Master of Electrical Engineering (MEE)
Electrical and Computer Engineering
First Committee Member
Number of Pages
This thesis presents the architecture and design of a JPEG compressor for color images using VHDL. The system consists of major parts like color space converter, down sampler, 2-D DCT module, quantization, zigzag scanning and entropy coDing The color space conversion transforms the RGB colors to YCbCr color coDing The down sampling operation reduces the sampling rate of the color information (Cb and Cr). The 2-D DCT transform the pixel data from the spatial domain to the frequency domain. The quantization operation eliminates the high frequency components and the small amplitude coefficients of the co-sine expansion. Finally, the entropy coding uses run-length encoding (RLE), Huffman, variable length coding (VLC) and differential coding to decrease the number of bits used to represent the image. The JPEG compression is a lossy compression, since downsampling and quantization operations are irreversible. But the losses can be controlled in order to keep the necessary image quality; Architectures for these parts were designed and described in VHDL. The results were observed using Active-HDL simulator and the code being synthesized using xilinx ise for vertex-4 FPGA. This pipelined architecture has a minimum latency of 187 clock cycles.
Compression; HDL; Image; Implementation; JPRG; Pipelined
University of Nevada, Las Vegas
If you are the rightful copyright holder of this dissertation or thesis and wish to have the full text removed from Digital Scholarship@UNLV, please submit a request to email@example.com and include clear identification of the work, preferably with URL.
Toomu, Arun Kumar Reddy, "Pipelined implementation of Jpeg image compression using Hdl" (2008). UNLV Retrospective Theses & Dissertations. 2387.