一种基于树莓派平台的脐橙品质分级方法

针对目前脐橙品质分级自动化程度较低,难以满足现代化生产运营需求的问题,设计了一种基于树莓派平台的脐橙品质分级方法。在S和V颜色分量图像上完成背景分割,并对脐橙表面的反光区域进行亮度矫正;提取脐橙的大小和着色度特征,并提出基于积分图的局部阈值分割算法,完成果面缺陷的检测;利用决策树算法实现脐橙品质的分级。试验结果表明:此方法对特级果、一等果、二等果、等外果的识别准确率分别达到96%、94%、94%、96%,单列输送线的分级速率达3个/s,准确性和实时性较高,能够满足实时环境下脐橙分级检测的要求。

Quality Grading Method of Navel Orange Based on Raspberry Pi Platform

To improve the low automation degree of navel orange quality classification,a method of navel orange quality classification based on Raspberry Pi platform is designed.The background is segmented on the S and V color components,and the brightness of the reflective areas on the navel orange surface is corrected.The size and color characteristics of navel orange are extracted and a local threshold segmentation method based on integral image is proposed to complete the detection of fruit surface defects.The decision tree algorithm is used to grade navel oranges.The experimental results show that the recognition accuracy of this method for extra-grade fruits,first-grade fruits,second-grade fruits and other external fruits reaches 96%,94%,94%and 96%respectively,with the grading rate of single conveyor lineis up to 3/s,and whose high accuracy and real-time can meet the requirements of grading detection of navel orange in real-time environment.