节点文献
Relief算法在关门车故障自动识别中的应用
Application of Relief algorithm to valve closing trouble automatic recognition
【摘要】 针对货车运行故障动态图像检测(TFDS)系统中的图像识别问题,提出一种基于Relief算法的解决方案。该方案采用梯度—共生矩阵的数字特征来描述一幅图像,然后通过Relief算法进行有效的特征选择,根据这些特征,通过模板匹配技术,最终检测出关门车故障。现场运用表明,该方案具有很高的实用性。
【Abstract】 A new method based on relief algorithm was proposed to solve image recognition problem in Trouble Frightcar of Moving Detection System(TFDS).A single image was described by numerical characteristics of co-occurrence matrix.Relief algorithm was used to evaluate the importance of every characteristics.Template matching was used in image recognition in TFDS.Experimental results showed that this method was efficient and applied.
【关键词】 共生矩阵;
Relief算法;
模板匹配;
图像识别;
【Key words】 co-occurrence matrix; Relief algorithm; template matching; image recognition;
【Key words】 co-occurrence matrix; Relief algorithm; template matching; image recognition;
- 【文献出处】 铁路计算机应用 ,Railway Computer Application , 编辑部邮箱 ,2007年01期
- 【分类号】TP274
- 【被引频次】18
- 【下载频次】217