Moving Target Detection In Image Sequences

Detection and tracking moving targets in serial images is the mostly research field in Computer vision and image coding. It also has applications in automaton navigation, Video Surveillance and monitoring Medicine image analysis and Video image compressing and transmitting.Serial images can be divided into two types by which background is still or moving. There are many reports in the first type and these reports mostly base on frame subtraction and background modeling. The other type is rarely researched and far away from matures. We discuss them all in this paper.Some common approaches in the field of detecting and tracking moving targets including the approach based on character, the approach based on optic flow, the approach based on frame subtraction and the approach based on background modeling are also introduced and compared with each other. Their advantage and disadvantage also with the field they fit are pointed out in this paper.With the background modeling method, a system of detection and tracking moving target for the static camera is realized there. The approach that is presented in this paper is applied to this system and gets a preferable result.In detecting moving targets in serial images with moving background, we first use image registration method to cancel the background's moving then adopt frame subtraction method to detect moving targets. In image registration, we put forward the concept of singular feature point and adopt it in image registration. Then we apply image registration to a system which detecting moving targets in serial images with moving background and get a preferable result.