Method and system for detecting crop diseases and insect pests
A technology for crops, diseases and insect pests, applied in neural learning methods, biological neural network models, image analysis, etc., can solve problems such as low accuracy rate and long training time, and achieve the effect of improving accuracy rate, low complexity, and short time
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
specific Embodiment approach 1
[0053] Specific implementation mode 1. Combination figure 1 This embodiment will be described. A method for detecting crop diseases and insect pests described in this embodiment, the method specifically includes the following steps:
[0054] Step 1, collecting images of crops to be detected, and performing sharpening processing on the collected images to obtain a sharpened image, and then intercepting images of crop leaf regions from the sharpened image;
[0055]Through sharpening, the boundaries of crop leaves can be highlighted, and then the image of the region of interest can be cut out according to the boundaries;
[0056] Step 2, after converting the size of the image intercepted in step 1 to a standard size, an image of a standard size is obtained;
[0057] Then align the obtained standard size image to the reference direction to obtain the processed image;
[0058] Set the size of the image, scale the intercepted image to the set size, and then select the reference d...
specific Embodiment approach 2
[0062] Specific implementation mode two: combination figure 2 This embodiment will be described. The difference between this embodiment and the first embodiment is that the specific process of the third step is:
[0063] Step 31. Taking the center of the processed image as the benchmark, divide the processed image along the circumferential direction, that is, take the center of the processed image as the center of the circle, and use the reference direction as the starting direction of the circle to divide the entire circle into 36 share;
[0064] Take the starting direction as the 0° direction of the circumference, take the image in the area greater than or equal to 0° and less than 10° as the first segment after segmentation, and use the image in the area greater than or equal to 10° and less than 20° as the first segment after segmentation Two copies, and so on, take the image in the area greater than or equal to 350° and less than 360° as the thirty-sixth copy after seg...
specific Embodiment approach 3
[0069] Specific embodiment 3: The difference between this embodiment and specific embodiment 1 or 2 is that in step 4, the image processed in step 2 is segmented to obtain sub-images after preliminary segmentation; the specific process is:
[0070] Step 41. Initialize the clustering center;
[0071] Step 42: Preliminarily segment the image according to the clustering centers to obtain sub-images after preliminary segmentation.
[0072] By segmenting the image, the local attention ability of the processed image can be improved during subsequent feature extraction, so as to avoid ignoring local features during feature extraction and improve the accuracy of pest detection.
[0073] Other steps and parameters are the same as those in Embodiment 1 or Embodiment 2.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More