✍ Scribed by Wenhua Wan; José A. Ventura
No coin nor oath required. For personal study only.
quirement, in addition to the high cost of equipment. Thus, it is important to develop machine vision techniques for This paper presents a procedure for segmenting planar curves, mainly the projected boundary contours of machined automated recognition and inspection of 3-D objects (espeparts, into straight-line segments and elliptical arcs. The break cially machined parts) using 2-D images, so that reliable points are divided into two types: corners and smooth joins. real-time quality control activities can be achieved in auto-The corners are detected by first applying adaptive smoothing mated manufacturing systems. Segmentation of digitized to the tangent orientation along the curve, then taking the planar curves is one of the most important elements in derivative of the smoothed tangent orientation, and finally loearly image processing, because a segmented curve can cating the high spikes on the derivative. The smooth joins are describe the object profile in a meaningful and compact first roughly located by a dynamic focusing fitting technique form to facilitate higher level vision processing, such as and then refined by an adjustment algorithm. The dynamic pattern recognition, shape matching, and geometric inspecfocusing fitting technique holds one end of a curve segment tion. In industrial applications, more than 90% of machined (which is bounded by a pair of adjacent corners) fixed and parts are bounded by flat and quadric surfaces [2]. Their scans it from the other end until it focuses on a component 3-D perspective projections are bounded by straight-line segment which fits either a straight line or an elliptical arc. This segments and elliptical arcs. Hence, we are interested in component segment is identified and the process is repeated in developing a procedure for segmenting digitized curves by the same manner for the rest of the curve. In the refining stage, each smooth join is adjusted to the left or to the right, point dividing the edge pixels of an image into intervals where by point, until the measurement of goodness of fit for the curve the data can be approximated by either straight-line segsegment is optimized. Tests of the procedure were performed ments or elliptical arcs.
with the boundary curves of three real object images.
📜 SIMILAR VOLUMES