论文部分内容阅读
目前对蔬菜病害的识别方法都有一定的局限性,难以满足现代农业要求。该文以计算机视觉技术为手段,结合图像处理与模式识别技术,重点分析了茄子病害叶片上褐纹病病斑的颜色、形状、纹理特征参数,提出了一种基于叶片病斑特征的茄子褐纹病识别方法。根据在HSI(hue-saturation-intensity)颜色空间中叶片上病斑色调不同的特点,利用H分量图像提取病斑,获取病斑图片,然后提取每个病斑区域的12个颜色参数、11个形状参数和8个纹理参数等共31个特征参数。再通过方差和主成分分析法选择20个分类能力强的特征参数组成分类特征向量,并随机选取35个非褐纹病病斑的特征向量与35个褐纹病病斑的特征向量组成的训练集,构建Fisher判别函数对测试集进行分类,试验结果表明,对茄子褐纹病的识别准确率达到90%,说明该识别方法可以对茄子叶部病害进行快速、准确识别,为田间开放环境下实现茄子病害实时检测提供了技术支撑。
At present, the identification of vegetable diseases have some limitations, it is difficult to meet the requirements of modern agriculture. In this paper, computer vision technology as a means, combined with image processing and pattern recognition technology, focusing on analysis of the eggplant disease leaf spot brown spots on the color, shape, texture parameters, proposed a leaf spot based on the characteristics of eggplant brown Wound disease identification method. According to the different speckles in the HSI (hue-saturation-intensity) color space, the H-component images were used to extract the lesion and obtain the lesion images. Then the 12 color parameters of each lesion area were extracted, and 11 Shape parameters and eight texture parameters a total of 31 characteristic parameters. Then 20 variance-based and principal component analysis methods were used to select the 20 most powerful feature parameters to classify the feature vectors and randomly selected 35 feature vectors of non-brown spot disease lesions and 35 feature spots of brown spot disease Set up the Fisher discriminant function to classify the test set, the test results show that the identification accuracy of eggplant brown spot disease reached 90%, indicating that the identification method can quickly and accurately identify the eggplant disease leaves open for the field environment Real-time detection of eggplant disease provides technical support.