Isolation Procedure Inference of User-Uploaded Images on Content Sharing Sites

Abstract

Usage of social media’s increased considerably in today world which enables the user to share their personal information like images with the other. This improved technology leads to privacy violation where the users are sharing the large volumes of images across more number of peoples. To provide security for the information, automated annotation of images are introduced which aims to create the meta data information about the images by using the novel approach called Semantic annotated Markovian Semantic Indexing(SMSI) for retrieving the images. To achieve this privacy settings for the people images we are using Adaptive Privacy Policy Prediction system. The proposed system automatically annotates the images using hidden Markov model and features are extracted by using color histogram and Scale-invariant feature transform (or SIFT) descriptor method. After annotating these images, semantic retrieval of images can be done by using Natural Language processing tool namely Word Net for measuring semantic similarity of annotated images in the database. Experimental result provides better retrieval performance when compare with the existing system.