JPEG Image Compression using FPGA with Artificial Neural Networks

252 Abstract— Image and video compression is one of the major components used in video-telephony, videoconferencing and multimedia-related applications where digital pixel information can comprise considerably large amounts of data. Management of such data can involve significant overhead in computational complexity and data processing. Compression allows efficient utilization of channel bandwidth and storage size. Typical access speeds for storage mediums are inversely proportional to capacity. Through data compression, such tasks can be optimized. One of the commonly used methods for image and video compression is JPEG (an image compression standard).Image and video compressors and decompressors (codecs) are implemented mainly in software as digital signal processors. Hardware-specific codecs can be integrated into digital systems fairly easily, requiring work only in the areas of interface and overall integration. Using an FPGA (Field Programmable Gate Array) to implement a codec combines the best of two worlds. The implementation of this work is carried out with JPEG algorithm with Artificial Neural Networks (ANN). The core compression design was created using the Verilog hardware description language. The supporting software was written in matlab, developed for a DSP and the PC. The implementation of this work was successful on achieving significant compression ratios. The sample images chosen showed different degrees of contrast and fine detail to show how the compression affected high frequency components within the images.