Ground Based Multi Spectral Imaging Sensor To Characterize The Oilseed Rape Nitrogen Stress

Precision agriculture is one of the most important aspects for agriculture development in 21st century. It would promote the change from extensive agriculture to knowledge and technology-based one. According to the condition of China, the acquisition and processing of field information is the premise and difficulty for precision agriculture practice. The major object of the variable fertilization is the optimization of output with the least pollution of the environment. The problems of fertilization in China are the impertinency structure and the low utilization. In order to enhance the utilization and increase crop output, it is no time to delay the study of the variable fertilization. Variable fertilization technique can be used during the whole growth period of crops, includes the collections of the nutrition elements content demand of different crops growth periods and various soil information, implementation of variable fertilization and obtained the graph of yield information.The major study of this task is to test nitrogen content of the oilseed rape with the multi spectral imaging sensing technology and reflectance information of crops from green, red and near-infra spectrum wave band.The major study and achievements were showed as follow:1. Nitrogenous fertilizer is one of the main nutrition elements that will influent the crop growth. This research offered a new-style fast testing way of crop nutrition content in field based on the present condition. Compared with the traditional testing method, the new one has the feature of non-destructive testing, real time information collection and processing.2. Through feasible study, experiment project was designed. The equipment was also chose properly according pre-experiment.3. The relation between nitrogen content of the oilseed rape and SPAD value was studied. The result indicated that there exists a linear relationship between the nitrogen content and SPAD. The correlation (R) was 0.864.4. Usually, the reflectance value of one leaf would be much difference because of the influence caused by the change of light intensity in different period of time. In order to reduce this influence, a calibration system, which contained a calibrated board and can be used to calibrate at any moment in everywhere, was established. This system enhanced the feasibility of testing nitrogen content of oilseed rape using multi spectral imaging sensing technology. An experiment platform was established withequipment support and this platform can be used in real-time information collecting and data processing.5. The correlation between the reflectance of oilseed rape canopy and plant's leaf chlorophyll was studied. The results indicated that the correlation between SPAD (soil plant analysis's development) values of oilseed rape plant and each value of GNDVI (Green and NIR channel), NDVI (Red and NIR channel) and ratio (IR/Green) was linear relationship and the correlation (R) was 0.94, 0.94 and 0.92, respectively.6. During the shoot process with multi-spectral imaging sensor (3CCD), the background color of soil would affect the reflectance testing of the whole plant. The image processing technology in MATLAB was used to remove this influence by separating the background from the picture and data analysis also performed in MATLAB. The data process software was programmed by MATLAB.