فهرست مطالب وجیهه ظریف

Nowadays, close competition and high cost of production cause the loss reduction and waste recovery to be one of the most concern of industrial and production companies and factories. For this, managers are looking for new methods to reduce the defect in the product line of the factories. Image processing and machine monitoring systems are technologies used in different fields of military, physician, agriculture, industry, etc. Purpose of this study is to use an image processing model to recognize the defect in the product line. As a case study, oil bottles produced by Shiraz Vegetable Oil Company are used. Input data to the program is the images of the intact and defected oil bottles and output is the final judgment of program about the correctness of the bottles. Image processing is performed using software MATLAB. In this study two different procedures are used to identify the intact and defected oil bottles. The first theory is based on the comparison of the area of intact and defected bottle images and the second theory compares the ratio of the height to the sum of some widths for two different images. Second theory uses edge function algorithm and the result obtained from this theory is more optimize with respect to the first theory. This theory is programmed without using the image of an intact object as reference image. This is one of the advantages of second theory with respect to the first one; because a small defect in reference image of the first theory can strongly affect the results. This research is also, consist of two recommended theories. The first theory is based on a reference point in the image and summarizing some of the distances from the edge of the bottle to the reference point. Such, that acceptable results can be obtained by comparing the summarized distances in the intact and defected bottle. The second recommended theory is based on the comparison of the center of area in the intact and defected bottles.