Hardware JPEG Decompression

Dan MacDonald, University of Windsor


Due to the ever increasing popularity of mobile devices, and the growing number of pixels in digital photography, there becomes a strain on viewing one's own photos. Similar to Desktop PCs, a common trend occurring in the mobile market to compensate for the increased computational requirements is faster and multi-processor systems. The observation that the number of transistors in integrated circuits doubles approximately every 18-24 months is known as Moore's law. Some believe that this trend, Moore's law, is plateauing which enforces alternate methods to aid in computation. This thesis explores supplementing the processor with a dedicated hardware module to reduce its workload. This provides a software-hardware combination that can be utilized when large and long computations are needed, such as in the decompression of high pixel count JPEG images. The results show that this proposed architecture decreases the viewing time of JPEG images significantly.