ACTIVITY RECOGNITION IN EXAMINATION HALL TO PREDICT UNFAIR MEANS

Abstract

Automatically recognizing human’s physical activities (a.k.a human activity recognition or HAR) has emerged as a key problem to ubiquitous computing, human interaction, and human behaviour analysis. In Our project we will be using computer image processing technique to classify and categorize a person s activity in an examination hall and on the basis of it forecast whether the person is involved or is going to be involved in unfair or unjust activities in an examination hall using video sequences. The need for such systems is increasing every day, with the number of (hundreds or thousands) of surveillance cameras deployed in public spaces, massive number of cameras calls for systems able. Our work focuses on three fundamental issues: (i) Collecting Dataset; (ii) Detecting Edges; (iii) Features This Selection; (iv) Defining Structure of a classifier. Our project will use MLP (Multilayer Perceptron) for classification purpose. For training MLP utilizes backpropagation, a supervised learning technique. The data that is not linearly separable can be distinguished. Our project’s main objective is to provide an easy way to predict use of unfair means during examination and to help achieve true merit. Many organizations, educational institutes and government sectors can ensure and provide capable and truly deserving employees/workforce for their respective institutes.