英文-基于图像处理的碎片拼接.docVIP

A Fast Reassembly Methodology for Polygon Fragment A new general methodology is introduced for computer-aided reassembly of the polygon image fragments. Since polygon image fragment has a characteristic that it has obvious corners, this methodology takes advantage of such characteristic, and detects the corners by Hough transform and fuzzy reasoning. Meanwhile, it computes the length of each side, compares the side length, finds out the sides with equal length in different fragments, and then computes the gray correlation of equal sides to determine the most optimal matching location for reassembly. Experimental results show that the reassembly methodology for polygon fragment based on corner alignment and gray correlation is effective and fast. This paper appears in:?Computer Engineering and Technology, 2009. ICCET 09. International Conference on, Issue Date:?22-24 Jan. 2009, Written by:?Xu, Gang; Xian, Yi、 ? 2009 IEEE INTRODUCTION The problem of image fragment reassembly is a novel and typical application in the field of computer vision, pattern recognition and image analysis.?In the computer vision domain, it can be approximately viewed as the jigsaw puzzle problem which has been researched for some years. At present, the hard, tedious and repetitive works of reassembling image fragments are still completed artificially in the fields of reassembling information fragments, judicial authentication and cultural relic fragments repair. So how to use computer to finish reassembly is one of the important problems for many archaeologists and criminologists. The hot research topic aims at irregular curve edge fragment whose contour is unique.Therefore, we can reassemble the fragments using their contours. However, another type of fragment, irregular polygon fragment, has seldom been researched. The most usual example is paper fragment from paper shredder, whose sides are straight lines. Because the sides on polygon contour are the same object in shape for computer, if we