Image Classification Using a Deep Convolutional Neural Network

This paper presents an efficient way to use deep convolutional neural networks (CNNs) to improve image classification systems’ performance. CNN automatically extracts local and global features from the normalized image. Different convolutional neural network configurations are used for classification, and an experimental study was conducted to assess the efficacy of the proposed system for image classification on CIFAR-10, CIFAR-100, and STL-10. The system has achieved high recognition rates when compared to state-of-the-art methods in this field.