A Study on the Method of Image Pre-processing for Recognition of Crop Diseases

Abstract

Regarding the cucumber powdery mildew, speckle and downy mildew as examples, the method of image pre-processing for recognizing crop diseases was studied. This paper compared the effect of two filters - simple filter and median filter, and at last we chose median filter to wipe out the disturbance of noise effectively, and two-apex method was applied to separate the disease images from the background. Disease spots were separated through performing image edge detection and snake model, and the latter got more desired result. Thus the image pre-processing made a good foundation for following effective characteristic parameters for the disease diagnoses and setting up pattern recognition system.