论文部分内容阅读
讨论了一种新的物体快速识别方法。这个方法非常适合于柔性自动生产线上移动工件的快速实时识别。该方法的基础是新发展的定向执行长度编码法(orientation run lengthcoding—ORLC),由此产生了一种新的物体影像样本形式,即:ORLC坐标样本。它的特点是样本数据量小,占用存储空间少,构造过程简单,柔性强,可根据不同的应用选择不同级次的样本形式,匹配处理速度快,匹配和识别信噪比高,不易产生错判。给出了ORLC方法的理论表达式和匹配判别式,并给出了ORLC坐标样本和实验数据。
A new method of object recognition is discussed. This method is very suitable for rapid real-time identification of moving parts in flexible automated production lines. The method is based on the newly developed orientation run length coding (ORLC) method, resulting in a new form of object image samples, namely: ORLC coordinate samples. It is characterized by a small amount of sample data, taking up less storage space, simple construction process, flexible, according to different applications choose different levels of sample forms, matching processing speed, matching and identification of high signal to noise ratio, not easy to produce wrong Judgment The ORLC method’s theoretical expression and matching discriminant are given, and the ORLC coordinate sample and experimental data are given.