基于机器视觉的扁弹簧分类方法研究

摘要

为实现扁弹簧的自动化分类,提出一种基于机器视觉的扁弹簧分类方法.首先对采集到的不同型号的扁弹簧图像进行预处理,对预处理图像提取黑色像素分布向量,作为特征参数向量.然后采用DDAG SVMs方法将不同型号的扁弹簧进行分类,最终对采集到的不同型号扁弹簧测试样本进行分类测试实验.结果表明,该分类方法效率高、准确率高,可以满足扁弹簧自动化分类的各项要求.%In order to realize the automatic classification of flat springs a new classification method based on machine vision is proposed.And the distribution vectors of black pixel is extracted as the feature parameter vector.Then,the SVMs DDAG method is used to classify the different types of fiat springs,and the test samples of flat spring are classified into different types.The result shows that this classification method is so efficient and accurate that can meet the requirements of the fiat spring automatic classification.