Abstract: In this paper a real time vision based system is proposed to identify the defect in the square washer. The entire frame work is based on the Arduino board and MATLAB. The images of square washer are taken from camera which is placed on top of conveyor belt. This images are processed in MATLAB. To extract the feature of washer area, height & width of washer is calculated and also area of hole is calculated. SVM is implemented for the classification of faulty and not faulty washers. The classified output is given to the Arduino board which will control the rotation of motor (motor is connected to conveyor belt) in clockwise / anticlockwise direction depending on not faulty / faulty washers. An LCD display is used for displaying status of washer. This paper attempts to study the defect detection technique on computer vision (MATLAB) platform which implements an industry assembly line using image processing.

Keywords: Conveyor Belt, Arduino, MATLAB (MATrix LABoratory), SVM (Support vector machine).