作物病害图像形状特征提取研究

Research on Shape Feature Extraction of Crop Disease Image

形状特征是作物病害图像识别的重要依据之一。为了解决形状特征平移、尺度和旋转变换不变性问题,笔者将不变矩理论引入作物病害图像形状特征提取中,将7个Hu不变矩参数定义为形状特征,并引入新的参数,对作物病害图像进行形状特征提取研究。通过主成分分析和统计分析方法验证,得到适合两种作物病害识别的形状特征,并将其应用到作物病害智能识别系统中。实验结果表明:系统识别率较高,能够满足实际应用需求。

Shape feature is one of the important basis for image identification of crop disease. In order to solve the prob- lem of translation, scaling and rotation transformation invarianee of shape feature, the invariant moment theory was intro- duced to shape feature extraction of crop disease image. Seven Hu invariant moment parameters were defined as shape features and new parameters were introduced for shape feature extraction research on crop disease image. The shape features suitable for two types of crop disease recognition were received and applied in crop disease intelligent recognition system. The results show that the system recognition rate is relatively high, and can meet the practical application requirements.