Automatic Control Of Chamfering Machine using Machine Vision

This paper proposes a solution to an Industry related problem by automating the process of chamfering. Before the chambering process takes place there is a process for laser engraving where a series of alpha numerical characters are being engraved in the metal sheet. The workers have to manually load the metal sheet on to the table and while loading the sheets they have to check for the laser engraved characters and it should be present on the top of the metal sheet. When the sheet is being put on the table, the machine(motor) is turned on and the process of chambering takes place. While loading the sheet, the workers sometimes unknowingly tend to put the sheet upside down so that the laser engraved part is on the opposite side. This makes the sheet totally unusable. For avoiding this problem, we have used Image processing where, a camera is used for the detection of the characters in the sheet. The machine gets turned on through a relay module only when the camera detects the presence of the characters and if it is not detected it does not gets turned on and remains in off condition. When the plate is not detected by the camera then a LED light is being setup so that it acts as an indication for the workers that the plate is being loaded upside down. By this process we can avoid the possible human error and thus the metal sheet can be saved. A suitable algorithm for detecting the laser engraved characters is being developed using popular OCR algorithm like Tesseract, and the filter used to detect the edges was Robert edge filter which was used to quickly compute the results and to reduce the idle time. Overall, this paper presents a solution to the problem by implementing suitable filters for Image processing and also to find out the accuracy of the applied filter.