Real-time Restoration of Motion Blurred Image Based on Four DSPs
Liu Wei①②; Zhu Ming①; Li Xiang-rong③
①Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130031, China; ②Graduate school of Chinese Academy of Sciences, Chinese Academy of Sciences, Beijing 100039, China; ③Dept. of Mech. and Elec. Eng., Qingdao University of Science & Technology Qingdao 266061, China
Abstract：To absolve image blur phenomena caused by relative motion between camera and objects, a real-time blur image restoration method is put forward, which is composed of three fixed point DSP and one float point DSP. Degeneration model is researched, from the point of view of software, wiener filtering is adopted to restore image. Two-dimensional FFT principle and important questions are presented executed by fixed DSP, the respective functions of four DSP is explained during the course of image restoration. For the image of 512×512，the time of two-dimensional FFT by fixed DSP is 17.7ms, the gross time of float point multiplication and division by float point DSP is 32.51ms, according to above time, the load of every DSP is calculated. From these datum, the hardware can complete real-time restoration of 512×512 image.
2006, Vol. 28 
