The first contribution of this paper is architecture of a multipurpose system, which delegates a range of object detection tasks to a classifier, applied in special grid positions of the tested image. The second contribution is Gray Level-Radius Co-occurrence Matrix, which describes local image texture and topology and, unlike common second order statistics methods, is robust to image resolution. The third contribution is a parametrically controlled automatic synthesis of unlimited number of numerical features for classification. The fourth contribution is a method of optimizing parameters C and gamma in LibSVM-based classifier, which is 20-100 times faster than the commonly applied method. The work is essentially experimental, with demonstration of various methods for definition of objects of interest in images and video sequences.
@article{arxiv.1310.7170,
title = {Object Recognition System Design in Computer Vision: a Universal Approach},
author = {Andrew Gleibman},
journal= {arXiv preprint arXiv:1310.7170},
year = {2013}
}