Research Of Stabilization Technology Based On Video Image Sequence

Video image stabilization technology has a wide range of applications in military investigations,civil video monitoring and the industry.Video image stabilization technology also has been gradually concerned by us.Although it is in the initial stage of medical and health,but it has a good development prospect.And because of its special needs,the resolution of the video must be bigger,the picture definition must be higher.More stable and jitter-free video.Based on this,we designed a suitable for high-resolution real-time image processing equipment.Firstly,this article analyzes the development of video image stabilization technology and development status of domestic medical industry,illustrates the video image stabilization in various fields of application and development prospects.Then discusses the basic principles of video image stabilization technology and algorithms,focusing on a detailed introduction to motion estimation,comparison and analysis of advantages and disadvantages of motion estimation algorithms.Then shows video image stabilization versatile system.Secondly,on the basis of traditional FAST,presents an improved algorithm for threshold adjustment,and for feature point matching and selection,ensuring the accuracy of feature extraction and matching.Motion filtering algorithm based on Kalman filter is designed,a distinction between the normal movement of the camera and random jitter.Thirdly,the motion compensation using nearest neighbor interpolation,the stable output of the video.Finally,the DSP+FPGA-based hardware platform into DSP algorithm(TMS320C6455).Design of a high-resolution real-time image processing equipment.Using FPGA(V5SX50T665)for image pre-processing,and logical sequence and external interface communication,effectively improve the processing speed of the signal processing equipment.From experimental results,the proposed algorithm stabilizes video,meets the requirements of real-time performance and reliability.