Explicit Content Image Detection
Explicit Content Image DetectionExplicit Content Image Detection
Explicit Content Image Detectionsipij
 
A Novel Efficient Medical Image Segmentation Methodology
A Novel Efficient Medical Image Segmentation MethodologyA Novel Efficient Medical Image Segmentation Methodology
A Novel Efficient Medical Image Segmentation Methodologyaciijournal
 
Automatic video censoring system using deep learning
Automatic video censoring system using deep learningAutomatic video censoring system using deep learning
Automatic video censoring system using deep learningIJECEIAES
 
IRJET- Survey on Detection of Crime
IRJET-  	  Survey on Detection of CrimeIRJET-  	  Survey on Detection of Crime
IRJET- Survey on Detection of CrimeIRJET Journal
 

Similar to SUB15786 (20)

Ai4
Ai4Ai4
Ai4
 
PAPER3
PAPER3PAPER3
PAPER3
 
Paper id 25201494
Paper id 25201494Paper id 25201494
Paper id 25201494
 
A Review on Matching For Sketch Technique
A Review on Matching For Sketch TechniqueA Review on Matching For Sketch Technique
A Review on Matching For Sketch Technique
 
Paper id 24201475
Paper id 24201475Paper id 24201475
Paper id 24201475
 
A novel enhanced algorithm for efficient human tracking
A novel enhanced algorithm for efficient human trackingA novel enhanced algorithm for efficient human tracking
A novel enhanced algorithm for efficient human tracking
 
Top Cited Articles in Signal & Image Processing 2021-2022
Top Cited Articles in Signal & Image Processing 2021-2022Top Cited Articles in Signal & Image Processing 2021-2022
Top Cited Articles in Signal & Image Processing 2021-2022
 
Skin Color Detection Using Region-Based Approach
Skin Color Detection Using Region-Based ApproachSkin Color Detection Using Region-Based Approach
Skin Color Detection Using Region-Based Approach
 
Announcing the Final Examination of Mr. Paul Smith for the ...
Announcing the Final Examination of Mr. Paul Smith for the ...Announcing the Final Examination of Mr. Paul Smith for the ...
Announcing the Final Examination of Mr. Paul Smith for the ...
 
IRJET - Explicit Content Detection using Faster R-CNN and SSD Mobilenet V2
 IRJET - Explicit Content Detection using Faster R-CNN and SSD Mobilenet V2 IRJET - Explicit Content Detection using Faster R-CNN and SSD Mobilenet V2
IRJET - Explicit Content Detection using Faster R-CNN and SSD Mobilenet V2
 
A Comparison of People Counting Techniques via Video Scene Analysis
A Comparison of People Counting Techniques viaVideo Scene AnalysisA Comparison of People Counting Techniques viaVideo Scene Analysis
A Comparison of People Counting Techniques via Video Scene Analysis
 
I045034956
I045034956I045034956
I045034956
 
Proposed Multi-object Tracking Algorithm Using Sobel Edge Detection operator
Proposed Multi-object Tracking Algorithm Using Sobel Edge Detection operatorProposed Multi-object Tracking Algorithm Using Sobel Edge Detection operator
Proposed Multi-object Tracking Algorithm Using Sobel Edge Detection operator
 
Elderly Assistance- Deep Learning Theme detection
Elderly Assistance- Deep Learning Theme detectionElderly Assistance- Deep Learning Theme detection
Elderly Assistance- Deep Learning Theme detection
 
Automatic motion tracking system for analysis of insect behavior
Automatic motion tracking system for analysis of insect behaviorAutomatic motion tracking system for analysis of insect behavior
Automatic motion tracking system for analysis of insect behavior
 
A Framework for Human Action Detection via Extraction of Multimodal Features
A Framework for Human Action Detection via Extraction of Multimodal FeaturesA Framework for Human Action Detection via Extraction of Multimodal Features
A Framework for Human Action Detection via Extraction of Multimodal Features
 
Explicit Content Image Detection
Explicit Content Image DetectionExplicit Content Image Detection
Explicit Content Image Detection
 
A Novel Efficient Medical Image Segmentation Methodology
A Novel Efficient Medical Image Segmentation MethodologyA Novel Efficient Medical Image Segmentation Methodology
A Novel Efficient Medical Image Segmentation Methodology
 
Automatic video censoring system using deep learning
Automatic video censoring system using deep learningAutomatic video censoring system using deep learning
Automatic video censoring system using deep learning
 
IRJET- Survey on Detection of Crime
IRJET-  	  Survey on Detection of CrimeIRJET-  	  Survey on Detection of Crime
IRJET- Survey on Detection of Crime
 

