【摘要】 图像分割是对图像特征提取与目标识别的前提,不同分割方法的精确度对特征提取及识别有很大影响。本研究对采集到的新鲜的感染稻瘟病的水稻叶片拍照,采用阈值分割法方对图像进行分割。实验结果表明,在直方图波谷处选取不同阈值进行分割,都能将水稻叶片图像中的病斑与叶片分割开,但阈值选取的差异会对分割结果产生很大影响。更多还原

【Abstract】 Image segmentation is the premise of image feature extraction and target recognition. Different segmentation methods have great influence on the accuracy of feature extraction and recognition. In this paper, through the threshold segmentation method, the collected fresh rice leaves infected with rice blast were photographed, and the threshold segmentation method was used to segment the image. The experimental results show that different thresholds at the trough of histogram can separate the diseased spots from the leaves, but the difference of threshold selection will have a great impact on the segmentation results.更多还原