Image recognition with neural classifiers in micromechanics and agriculture

Neural classifier for multi-purpose image recognition systems is developed. One of the applications is an adaptive control system based on image recognition system for micromechanics where the limited receptive area (LIRA) neural classifier is proposed for texture recognition of mechanically treated metal surfaces. The performance of the proposed classifier was tested on image database with four texture types corresponding to metal surfaces after milling, polishing with sandpaper, turning with lathe and polishing with file. The promising recognition rate of 99.8% was obtained. Another application is agriculture, where vast amounts of pesticides are used against the insects. In order to decrease the required amount of pesticides it is necessary to locate the precise form distribution of the insects and caterpillars. In this case the use of pesticides will be local. In order to automate the task of recognition of caterpillars we propose to use a web-camera and a personal computer. In this article we propose the LIRA neural classifier for the tasks of recognition of texture images of metal surfaces and texture images of caterpillars. We use an image database of caterpillars of different forms, sizes and color, distributed in different amounts and positions. The main idea is to recognize the difference between the textures corresponding to the caterpillars and real world background. In the article we present the structure and the algorithms of recognition with LIRA neural classifier, and results of its application in both tasks. Key-Words: texture recognition, LIRA, neural classifier, micromechanics, agriculture, larvae recognition