REAL-TIME COLOR CLASSIFICATION OF OBJECTS FROM VIDEO STREAMS

Real-time Color identification and differentiating those colors of a moving object from a live video is a fundamental step in many real-time applications such as video surveillance, bio-metric identification process, etc., Object Detection is based on two parameters such as Objective and Subjective dimensions. Where the object consists of attributes like color, texture, shape, size and scale, however the subjectivity is guided by the observation of the image interpreters. Due to brightness and contrast sometimes webcam can hardly detect the expected color of the objects and because of the similarity of tracking, environment background color and the color of objects gets unexpected pixel value. This project will aim to implement object detection and color classification in MATLAB GUI. Where the objects will be detected and the color of the object will be classified in a real world scenario for both still and real-time image. This task includes image segmentation process for still image, where only the color and the count of objects will be recognized from a still image. However, for live video processing the region of the object as well as the color of the object will be recognized