基于Lab颜色空间和遗传算法的苦苣菜叶片病斑分割

Segmentation of common sow thistle leave disease spots based on Lab color space and genetic algorithm

随着可耕地的不断减少,植物生长柜的出现可以高效地给人们提供更加高质量的绿色蔬菜。但有时因营养液成分配比不均或一些不可预知的因素存在,使植物生长柜中的蔬菜出现了病斑。这种病害会对植物生长柜中的其他蔬菜造成影响,因此在对其进行识别和诊断时,首要的就是进行分割并提取病斑。本文采用一种在Lab空间下用遗传算法对苦苣菜叶片病斑进行提取的方法,把图像从RGB模型转换到Lab模型,把彩色图像转换成Lab图像;再采用遗传算法快速寻取最佳阈值,对叶片进行分割,完成对叶片正常区域和病斑区域的分离,最终还原出病斑的彩色图像。结果表明,所提出的方法能够精确地提取出叶片病斑区域。

With the continuous decrease of arable land, the occurrence of plant growth cabinet can efficiently provide people with more high quality green vegetables. But sometimes uneven ratio of nutrient solution components or the existence of some unpredictable factors makes vegetables appeared disease spot in the Plant Growth Cabine. In order to avoid this kind of disease affect growth of other vegetables in the Plant Growth Cabine, we need to be recognition and diagnosis, the priority is splitting and extracting the disease spot. This paper puts forward a method to extract the disease spot of common sow thistle leave by the genetic algorithm in lab space. Firstly, Switch the image from the RGB model to Lab model, and transform the color image to Lab images ; Secondly seek out optimal threshold based on the genetic algorithm lastly, the blade was divid- ed, complete the separation of leave normal area and the disease spot area ; Finally restore the color images of disease spot area. The test results show that the proposed method can accurately extract leave disease spot area.