Image Compression Using AMBTC with Artificial Neural Networks

Image compression is one of the most crucial image processing techniques widely used in multimedia signal processing. Traditional lossy image compression techniques like block truncation coding (BTC) should still be improved to achieve higher compression rates. The developments of BTC mainly concentrate on improving the reconstructed image quality and reducing the bit rate. This study presents a new approach that uses absolute moment block truncation coding (AMBTC) with artificial neural networks (ANNs), which are used in solving several problems, like function approximation problems. The suggested approach aimed to minimize the bit rate further to increase compression's effectiveness. The simulation results demonstrated that the proposed method could achieve lower bit rates (BR) while maintaining acceptable PSNR values.