基于图像处理的橡胶树割痕提取与死皮识别

橡胶树死皮病是影响橡胶树乳胶产量的一个重要原因。通过从图像中分离出橡胶树割痕的特征割线及乳胶来判定死皮等级可大大降低人工判定死皮级别的误判率。在考虑分割算法时,鉴于橡胶树图像分割的难点,改进传统的大津阈值算法,选用一种加权的Otsu法并与图像旋转相结合,可得到理想的分割阈值,分割效果极佳。同时,定义死皮的判定准则,对实验基地选取的260份图像进行实验,结果表明,该方法实现了橡胶树割痕的提取以及乳胶的分离,其死皮等级判断准确率高达92. 4%。该方法可有效预防死皮病的进一步蔓延,具有一定的实验价值。

EXTRACTION OF CUT MARKS AND IDENTIFICATION OF TPD OF RUBBER TREE BASED ON IMAGE PROCESSING

The tapping panel dryness(TPD)is an important factor affecting latex yield of rubber trees.The grading of TPD can be determined by separating the characteristic cut lines and latex from the image,which greatly reduces the misjudgment rate of artificial determination of the TPD grade.In consideration of the segmentation algorithm,this paper improved the traditional Otsu threshold algorithmin view of the difficulty of the image segmentation of the rubber tree.A weighted Otsu method was selected and combined with the image rotation.The ideal segmentation threshold was obtained,and the segmentation effect was excellent.We defined the criterion of TPD.Experiments on 260 images selected from the experimental base were carried out.The experimental results show that the method has realized the extraction of cut mark from the rubber tree and the separation of latex.The accuracy rate of judging TPD grade is up to 92.4%.It can effectively prevent the further spread of TPD and has certain experimental value.