基于计算机视觉和神经网络的芒果外观等级分类研究

Research on Mango Appearance Rank Classification Based on Computer Vision and Neural Network

针对目前对芒果外观品质分级还是采取人工分级的不足,提出了一种基于计算机视觉和BP神经网络的芒果外观等级分类方法。首先,通过计算机视觉技术获取芒果图像,并利用基本的图像处理方法对芒果图像进行预处理。其次,根据芒果外观特征对芒果外观等级分类的影响,选择芒果的小波特征、缺陷面积所占百分比、颜色H分量值、芒果横径和果形指数等特征作为芒果外观等级分类的特征参数。最后,将提取的8个特征参数作为BP神经网络的输入,以芒果的3个等级分类为输出,建立芒果外观等级分类的神经网络模型,实现了芒果的外观等级分类。试验结果表明了该方法的有效性,识别率达93.3%。

For the shortage of artificial rank method in mango appearance rank classification,a method of mango appearance rank classification was presented based on computer vision technology and BP artificial neural networks. First,through the computer vision technology gaining mango image,the basic image processing method was used to carry on pre-processing to the mango image. Secondly,according to the effect of the appearance characteristic to mango appearance rank classification,the mango＇s wavelet characteristic parameter was chosed,the flaw area percentage,the color H component value,the transverse diameter and the shape index as its characteristic parameters of appearance rank classification. Finally,using these selected eight characteristic parameters as the input and the three rank classification results as the output,the BP neural network model of mango appearance rank classification was established and the mango appearance rank classification was realized. The test results indicated that the presented mango appearance rank classification method is effective and the recognition accurate rate has reached 93.3%.