About

Child pornography is criminal offence and a repercussion of child trafficking. Tens of thousands of videos are uploaded on many websites throughout the world in a day. This is a global social problem which needs to be addressed using latest technologies. We intend to prevent further addition of such videos on the internet and even remove such contents or report to the authority. We have developed an video analysis platform, which internally works on advanced image recognition using machine learning and neural-network concepts. It recognizes if a given pornographic video has child pornography. The recognized videos have certain accuracy associated. The video is then further reported to the website/manual testers. The primary objective of this web app is to reduce load of manual testers(belonging to pornographic websites for checking if a video contains child pornography) from manually viewing 100,000 videos to only 5000 videos. We further analyze who is uploading the video, from which area it is being uploaded and then mark certain areas as hotspots for such activities. This is helpful in tracking the uploaders. This location tracking further helps us to predict the probability of a video being a child pornographic video if it was being uploaded by a particular user from a particular location . The data of the location of uploader, tag of the video (child pornography or not as marked by manual tester), email of the uploader (gathered from the website from where the video is uploaded) is used to train classifiers to predict before hand if the video being uploaded is a child pornographic video or not depending on its location, email id of uploader data. Thus we aim to curb this social problem of child pornography which leads to human trafficking, sex offenses etc via our system. This will serve the common populace by helping to arrest all antisocial elements and criminals who are uploading such videos. Thus this will help to make the society as well as the internet a clean and a safe place

Comments