节点文献
基于数学形态学和纹理的切屑图像处理及识别
Image Processing and Recognizing of Chip Shape Based on Mathematical Morphological
【摘要】 根据切屑图像的特点,阐述了数学形态学在切屑图像降噪及边缘提取中的应用。利用所述方法对几种不同类型的切屑图像进行处理,并进行纹理特征提取,利用SOM神经网络对图像进行分类,最终将其应用于切屑图像的识别。经检验证明,该方法利用形态学对切屑图像处理及纹理分析是可行的,并且有较好的效果。
【Abstract】 For the defects of supervisions of the chips’ shape,the images are processed for eliminating the noise and detecting the edges of the image based on the theory of mathematical morphological.Then the images can be recognized by the characteristics.The theoretical algorithm for classifying the images based on SOM is used for recognizing and is proved to have a good feasibility.
【关键词】 图像识别;
数学形态学;
形态学滤波;
纹理;
切屑形态;
【Key words】 image recognition; morphological; morphological filter; texture; chip shape;
【Key words】 image recognition; morphological; morphological filter; texture; chip shape;
【基金】 江苏省高校自然科学基金资助项目(编号03KJ0460193)
- 【文献出处】 苏州大学学报(工科版) ,Journal of Soochow University(Engineering Science Edition) , 编辑部邮箱 ,2006年04期
- 【分类号】TP391.41
- 【被引频次】17
- 【下载频次】177