基于儿童生长发育检测的图像切割拼接拟合算法

Image Cutting Splicing Fitting Algorithm Based on Child Growth

摘要: 在医学中，判断儿童骨骼发育健康状况是一个非常复杂的过程，而儿童髋骨健康生长，更是影响儿童生长发育的一个重要因素。因此对问题骨骼的早期检测判断对儿童的生长发育有决定性的影响。传统的骨骼成长状况检测单纯依靠人眼估算CT、X光图片，不能定量精确地确定对骨骼的发育情况进行评测，无法为后期医学诊断提供丰富的参考数据。因此结合图形图像学知识，对骨骼生长的医学检测方法的研究是医学研究中一个热点。本文利用MATLAB在图像处理方面的强大优势，在计算机上实现了对儿童髋骨图像快速的切割、处理、拼接以及最佳点拟合检测，并通过计算两块骨头所形成的圆弧所拟合出椭圆的圆心的吻合度来分析两块骨骼的吻合度，为判断儿童骨骼发育状况提供信息。

Abstract: In medical research, how to judge children’s bone health is a very complicated process. Healthy growth of children’s hip bone is also an important factor which affects children’s growth and development. So early detection and judgment on bone problems have a decisive influence on children’s growth. Traditional bone growth detection relies on eye estimation of CT, X-ray images, which can’t quantitatively and accurately determine bone development and can’t provide a wealth of reference data for late medical diagnosis. Therefore, combined with image knowledge, the study on medical detection method of bone growth is a hot spot in medical research. Using powerful advantages of MATLAB in image processing, fast cutting, splicing, fitting and best point detection on child hip image can be achieved on computers. Through calculating arc formed by the two bones which fitting out to be the center of the ellipse, we can use the result to analyze inosculation of the two bones and to provide information for judging child bone development situation.