Font Size: a A A

The Research Of Electronic Image Stabilization Technology Based On FPGA

Posted on:2013-07-17Degree:MasterType:Thesis
Country:ChinaCandidate:J LiFull Text:PDF
GTID:2248330374951645Subject:Signal and Information Processing
Abstract/Summary:PDF Full Text Request
With the rapid development of information technology, image stabilization technology has been paid extensive attention in many fields. Electronic image stabilization is a new generation of image stabilization technology after mechanical image stabilization and optical image stabilization, and it stabilizes the jitter of the video image sequence by using electronic equipment and image processing technique. Compared with traditional image stabilization system, its character is small power, high accuracy, carry conveniently and widely used in the military field and civilian equipment. The research on electronic image stabilization has become an important subject.In this paper, the basic principle and treatment process of electronic image stabilization are elaborated. The gray projection algorithm for motion estimation as the image stabilizing algorithm has been studied, and the algorithm is improved in order to adapt to the difference between rotating and a small moving object in background. Electronic image stabilization system based on FPGA is designed and implemented. The specific research contents are as follows:Firstly, electronic image stabilization system framework diagram based on FPGA is designed based on the analysis of the electronic image stabilization principle, system composition and evaluation method, combined with a variety of image stabilizing algorithm design characteristics.Next, focus on the gray projection algorithm for image stabilization. Using MATLAB as the algorithm research platform, algorithm is analyzed In accordance with the pretreatment, gray projection, projection filter, motion estimation and motion compensation process. In view of the traditional gray projection algorithm can only detect the displacement vector one horizontal and vertical directions, the algorithm is improved. The improved gray projection algorithm can detect the image rotation angle; combination of block matching and gray projection method to eliminate the effect of moving objects in motion estimationThen, electronic image stabilization platform based on Virtex2Pro is designed. Xilinx’s Virtex2Pro FPGA integrated within the two PowerPC405core processors, using the IP implant and CoreConnect interconnect technology, the system structure of software and hardware are very flexible, functional components implementation is also simple. Taking PowerPC as processor core, image stabilization system hardware platform is built in Xilinx EDK XPS, including IP calling, DDR SDRAM interface, VGA display and the bus connection, and the key part of the test.Finally, implemented electronic image stabilization system based on Virtex2Pro. After hardware platform building successfully, algorithm developed by VC6.0was developed again in the SDK environment. The algorithm was transplanted to the hardware platform, and the electronic image stabilization system was implemented. Jitter image sequences which ware shot in laboratory environment was treated by using the stabilization system, and achieved good results of image stabilization.
Keywords/Search Tags:electronic image stabilizing, Gray projection algorithm, Virtex2Pro, motion estimation, motion compensation
PDF Full Text Request
Related items