| Microscopic Halftone Image Segmentation |
| |
| 作者姓名： | 王永刚 杨杰 丁永生 |
| |
| 作者单位： | [1]InstituteofImageProcessing＆patternRecognition,ShanghaiJiaotongUniversity,Shanghai200030 [2]CollegeofInformationScienceandTechnology,DonghuaUniversity,Shanghai200051／／DigitizedTextile＆FashionTechnologyEngineeringResearchCenterofNationcdEducationMinistry,DonghuaUniversity,Shanghai200051 |
| |
| 摘 要： | Microscopic halftone image recognition and analysis can provide quantitative evidence for printing quality control and fault diagnosis of printing devices, while halftone image segmentation is one of the significant steps during the procedure. Automatic segmentation on microscopic dots by the aid of the Fuzzy C-Means (FCM) method that takes account of the fuzziness of halftone image and utilizes its color information adequately is realized. Then some examples show the technique effective and simple with better performance of noise immunity than some usual methods. In addition, the segmentation results obtained by the FCM in different color spaces are compared, which indicates that the method using the FCM in the f1f2f3 color space is superior to the rest.
|
| 关 键 词： | 图象处理 图象分割 FCM 色彩印刷 色彩空间 |
Microscopic Halftone Image Segmentation |
| |
| WANG Yong-gang,YANG Jie,DING Yong-sheng.Microscopic Halftone Image Segmentation[J].Journal of Donghua University,2004,21(5):83-87. |
| |
| Authors: | WANG Yong-gang YANG Jie DING Yong-sheng |
| |
| Institution: | 1. Institute of Image Processing & pattern Recognition, Shanghai Jiaotong University, Shanghai 200030
2. College of Information Science and Technology, Donghua University, Shanghai 200051;Digitized Textile & Fashion Technology Engineering Research Center of National Education Ministry, Donghua University, Shanghai 200051 |
| |
| Abstract: | Microscopic halftone image recognition and analysis can provide quantitative evidence for printing quality control and fault diagnosis of printing devices, while halftone image segmentation is one of the significant steps during the procedure. Automatic segmentation on microscopic dots by the aid of the Fuzzy C-Means (FCM) method that takes account of the fuzziness of halftone image and utilizes its color information adequately is realized. Then some examples show the technique effective and simple with better performance of noise immunity than some usual methods. In addition, the segmentation results obtained by the FCM in different color spaces are compared, which indicates that the method using the FCM in the f 1f 2f 3 color space is superior to the rest. |
| |
| Keywords: | image segmentation color halftone printing FCM color space |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
