A robust digital image processing method for measuring the planar burr length at milling

Abstract

The burr formation is one of the difficulties at the cutting processes; moreover, its size and characteristics show changes along the edge; therefore, it is hard to measure. The main aim of the paper is to develop a novel algorithm that is suitable for measuring the planar burr length of such contours that cannot be scanned below. The proposed method uses a digital microscope attached to a machine tool and using existing image processing algorithms in combination with novel approaches. The advantage of the current method is that, it allows collecting the height of burr along the workpiece automatically since it is able to handle and avoid interruptions, branches and provide uniform thickness for the determined contour. Another advantage is the extended measurement ability for workpieces which contains inner contours and cannot be illuminated from below. The method was tested with sample images and also with face milling experiments, and the results were validated with the data originating from a coordinate measurement machine. The result shows a good agreement at the validation. The available accuracy mainly depends on the applied measuring equipment (resolution and magnification of camera) and set up. The theoretical accuracy of the proposed method can even approximate the magnitude of the pixels. In this research work, the achieved accuracy was ±0.06 mm in the best case, and ±0.1 mm in the worst case with the used experimental device.