Taking Acount Of People By The Use Of Image Feature Based On Simple And Fixed Backgroud

The technology based on image processing and machine vision is in the forefront of the Intelligent Recognition of image processing and it is the important research branch of artificial intelligence plant. The identificational method is to find a suitable model, and then using machines to achieve simple intelligent judgment. In this paper, we study the recognition and judgement of the target of human beings. We use the normalization program and an active contour to extract algorithm information of the object and image characteristics from sampling image .Then, we make use of some prior knowledge of this information to reach the judgement which cause to the identification of "people" Goals.Target detection is the premise of recognition. According to Gray weighted analysis we can find an interesting target which is close to us the image point (that may be the goal points). On basis of this, we can establishmen a Cartesian coordinates and segment from the sampling image to solve the numbers of targets detection problem.Target matching and target tracking is the basic ideology of identification area, in the light of the issue of uncertained sampling image.It is hard to gain the position, Morphological differences in perspective .So matching the target directly is obviously not feasible. In this paper, incording to the idea of normalization program, such as angle correction, coordinate translation and telescopic coordinates, we can obtain standard sizes and shapes of objects. In a different with traditional algorithm ,it doesn't use standard templates as a reference. thus we can prevent the other targets from getting similar size and morphology of"people"targets.Contour detection is always be the is the most fundamental and most difficult problem of objective recognition algorithm, traditional contour detection program often suffers from the inactual results of sampling map greatly, which causes Fuzzy result of contour extraction and breakpoints in contour or even serious distortion of the contours.The usual solution is to adopt corresponding amendments to the follow-up algorithms for object contours to maintain continuous. Based on the goal of normalization and active contour algorithm issue in this paper, wo can solve the problem to get profile of the target close to his own continuity and void the complex follow-up rectification works.Under normal circumstances , the target recognition can often be solved by identifing the target or extracting its invariant feature as soon as possible. Thus, it will not affect judgment of other goals. However, "people" target patterns will have random changes over time. So it is not easy to extract its invariant features. Giving the statistical number of objectives is not allowed to repeat the same individual counts. Based on result of previous target detection and the identification of goals in current , we can use the integrated calculation method to determine the necessity of identifing and judging of the current sampling plan . In a result , we solve the problem of counting repeatly and reduce the amount of date in a real-time processing.