基于机器视觉图像提取的马铃薯内部病虫害特征识别

构建二维马铃薯内部病虫害视觉图像采集模型,对采集的马铃薯内部病虫害视觉图像进行分块融合检测,根据马铃薯绿叶素纹理分布进行病虫害的特征检测,提取马铃薯内部病虫害视觉分形特征量,采用表面纹理配准和分块自适应检测方法进行病虫害的特征点标定,结合小波变换方法进行马铃薯内部病虫害视觉图像的特征分解,根据颜色梯度变化的差异性实现机器视觉下的马铃薯内部病虫害特征识别。仿真结果表明采用该方法进行马铃薯内部病虫害特征识别的准确率接近90%,提高了马铃薯内部病虫害的防治和识别能力。

Recognition of plant diseases and insect pests in potato based on machine vision image extraction

In order to improve the detection and recognition ability of potato internal diseases and insect pests, image visual feature recognition method was used to detect potato diseases and insect pests, and a method of potato internal diseases and insect pests feature recognition based on machine vision image was proposed. A two-dimensional visual image acquisition model of potato internal diseases and insect pests was constructed, and the visual images of potato internal diseases and insect pests were detected by block fusion, and the characteristics of diseases and insect pests were detected according to the distribution of potato green leafin texture. The visual fractal features of potato internal diseases and insect pests were extracted, the surface texture registration and block adaptive detection methods were used to calibrate the feature points of diseases and insect pests, and the wavelet transform method was used to decompose the visual images of potato internal diseases and insect pests. According to the difference of color gradient change, the characteristics of potato diseases and insect pests under machine vision were recognized. The simulation results show that the accuracy of the method is close to 90%, which improves the ability of the prevention and identification of the internal diseases and insect pests of potato.