FPGA-based CMOS Video Image Enhancement Algorithm Research And Application

In this present era of mobile Internet and large data, the video image related applications in various fields. the video image acquisition process vulnerable to the impact of the changing external environment, resulting in image contrast is low, dark gray and gray-scale dynamic range contraction and other issues degradation of image quality caused by the need to enhance image processing. Image enhancement, while maintaining the original semantics, targeted enhancement, improved video image quality, ease of subsequent analysis and processing. Based on the research of image enhancement algorithms, the paper proposed an improved adaptive scene transform image enhancement algorithm, combined with FPGA technology research, using Xilinx’s Virtex-4SX series FPGA hardware implementation of adaptive enhancement algorithms.This paper studies from CMOS image sensor technology, analyzes the principle and the image sensor pixel structure, the choice of video capture OV8825CMOS chip as the front-end chip, combined with SCCB bus control technology, drive control.Video enhancement algorithm is the theoretical basis of the paper, the date of birth on the image enhancement algorithms from experiencing continuous research and improvement, from the traditional classical histogram equalization algorithm to fuzzy theory enhancement algorithm are improved in the study. On the basis of research, the paper proposes an adaptive scene video image enhancement algorithm, the algorithm divides the scene into too dark, too bright, ideal for three different type, propose appropriate transformation functions for different scenarios, referring to different scenarios to determine the threshold of a type different gray, function transformation performed according to the judgment result, completed adaptive image enhancement.Theoretical studies need support platform, combined with FPGA technology, Paper presents a general framework and ideas adaptive enhancement algorithms on FPGA module, Completed the circuit design and FPGA peripheral interface configuration mode, Use ChipScope Pro line debugging tool to complete initial performance debugging platform.Finally, the specific design of the adaptive enhanced FPGA implementation, and is divided into two steps to complete the design, front-end to complete the acquisition module design, simulation debugging, provides the signal source for the subsequent video processing; complete the core processing algorithm design, simulation debugging read out and displayed by Matlab software. Compared with the software implementation of the algorithm processing, consistent results, but the processing speed increased more than one level, real-time video processing requirements.