The Develop Of Real Time Object Detection System Based On Machine Vision

The image of the specific objectives is to conduct real-time detection machine vision applications in key technologies, Object detection involves two phases of processes: The verification of an object in image sequences and locating it. Object detection results depend on the design of algorithms to a great extent. Therefore, it is still a hot point to research an accurate and robust algorithm in computer vision. Object detection algorithm is essentially a type of combinatorial optimization problem, in the traditional object detection algorithm based on various heuristic optimization joined by more and more attention. The issue stems from the Guangdong Provincial Natural Science Foundation (07001748): "Based on the Election Algorithm of Parallel Multi-object Detection Method."This dissertation will object detection algorithm is applied to the Election Algorithm, in the algorithm on the basis of the study, developed a machine vision based on the real-time object detection system the core module. Main contents include the following aspects:First of all, according to the actual needs of the project, select the appropriate equipment, in the advanced development platform, using existing software and related hardware configuration, built machine vision systems development platform, to achieve the objectives of the real-time detection;Secondly, on the Election Algorithm of thinking, framework and characteristics, a detailed summary of the Election Algorithm of calculation process, and the use of genetic algorithms function of the standard tests to verify the Election Algorithm;Thirdly, outlines the image segmentation and image matching the definition , use and its development status and trends, and on this basis by the Election Algorithm based on the adaptive algorithm block image segmentation algorithms and image matching algorithms, and using standard test image and the performance of the algorithm effects of the experiments and analysis;At last, summed up the object detection involves several key technologies :Image feature extraction, object modeling and object detection algorithm design, based on the Election Algorithm in the adaptive algorithm block image segmentation algorithms and image matching algorithm based on the research, design the object detection algorithm; In the OpenCV image processing functions on the basis of this dissertation, with the object detection algorithm using VC + +6.0 for development tools, in the Windows XP operating system, based on the development of machine vision technology object detection system the core modules: Image matching module, and a verification system, using standard test image on the algorithm performance and results were verified. OpenCV is Intel (?) open-source computer vision library, a series of C function and a small amount of a category C + +, and the image processing and computer vision of many common algorithm.In the final dissertation, a summary of the full text, and further study the direction and prospects were discussed.