Research on Image Recognition of Extension Strategy Generating System
|Course||Mechanical and Electronic Engineering|
|Keywords||Image Recognition Extenic ESGS Extension Model|
The technology of image recognition is a kind of new-style technology, which is developed due to the theory of present computer technology, image processing, artificial intelligence and pattern recognition. Its primary research content is to study the classification and description of some objects or process. To develop the system which could process some information automatically for achieving image classification and identification is image recognition purpose. The field of investigation in the image recognition is far-ranging. For example：the license plate recognition in the system of traffic monitor; identify the spare parts in the machine processing; find out the ailing cells via medical image; identify the forest,lake and some other specific facility through the remote sensing images; sorting the letters from post system;the recognition of face and fingerprint; signature identification and so on. Image recognition from the acquisition, pretreatment, segmentation to the final recognition is a subjective selection algorithm. Based on the study of the Extension Strategy Generating System (ESGS), this paper sought to use the extension theory, combined with the image processing technology for image recognition method.Firstly, the theory of extenics and related algorithm of image recognition is analyzed. Based on knowledge representations, the extension model of image recognition system with basic elements is set up. The divergence analysis is carried out by using the extended properties of the elements, achieving to describe information and knowledge formally.After that, combined with database technology, the knowledge database is set up in a established extension model. The problems related with current knowledge are processed. The scheme knowledge is generated by the extension reasoning principles. Combined with genetic algorithm to optimize the path, the solution to this problem is generated and assessed by priority evaluation, in order to generate the optimal strategy system.Finally, based on the above-mentioned theory and algorithms, the friendly ESGS of image recognition is developed including system management, image acquisition, scheme generation and system evaluation on Visual Studio2008platform. The results of this study will further validate the feasibility of the extension in combination with artificial intelligence, which can develop more and better intelligence software, and further enrich the ESGS.