Research On The Detection System Of The Greenhouse Vegetable Diseases Degree Based On Android Platform

Aimed at the lack of the traditional disease detection methods in greenhouse vegetable, this paper combines the Android development technology with computer image processing technology to study the greenhouse vegetable diseases and develop a non-destructive degree detection system of greenhouse vegetable diseases based on Android platform. The proposed detection system makes great contribution to the leaf disease degree detection and the disease prevention of the greenhouse vegetable.The proposed disease degree detection system was built by the Eclipse and other software based on the Android development environment. After lots of research and analysis of the image processing algorithms, the CIE1931-RGB color space was chosen as the final color space model, the automatic threshold segmentation algorithm was chosen as the main segmentation method of the vegetables leaf, and the fuzzy C-means clustering algorithm was chosen as a method of extracting the vegetables leaf lesion. Finally, the language Java was used to finish the coding tasks. Firstly, the collected disease leaf images were loaded into the disease detection system, and some image preprocesses were work on it. Secondly, the leaf area and lesion area were segmented and extracted, and the pixels number of this area was counted. Thirdly, the pixels number of the leaf area and lesion area were compared, and the disease level was judged by the standard grade criteria. Then we used several leaves to define the disease index of all greenhouse vegetables. Finally, the results of the disease detection and the prevention or control measures were displayed on the screen. Users could upload the data to the server through the mobile network, and they were also able to view the greenhouse vegetables disease on the website through personal computer.The results show that the user could use the detection system combined the Android mobile application development technology with the image processing technology to require the disease detection and the prevention or control measures through some simple operations. The proposed detection system realizes the greenhouse vegetables non-destructive disease detection in real time with enough reliability.