SUB15786

  • 1. International Journal of Science and Research (IJSR) ISSN (Online): 2319-7064 Index Copernicus Value (2013): 6.14 | Impact Factor (2013): 4.438 Volume 4 Issue 1, January 2015 www.ijsr.net Licensed Under Creative Commons Attribution CC BY Immoral Scene Censoring System Sam V Varghese1 , Tibin Thomas2 , Alpha Vijayan3 1 Mahatma Gandhi University, Mar Baselios Christian College of Engineering and Technology, Kuttikkanam, Idukki Dist, 685 531, Kerala, India 2 Mahatma Gandhi University, Mar Baselios Christian College of Engineering and Technology, Kuttikkanam, Idukki Dist, 685 531, Kerala, India 3 HOD, Mahatma Gandhi University, Mar Baselios Christian College of Engineering and Technology, Kuttikkanam, Idukki Dist, 685 531, Kerala, India Abstract: Harmful contents are rising in television day by day and this motivates the essence of more research in fast and reliable obscene and immoral material filtering. Adultery content image recognition is an important component in each filtering system. We use Skin detection process of finding skin colored pixels and regions in each frame of a video. This process is typically used as a preprocessing step to find regions that potentially have human faces and limbs in images. Many computer vision approaches have been developed for skin detection. Immoral imagery or as they put it pornography has been there for a long time and with the advent of web it has only just grown. Adultery is considered as obscene material whose intention is to provoke sexual arousal. So, it should be managed and certainly shouldn't be popping where it is not meant to. So we propose to design a system that takes a frame and detects if it is a pornographic image or not. We intend to be using SVM (Support Vector Machines) to exploit primitive information from different images and use the knowledge to determine whether given image is pornographic or not. Keywords: Adultery detection, Adultery censoring media player, Skin detection, Computer vision 1. Introduction creasing network bandwidth and storage capacity, video streaming, and web services like YouTube allow us to generate and share more digital video content than ever before ( example, 20 hours of video are uploaded to YouTube at each minute). Correspondingly, new methods must be developed to cope with such huge loads of video data. One challenge in this context is the automatic detection of adult video content. Applications of this project include the recursion of personal video digest (e.g., for child protection) or digital forensics, where law enforcers need to detect the illegal pornographic content. The image domain, such technology has already been developed and is also applied in practice. Regarding pornographic video material, the straight forward approach would be to extract representative key frames and apply image classification techniques on them. In this paper, we demonstrate that better results can be achieved by enriching this standard approach with motion information. We compare key frame-based methods (skin detection and bag-of-visual-words) with two motion analysis approaches (periodicity detection and motion histograms).Our evaluation is performed on real-world adult web videos and intensive content from the web portal YouTube. Results show that a significant improvement can be achieved by combining both information sources (image content and motion) in a late fusion step. Most work regarding the detection of pornographic material has been done for the image domain. Forsyth et al[1] proposed to detect skin regions in an image and match them with human bodies by applying geometric grouping rules. Wang et al. presented a system that achieves a speedup by a fast altering of icons and graphs. Successive steps include the detection of skin areas and nearest-neighbor classification. Popular data scientists " Jones and Rehg" focused on the detection of human skin by constructing RGB color histograms from a large database of skin and non-skin pixels, which allows to estimate the skin probability of a pixel based on its color. For adult image classification simple features of the detected skin areas are fed to a neural network classier (we will use similar approach in our evaluation). Rowley et al. used Jones' skin color histograms in a system installed in Google's Safe search. A manually dined color range is used for deciding whether a pixel belongs to a skin area. The area's shape is then described by normalized central moments and matched to samples in a database. The other modalities have been employed for adult video content classification. Repeal combined skin color estimation with the detection of periodic patterns in a video's audio signal (The method could similarly be applied to the motion signal). For periodicity detection, the surface of the lines through local maxima and minima in the signal's auto correlation function is computed. 2. Literature Review This paper focuses on how to detect immoral scenes in video by checking each video frame. Here we compare much adultery scene detection method. But we mainly focus in skin detection methods. D.A. Forsyth, [1] proposed an automatic system for telling whether there are immoral scenes present in an image. The system converts skin-like pixels using combined color and texture properties. These are then fed to a specialized grouper, then attempts to detect a human figure using geometric constraints on human structure. This method offers an alternate view of object recognition, thus an object model is an organized collection of grouping hints obtained from a combination of constraints on color and texture and constraints on geometric properties such as the Paper ID: SUB15786 2117
