Boundary Detection Algorithm for Tracking Coloured Objects using Opencv

Abstract: In this research work, an open CV based image processing algorithm proposed to track objects in image. In the present research work, a different approach is proposed for segmentation of image data in the presence of intensity inhomogeneities. This formulation allows the labeling of a pixel to be influenced by the labels (colored objects) in its immediate neighborhood. The neighborhood effect acts as a regulariser and biases the solution towards piecewise homogeneous labeling. In this work, optimal texture features are identified for classification of colored objects. The texture features are extracted from the ROI in images. The classifier is trained to classify the features into different colored objects.

Keywords: FL 2440 Board, OpenCV, Linux, C++