퍼지 추론을 이용한 비젼 시스템의 영상인식 알고리즘 및 제어모듈 구현에 관한 연구

Abstract

Fuzzy inference is an important issue in research on the fuzzy set theory. Because of the suitability for representing uncertain values, fuzzy numbers have been widely used in many applications.

In this paper proposed detecting specific position of the object using fuzzy inference and image processing.

Recently instead of classify the object by shade using existing analog sensor, apply a theory to detect of the object information by using image data from CCD camera. It takes interest in the morphology of image, fuzzy inference and image analysis system on knowledge base. and now days, image processing method was used in the most part of field. So, in this paper proposed detecting method on specific position of specific object(fish) about real time gray level input image using image processing method(histogram analysis, binary, erosion, dilation, projection) and application case in intelligent processing system using the fuzzy inference algorithm and based on the computer with machine vision system. We obtained dynamic threshold value, contrast value, brightness value ,erosion and dilation value using fuzzy inference.

This processing system is inspected by IBM-PC interface digital I/O card, image grabber board, micro-controller and sensor, etc, so the operating data(the position of cutter, cutting operation, conveyor of movement, detecting fish, observation of sensor data, transmission of cutting data, etc) of processing system are monitoring and control on IBM-PC monitor during the processing time.

A view of a component control machine in this paper, one frame image as captured by CCD camera of effective pixel resolution 320×240. If the positioning of the carriage is based on accurate gauging of the desired position, the image contains the necessary information.

Finally, a control system processing work-cell was presented as a system to which the model could be applied. The results of a practical implementation of the system were given. We conformed that our study had better performance than conventional one.