  • 2. International Journal of Science and Research (IJSR) ISSN (Online): 2319-7064 Index Copernicus Value (2013): 6.14 | Impact Factor (2013): 4.438 Volume 4 Issue 1, January 2015 www.ijsr.net Licensed Under Creative Commons Attribution CC BY structure of individual parts and the relationships between parts. This system is not as accurate as some recent object recognition algorithms, but it is performing a much more abstract task (find a nude rather than find an object matching this CAD model). Seyed Mostafa, Hossein [2] in his method, describes the Fourier descriptors and signature of boundary of skin region is used as shape descriptor features. Fourier descriptors describe the shape in terms of its spatial frequency content. Fourier descriptors are the boundary based descriptors in 2D space using the Fourier methods. Next, important features for pornography which are more distinctive and less correlated are found based on Self Organizing Feature Maps (SOFM) and correlation analysis. Jiann-Shu Leea, Yung-Ming Kuob [3] As can be seen in above methods, none of them consider the inference coming from special lighting and color altering. There exist a large number of naked pictures taken under special lighting. Usually, warm lighting is applied to make skin tone look more attractive, while human skin color deviates from the normal case at the same time. Dealing with the special lighting effect in the naked images is a difficult task.A feasible solution for the problem is to adapt the adopted skin Chroma distribution to the lighting of the input image. Based on this concept, a new naked image detection system is proposed. In these paper, they develop a learning-based chromatic distribution-matching scheme that consists of the online sampler. V.A.Oliveira,A.Cohini,[4] This work describes an implementation for skin detection which relies on the H channel to characterize the skin colors range. Here Open Cv library is used for image Processing. Proposed system's program initially converts RGB images to HSV one. H channel is used to characterize the colors range for skin detection’s color space.It is more related to human color. Skin color in channel H is characterized by values between 0 and 50, in the channel S from 0.23 to 0.6 for Asian and Caucasian ethnics. In this work they used images from different people, from different places of the world. Their detection range is better as compared with others. Martin Stuetz, Martina Lindorfer, [5] In this paper, (svm) they present an adaptable solution for detecting nudity or pornography in color images. They combine a novel skin detection approach with machine learning techniques to alleviate manual image screening. Not all images have the same quality. Some are over or underexposed skin may contain compression artifacts or are simply of a small size and therefore not rich in detail. In order to use a pixel-based skin detection mechanism, the important is to equalize input pictures as good as possible. Naturally, a lot of possibilities to equalize images exist. For the purpose of skin detection converting pixel onto RGB color cordinate alone is not effiecent.Main skin clor space use can hsv[4], YcbCr[6] and combination these.but better is hsv color space.by our literature review we analysised that for better skin detection huge saturation view (hsv) is effiecent. 3. Proposed System In our proposed system we develop a application Which use the technology described by Martin Stuetz, Martina Lindorfer, [5].Main technique in our system is skin detection approach. Block diagram of prosed application is show in below(fig).First we anlaysis video in each frame and check whether any immoral scene in the frame.To retrive video frames we use openCv and for skin detetion and nudility test skin sheriff method [5] is used.if a particular cotinous range of skin is identified(exposed) and which matches whith human structure .we impose some restriction in displaying the video.Also we add a parental facility in application.we use this feature as a filtering option in an video player. Figure:Proposed System 4. Future Work Shape classification take large amount o f time or even minutes per image, thus classifications are not currently applicable to such a large number of images[7]. In a next step, we will devise a method to estimate the age of individuals on contraband images and categorize them in adults and underage persons.we can also this techiniq ue as tool for forensics to automatically screen files. 5. Conclusion In this project, we have addressed the automatic detection of pornographic content in videos, a problem with practical applications in content altering and digital forensics. Our key Paper ID: SUB15786 2118
  • 3. International Journal of Science and Research (IJSR) ISSN (Online): 2319-7064 Index Copernicus Value (2013): 6.14 | Impact Factor (2013): 4.438 Volume 4 Issue 1, January 2015 www.ijsr.net Licensed Under Creative Commons Attribution CC BY contribution lies in the fact that we evaluate both image features and motion information as discriminative clues for the detection of pornographic material. To the best of our knowledge, the study presented in this paper is the rest one that compares both feature modalities. Particularly, we show that sign cant improvements can be achieved by a combination of image features and motion in a simple late fusion step. References [1] D.A Forsyth, Jinging Fu “Automatic Detection of Human Body “ 2013 [2] Seyed Mostafa, Hossein Rahmani, Reza Mohsen Ebrahimi, Amer Namazi “Skin Novel Scheme for Intelligent Recognition of Pornography”, 2012. [3] Jiann-Shu Leea,Yung-Ming Kuob,Pau-Choo Chungb, E- Liang Chen” Naked image detection based on adaptive skin color”, 2011. [4] A.Oliveira, A. Cohini “Skin Detection using HSV color space”, 2011 [5] Christian Platzer, Martin Stuetz, Martina Lindorfer “Skin Sheriff: A Machine Learning Solution for Detecting Explicit Images,” 2013. [6] Jorge Alberto Marcial Basilio, Gualberto Aguilar Torres, Gabriel Explicit Image Detection using YCbCr Space Color Model as Skin Detection”, 2013. [7] Christian Platzer, Martin Stuetz,Martina Lindorfer Large Scale Image-Based Adult Content Filtering‖ , 2010 Author Profile Sam V Varghese is currently pursuing B.Tech Degree in Computer Science and Engineering in Mahatma Gandhi University Kottayam, India, Tibin Thomas is currently pursuing B.Tech Degree in Computer Science and Engineering in Mahatma Gandhi University Kottayam, India. Alpha Vijayan is currently working as HOD in Computer Science and Engineering department in MBCCET, Peermede, Idukki, Kerala, India Paper ID: SUB15786 2119