基于图像的2维板件排样方案评价方法研究  

Research of Layout Estimating Method of 2-D Parts Packing Based on Image

作　　者：罗立宏[1,2] 冯开平[2] 叶家玮[1] 

机构地区：[1]华南理工大学交通学院,广东广州510641 [2]广东工业大学数字媒体系,广东广州510006

出　　处：《四川大学学报（工程科学版）》2013年第6期162-169,共8页Journal of Sichuan University (Engineering Science Edition)

基　　金：国家自然科学基金资助项目(50575046)

摘　　要：对于2维不规则件排样的重叠、出界检测和方案评价,传统方法都是使用图形学的方法计算,计算量大。提出一种基于图像和颜色统计的新方法。先基于面积和直方图的方法进行研究,再提出基于边的方法,然后又提出2种改进方法:动态尺度评价和局部重绘整体评价。对每一种方法都进行了时间性能测试和复杂度分析。结果表明:基于面积的方法对于小尺寸图像性能尚可,但大尺寸图像时间耗费大,时间为图像尺寸的平方级;基于边的方法时间性能有很大改善,时间耗费为图像尺寸的线性级别;动态尺度评价能基本克服离散方法中尺寸(或精度)与时间(或速度)的矛盾,时间耗费随尺寸的增长低于线性级;局部重绘整体评价则对于大规模排样的时间有大幅改善。最后以3个实例证明了这些方法的有效性。The traditional methods have used graphics to treat the problem of the overlap or outside testing and layout estimating of 2-D irregular nesting. In the present work, new methods based on image and color statistics were proposed, which included a method based on area and histogram, and a method based on edge. Time performance test and complexity analysis were carried for all these methods. The results showed that in small image the performance of the method based area is barely OK, but in large image the consuming time is long, being square level of the image dimension. The method based on edge has better performance, especially in large image, which time being linear level of image dimension.

关 键 词：排样 不规则件 评价 重叠检测 图像 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]