基于计算机视觉的苹果自动分级方法研究

Research on Apple Grading Method Based on Computer Vision

研究苹果图像自动分级优化问题,传统方法采用单一特征进行分级,难以全面描述苹果状态,导致分级精度低。为提高苹果分级精度,提出计算机视觉的苹果自动分级方法。首先对计算机视觉采集的苹果图像多种特征进行提取,然后采用主成分分析对特征进行选择,最后建立基于最小二乘支持向量机的苹果自动分级模型。采用苹果数据对分组方法的性能进行仿真测试,实验结果表明,相对于其它分级方法,计算机视觉方法不仅提高了苹果自动分级效率,克服了传统方法的缺陷,而且加快了苹果自动分级速度,为水果品质分级等领域提供广泛的应用前景。

Based on computer vision, the paper proposed an apple automatic grading method. First, apple image features were extracted, and then the principal component was analysez to select the features. Finally, the automatic apple grading model was built based on least squares support vector machine. The experimental results show that, compared with other classification methods, the computer vision method not only improves the apple automatic grading efficiency, but also accelerates the speed of apple automatic grading, and has a wide application prospect in fruit quality grading and other fields.