Abstract:In order to meet the demands of collecting and analyzing services of farmland crop information, a crop information collection and analysis tool was designed based on cloud service architecture with the combination of the smart phone terminal hardware, WeChat applet software and cloud service platform. The system mainly included Tencentcloud server module and mobile phone WeChat applet module. The MySQL was used to build a database for data storing, processing and download on cloud server. The WeChat applet was developed by CSS, JavaScript and applet packaged components. It was used to realize the interaction between collecting and uploading data and information feedback. In order to apply and test the system, the survey of biomass indication parameters of wheat in the field was taken as an example. The researches were carried out for the calculation of canopy coverage and plant rowspacing. More than 100 sampling images of wheat were captured at seedling period, and they were uploaded from applet to background and processing. After the image preprocessing following the target area detection by using Hough transform, image mask segmentation and image enhancement with erosion processing, the canopy of wheat was segmented and the canopy coverage was calculated by using HSV color space to highlight the pixel of sampling plants. The algorithms were proposed to extract the peak line of plant by projection and filtering method. Then, it was used to calculate the plant line spacing in the row. The linear regression model was established to indicate the fitting accuracy between the line spacing of the image recognition pixels and the measured values. The result showed that the modeling accuracy R2 reached 0.911. It could provide a technical support for crop information detection and investigation in the field.
