Design And Implementation Of System Of Measuring The Leaves Of Arabidopsis

In the plant physiology about Arabidopsis, researchers often need to raise the Arabidopsis seeds and observe the growth processes. Due to the limited number of experimental seeds, Arabidopsis thaliana seeds which have been exposed to the space radiation is particularly precious in the research of biological effects. In the particular ground environment, the Arabidopsis seeds are cultivated and grow into seedlings. The data of the leaves’areas of Arabidopsis seedlings which are observed continuously without damage is important for analysis in the later experiments. So it is needed to develop a software to automatically separate the leaves from the entire Arabidopsis plants and separate the overlapping leaves and measure the single leaves without damage the plants.The development tool of software system is Visual Studio 2010. Because of using digital image processing technology, the function modules of software uses OpenCV image processing library on.NET platform package, namely the Emgu external library. This software uses the YUV color space transformation method to extract the Arabidopsis from complex background. With the morphological method in image processing and Sobel edge detection algorithm, extraction of single blade can be easy. When extract the leaves from the Arabidopsis plants, we will face the problem of overlapped leaves. To solve this problem, we use the geometrical "AND" of Otsu and Sobel algorithms. This method can help us to gain the complete outline of the single leaf contained in the overlapped leaves. Then we can compute the area data of the single leaf.The experimental results show that the software can effectively eliminate the influence of the complex background and extract the target plant Arabidopsis and measure the areas of leaves. The software is proved to be fast, efficient and practical.