【作者】 赵海英； 杨一帆； 徐光美；

【Author】 ZHAO Hai Ying~(1,2)),YANFG Yi Fan~(3)),XU Guang Mei~(4)) 1)(College of Computer Science and Technology,Xinjiang Normal University;Urumqi Xinjiang,830054) 2)3)(School of Information Engineering University of Science and Technology Beijing,Beijing,100083) 4)(Information college of Beijing Union Universit,Beijing 100101)

【机构】 新疆师范大学计算机科学技术学院； 北京科技大学信息工程学院； 北京联合大学信息学院；

【摘要】 新疆民间织物图像具有鲜明的民族特色,为了生成和创新民族民间织物图像,如何自动分割织物基元是一个关键问题。论文以问题为驱动,提出了一种基于边缘形态变换的彩色织物图像分割算法。首先,对彩色图像进行颜色空间的变换,然后根据图像特征提出调整图像大小的前处理标准,再利用Canny算子检测边缘,对检测边缘进行形态变换后处理,最后标记变换图像的封闭区域进行织物图像的分割。实验数据表明:该算法虽然是几种常用算法的组合,但较好解决了民间织物彩色图像分割问题,而且对不同类型的织物图像,均可以获得令人满意的分割结果。更多还原

【Abstract】 Xinjiang folk fabric image has distinct national characteristics.It is a key problem that how to segment the folk fabric images,in order to generate and innovation.Taking problem driven,this paper puts forward a segment algorithm based on edge morphological transform for color fabric image.First,color space transformation is implemented,then the image size norms are proposed according to the characteristics of image,and Canny operator and morphological transform are executed to detect edge of image as post-processing.Finally,based on closed mark transform area fabric image segmentation is carried out.Experimental data shows that the algorithm by combination of several common algorithms,but better solved the folk fabric color image segmentation problem,which can obtain satisfactory segmentation result for different types of fabric image.更多还原