Down Category Recognition System Based On Support Vector Machines

Down is the ramification of birds'skin, and it mainly keeps them warm besides helping flying. Because of its characteristics, light, soft, warm, and so on, nowadays, it becomes more and more popular with people in every country.Now, down category recognition is often done by man with a microscope. This kind of identifications demands a great deal of training and practice experiences, at the same time, the people who examines downs, must know well about the characteristics of goose down , duck's down, chicken's down and so on. Even more badly, many artificial factors in the process may result in different examination results of same sample. Furthermore, if someone examines the down for a long time, he will feel very tired, which makes even more faults. So, developing down automatic recognition equipments is badly needed for the sake of improving the objectivity and accuracy. But now, at home and abroad, down automatic recognition system is not reported.On the base of analyzing and tallying up the structure of goose down and duck's down under the microscope, after researching, we develop an automatic recognition system for down based on SVM. In the paper, our main work has been done as follows:1. According to analyzing and tallying up the structure of goose down and duck's down under the microscope, we get the judging characters that are fit for the automatic recognition of computer.2. Study the methods of image processing of the image of down. Methods of image enhancement, image smoothness and increasing contrast are mainly researched, and we get the suitable method for the image of down. After researching the image segmentation, we get the two-value image of the image of down with the threshold of it.3. On the base of researching theories of SVM and morphology, a new method of down automatic recognition based on SVM and morphology is proposed. The triangle nodes of two-value image of down is to be recognized with SVM, then the triangle nodes which have been recognized will be matched each other and the distance between the matched triangle nodes is calculated, in the end, the down category is recognized.4. The whole project of the system is proposed and the software of down category automatic recognition system is designed. In the end, we develop a system with friendly interface and simple operation with VC++ 6.0.In the paper, the down category automatic recognition is implemented using the technology of image processing, SVM, morphology and so on, and the system is developed by VC++6.0. The system can recognize the down category without manual work. It can get over some disadvantages and inconveniences of manual identification, and decrease the faults of manual identification. Besides, the system can improve the level of down category recognition and our study has a wide range of further use and practicality. Even it fills up the blank in the domain at home and abroad.