ABSTRACT
ABSTRACT: Object category classification is one of the most difficult tasks in computer vision because of the large variation in shape, size and other attributes within the same object class. Also, we need to consider other challenges such as the presence of noise and haze, occlusion, low illumination conditions, blur and cluttered backgrounds. Due to these facts, object category classification has gained attention in recent years. Many researchers have proposed various methods to address object category classification. The main issue lies in the fact that we need to address the presence of noise and haze which degrades the classification performance. This work proposes a framework for multiclass object classification for images containing noise and haze using a deep learning technique. The proposed approach uses an AlexNet Convolutional Neural Network structure, which requires no feature design stage for classification since AlexNet extracts robust features automatically. We compare the performance of our system with object category classification without noise and haze using standard datasets, Caltech 101 and Caltech 256.
