Research On Nitrogen Detective System Based On Image Processing

This paper introduces the detection of research and application of image processing technology in the plant leaf. Discusses the key hardware and image data transmission in the selection of plant leaves image collection, using scanner to obtain floral blade image, and combined with the characteristics of digital image processing to deal with the digital information of leaf image, and then a series of processing the blade image analysis filter, edge detection, and the image segmentation, the main image processing algorithm is. Puts forward an improved edge detection algorithm--improved Sobel algorithm for different leaf image is adaptive, and the calculating accurate, fast speed, can meet the real-time. The experimental results show that the proposed algorithm can effectively used in leaf nitrogen detection system, and has achieved satisfactory effect. To analysis the color characteristic value and nitrogen content of the relationship between a regression model. Using Delphi 7.0 of the above algorithms for the programming, and verifies the accuracy and feasibility of the method for the next research, and laid a solid foundation.