A Dynamic JPEG CODEC with Adaptive Quantization Table for Frame Storage Compression

Joint Photographic Experts Group (JPEG) is one of the most advanced compression formats for still images. However, JPEG is still limited and inefficient in real-time image processing and hardware implementation. This paper presents a dynamic JPEG CODEC (the encoder and decoder) to compress the frame storage. Based on the pipeline structure, the system uses ping-pong buffers to support the circuit across two clock domains. In addition, through dynamically adjusting the quantization table, the frame storage compression can fluctuate around 80% with high quality. Also, the simplification of compression decreases the initialization time of the decoder. The proposed word can yield a reduction of the frame storage of 76.60% and an improvement of the latency of 77.10% compared to the baseline JPEG CODEC.