DCT-CompCNN: A Novel Image Classification Network Using JPEG Compressed DCT Coefficients

The popularity of Convolutional Neural Network (CNN) in the field of Image Processing and Computer Vision has motivated researchers and industry experts across the globe to solve different challenging research problems with high accuracy. The simplest way to train a CNN classifier is to directly feed the original RGB pixel images into the network. However, if we intend to classify images directly with its compressed data, the same approach may not work better, like in case of JPEG compressed images. This research paper investigates the issues of modifying the input representation of the JPEG compressed data, and then feeding into the CNN. The architecture is termed as DCT-CompCNN. This novel approach has shown that CNNs can also be trained with JPEG compressed DCT coefficients and subsequently can produce a good performance similar to the conventional CNN approach. The efficiency of the modified input representation is tested with the existing ResNet-50 architecture and the proposed DCT-CompCNN architecture on a public image classification datasets like CIFAR-10, Dogs vs Cats and MNIST datasets, reporting a better performance.