The Technology Of Multiple Cameras Panoramic Image Mosaicing Real-time Video Processing

Nowadays,with the development of science and technology,people's demand for electronic products is getting higher and higher.In real life,many electronic products have reached or even exceeded the needs of people,For example,in recent years,the camera industry has been developing rapidly.Among them,video surveillance has been widely used in daily life,traffic surveillance and even military field.However,the existing camera still can not meet the needs of panoramic shooting.The narrow angle of the camera has brought a lot of obstacles to the development of video surveillance.However,due to the limited range of photographing,a single image acquisition device is very likely to be impossible to achieve the need for clearer and higher resolution images.Therefore,in order to get a clearer and high resolution image,a multiple image acquisition can be used to collect images and complete the stitching a panoramic image.In order to solve the problem of a series of multi view monitoring images in the traditional monitoring system,the real-time panoramic monitoring system can be realized by using multi camera real-time splicing technology,which can be used in the field of security monitoring and control.The real time video processing technology of multi camera panoramic image mosaic and the architecture of multi camera panoramic image real time mosaic system are the main research and introduction of this paper.The basic theoretical knowledge and research background of real-time video processing technology for multi camera panoramic image mosaic is the content of the first and first brief introduction.After that,we set up a real-time mosaic system for multi camera panoramic images and provide a simple analysis and functional description of the overall scheme and architecture.Then,the key points of image registration and image fusion are discussed in detail,which are the key links in the two multi camera panoramic image mosaic techniques.The common method of image registration and image fusion is introduced and analyzed.At the same time,the combination effect and speed of the method are compared and analyzed with the aid of experiments.The principle and implementation of the SIFT feature matching algorithm are introduced and analyzed.The image registration part is completed by using the robust RANSAC algorithm and the feature matching of the LMedS algorithm.The image fusion part uses the nonlinear parameter weighting method to ensure the image fusion quality and the contrast analysis of the image and then the color of the image.Make corresponding correction to ensure the effect and speed of panorama.At the end of this paper,the design of the multi camera panoramic real-time image mosaic system is tested,analyzed and verified.It is proved that the proposed scheme is feasible and the effect of panoramic image splicing is better.After that,we can study the technology related to the real-time mosaic system of multi camera panoramic images.Application development provides some experience and reference value.