Plant health detection method and device based on image recognition

The invention relates to the technical field of plant health informatization detection, in particular to a plant health detection method and device based on image recognition. The method comprises thesteps of firstly collecting an infrared image of a to-be-detected plant in real time; preprocessing the infrared image through a gray scale enhancement method to obtain a gray scale enhanced image; extracting the texture features of the gray scale enhanced image; and finally, comparing the textural features with the expert database image features, and detecting the plant health condition. According to the present invention, by improving the accuracy of the plant image comparison, the accuracy of the plant health detection is improved. 本发明涉及植物健康信息化检测技术领域，具体涉及一种基于图像识别的植物健康检测方法及装置，首先实时采集待检测植物的红外图像；进而通过灰度增强方法预处理所述红外图像获得灰度增强图像；接着提取所述灰度增强图像的纹理特征；最后将所述纹理特征与专家库图像特征比对，检测出植物健康状况，本发明通过提高植物图像比对的准确率，从而提高植物健康检测的准确度。