Research On Digital Image Stabilization And Panoramic Image Enhanced Display Technology Based On Vehicle Platform

Video stabilization,panoramic image stitching,moving object detection and enhanced display technology have always been very hot topics in the field of imaging.With regard to the low degree of instrument integration and single function in this area,integrating the three functions of image stabilization,image stitching,and moving target detection and enhanced display into one set of equipment is the trend of instrument development.The integrated equipment has a huge application market and Value.The dissertation focuses on the sight-seeing system and uses digital image processing technology to optimize the structure of the existing sight-seeing system.For its application scenario,a real-time video image stabilization algorithm combining optical flow method and Kalman filter is proposed to solve the problem of video jitter in the process of vehicle platform.A panoramic image stitching algorithm was designed to extend the function of the sight-seeing system.According to the characteristics of human vision,combined with the acquired digital panoramic image,using the background difference method,the detection of the moving target in the scene and the color enhanced display are realized.Replace the Dove prism in the sighting sight system with a digital image processing method to obtain an always upright image;In the image stabilization algorithm,the optical flow method is used to calculate the local motion vector of the video image.In combination with the kalman filter commonly used in the control field,the active motion component and the random jitter component in the global motion are separated,and only the active motion is obtained by the digital image processing technology.Stable video sequence,stable video motion smooth and natural,clear and smooth picture,real-time performance,and meet the basic requirements of video stabilization;In the method for stitching images of peripheral vision,a panoramic image is stitched using an absolute rotation angle of a sighting sighting system.In order to ensure the integrity of the information,the patchwork in the panoramic image is not smoothed,and the stitched image is complete and natural.The moving object detection uses a simple background difference method,uses the open operation in morphology,and easily removes the pseudo matching point in the difference image,and has no effect on the true moving target with a size greater than 1 m;According to the characteristics of human eyes that are more sensitive to red information and fewer red targets in the application scene,the luminance value of the corresponding point of the moving target is superimposed on the red channel of the original image,and the specific movement objects before and after the movement are marked.position;Using opencv’s image display function and mouse clickback function to create a simple interactive interface,the interface operation is simple,and successfully integrates the three functions of video image stabilization,circumferential image stitching,motion object detection and enhanced display in panoramic images.Overcome the single drawbacks of the instrument function.Simplifying the structure of the system,and developing the system toward the direction of integration,compactness,digitization,and electronics,is in line with the mainstream of modern instrument development.Improved sight-aiming systems have great application prospects in security,surveying,and mapping.