基于图像处理技术的杂草特征提取方法研究

利用计算机视觉技术将杂草从背景中识别出来进行定位喷洒农药已成为精细农业研究的热点。选取颜色空间OHTA中I＇2分量作为特征量；利用基于遗传算法的自动阈值选取方法对特征分量巧进行阈值分割初步分离杂草与小麦；通过颜色聚类和形态滤波获得准确的杂草区域。实验结果表明：直接在彩色空间进行分割，可提高彩色图像的分割效果，利用该方法获得的杂草平均正确识别率达到90．47％。

Research of weed feature extraction method based on image processing technology

Computer vision technology used to weed identification from background which to spray herbicide in a precise position has become a hot research of precision agriculture.A weed identification method that takes OHTA as color space and I2′as characteristic variant is presented,automatic threshold selection method on the basis of genetic algorithms for the feature component I2′is used to separate weed and wheat preliminarily,and then color clustering and morphological filtering is used to achieve the accurate weed regions.The experiment results show that separating directly in the color space would improve the segmentation effect of color images,and the average correct identification rate reaches 90.47 % using the method proposed before.