节点文献
数显数字字符图像特征提取算法的研究与实现
Study and Realization for Numeral Instrument Image Feature Extraction Method
【摘要】 由于数显数字字符有其自身特点,相对于其它类型字符识别存在着新的难点,针对此类数显数字字符采用将细化后的字符图像看作是一幅连通图,选择闭合曲线作为其整体特征对十个数字字符进行初次分类,将笔画端点所处字符图像中四个子区域的位置作为主要的细节特征,对字符进行识别。测试结果表明该算法是可行、有效的。
【Abstract】 A thinned image is viewed as a graph.In the procession of image recognition of numeral instrument,the author of the paper selects closed curves as integrate features of those character images,makes the sub region where the end of last drawing of numbers locate of the character images as the major specific features,and then recognizes the character of numeral instrument.The experiment results shows that the proposed approach is practical and effective.
【关键词】 字符分割;
数学形态学;
特征提取;
字符识别;
【Key words】 character segmentation; mathematical morphology; feature extraction; character recognition;
【Key words】 character segmentation; mathematical morphology; feature extraction; character recognition;
【基金】 安徽省高等学校自然研究项目(2005KJ075)
- 【文献出处】 计算机技术与发展 ,Computer Technology and Development , 编辑部邮箱 ,2007年11期
- 【分类号】TP391.41
- 【被引频次】4
- 【下载频次】170