FPGA implementation of a predictive vector quantization image compression algorithm for image sensor applications

This paper presents a hybrid image compression scheme based on a block based compression algorithm referred to as Vector Quantization (VQ) combined with the Differential Pulse Code Modulation (DPCM) technique. The proposed image compression technique called the PVQ scheme results in enhanced image quality as compared to the standalone VQ. The generated codebooks for the PVQ scheme are more robust for image coding than that of the VQ. This made our system a suitable candidate for developing on chip image sensor with integrated data compression processor. The proposed system was validated through FPGA implementation. The resulting implementation achieved good compression and image quality with moderate system complexity.