Corn Leaf Diseases Diagnostic Techniques Based on Image Recognition

Corn leaf diseases are automatically recognized by digital image processing techniques and pattern recognition method. The method is separated into three steps. First, the gray-scale images are gotten from color images which were caught by numeral camera, which is enhanced by histogram equalization method, and the unwanted noise is removed from the image. Secondly, the disease spots were segmented from leaves based on the iterative threshold method and morphological methods. Finally, the shape characteristic parameters of disease spots, such as area, perimeter, rectangularity, circularity and shape complexity, are extracted, which are used to identify and diagnose diseases. The results show that the corn leaf diseases of the 30 images could be well diagnosed with a diagnostic rate of 80%.