Hall Ticket Fraud Detection System using ANN Classifier

Document image analysis is a major research direction in the field of image recognition systems for identifying and classifying image objects in documents. Facial recognition is the most widely used and effective system in biometrics. Existing methods have been working on document image analysis and face recognition areas separately. This paper attempts to developa new system that verifies the identity of the person taking the test by comparing the name, seat number, and photo of the ticket. Broadly speaking, the lounge ticket identification system consists of 5 steps: pre-processing, segmentation, face detection, feature extraction and ANN classification. For feature extraction, we use 10 shape features from a new tessellation of the lounge ticket image and 6 features from the object appearance histogram, and pass these features to an ANN classifier to detect lounge ticket fraud. The proposed algorithm was tested on data sets from two different institutions with an average accuracy of 94.12%.