基于统计模式识别的植物病害彩色图像分割方法

根据植物病害彩色图像的特点,提出了用颜色空间作为特征空间,利用统计模式识别的监督分类方法,采用基于Fisher准则的线性判别函数来对彩色图像进行真彩色二值化分割。结果表明:该方法具有快速、有效和准确的特点,对彩色植物病害图像的分割处理十分有效。

Color image segmentation method based on statistical pattern recognition for plant disease diagnose

A supervision and sorting technique on the basis of statistical pattern recognition by using color as characteristic space was proposed according to the properties of plant disease image.Real color binarylization segmentation to the color image was carried out through applying Fisher criterion.The experimental result shows that this segmentation method possesses high speed with high accuracy and is effective in plant disease image segmentation and processing.