The Research And Application On Cucumber Sclerophthora Macrospora Experimental Based On The Image Processing

This paper based on computer image processing, research cucumber downy mildew disease identification and diagnosis by computer image processing technology. As well as the development of computer processing power , the image processing and recognition technology is enhanced. In the field of agriculture, we use the image processing technology scientific and rational, we can identify the pest and disease correctly, identify the nutrition of plant and so on, prevent pests and diseases, reduce the lose of agricultural, increase peasants ' income, it has a positive and far-reaching significance.Image processing technology,the morphological feature extraction of sick leaves, pattern recognition, software system development and experimental verification are applied to process and recognize cucumber sclerophthora macrospora in this paper.1. Image processingTo improve image quality, we should deal with the selected image preprocessing. First of all, to enhance the image, introduced some methods to enhance the image, such as based on gray-level histograms, image smoothing, image sharpening; to segment the image, edge detection and marking, these are the preparatory work for the next feature parameter extraction.2. Extract the parameters of the sick leaves'morphological characteristicsWe extract the characteristic parameters separately from the color, texture, shape lesion of the sick leaves.Use HIS and CIE1931-RGB standard colorimetric system to extract the color feature. Use gray level co-occurrence matrix to extract the texture feature of images; extract the parameters on leaf lesion in disease part shape feature. Build recognition disease recognition model of leaves, established pattern recognition classifier; fuzzy identification with principles of membership identification method. 3. The development and experimental realization of the computer image processing and recognition software system.Prepared image processing program code in MATLAB, formed applicable image processing and recognition system.Selected the disease of cucumber leaves in seedling and adult stage two phases respectively, in accordance with the sampling time order, labeled seedling leaves sample 1, sample 2, and sample 3; with the adult time order, labeled adult leaves sample 1, sample 2, and sample 3. In the process of collecting, combined with artificial identification test, then dispose and identify the image compare the results of artificial recognition, to perfect the software system.Based on the research and development of image processing software, we realize the cucumber disease recognition. For computer image processing techniques in the identification of crop diseases , we provide the references.