Discriminative Canonical Correlation (DCC) Based Object Detection in Images Using Machine Learning

Abstract

Efficient and accurate object detection has been an important topic in advancement of computer vision systems. Object detection is basically invariant to the dramatic changes caused in objects’ appearance such as location, size, viewpoint, illumination, occlusion and more by the variability in viewing conditions. This paper describes the Discriminative Canonical Correlation (DCC) based Object detection in images using Machine learning. The invariant features namely shape and texture invariant features of the objects are extracted separately with the aid of suitable feature extraction techniques. The pattern recognition algorithms, like Discriminative Canonical Correlation (DCC) and Machine learning algorithm as Support Vector Machine (SVM) is employed. A large collection of object images consists of 1000 objects recorded under various imaging circumstances. The experiments clearly demonstrate that this presented approach significantly outperforms the state-of-the- art methods for combining shape and texture features. Accuracy of Discriminative Canonical Correlation (DCC) and Machine learning algorithm showed the significant improvement in object detection.