Analysis and Testing of Weed Real-time Identification Based on Neural Network
Contrasting the two green strength genes of soil, wheat, corn, and the weed, the paper designed a system to identify the weed from the crop. It used the 2G-R-B and BP neural network, with the help of pixel-position-histogram diagram, to calculate the area and position of weeds. The result showed that it could identify the weed from the field and crop with an accuracy of 93%. The program gave the result that running time of identifying weed in wheat field was 273.31ms. As far as the corn was concerned, the time was 321.94ms, In a word, the system can satisfy the request of real-time.
Keywordsvariable spray weed identify picture disposal visual c++
- Blackmore B.S. Developing the principles of precision farming. Proceeding of Agrotech, Barretos. Brazil, 1999, 11(6):15-19Google Scholar
- Ji Shouwen, Wang Rongben, Chen Yajuan. Research on Recognizing Weed from Corn Seedling by Using Computer Image Processing Technology, Transactions of the Chinese Society of Agricultural Engineering, 2001(02) (in Chinese)Google Scholar
- Shi Yan, Study on the System of Spraying Rate Varied by Pressure of Liquid Chemical Application, Chinese Agriculture University, 2004 (in Chinese).Google Scholar
- Shi Yan, Qi Lijun, Fu Zetian. Model development and simulation of variable rate of pressure spray. Translations of the CSAE, 2004 (in Chinese)Google Scholar