Fruit defect classification method based on compressed sensing
A classification method and compression sensing technology, applied in the direction of optical testing flaws/defects, etc., can solve the problems of limited promotion and application, complicated processing process, long execution time, etc., to promote economic development, fast classification speed, and high accuracy. Effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0022] The present invention is a method for grading apple defects based on compressed sensing, using first-class fruit (apples without surface defects) as the measured object, including the following steps:
[0023] Step 1, obtain the main view and two left and right side views of the tested apple through the CCD camera;
[0024] Step 2, respectively extracting the R component images corresponding to the RGB images of the left and right side views, and adopting a color image space mean filter to smooth and filter the R component images of the left and right views respectively, so as to reduce the noise of the image;
[0025] Step 3: Transform the left and right side views from the RGB model space to the HIS model space, extract the corresponding H component images, and use the color image space mean filter to smooth and filter the H component images of the left and right views respectively to reduce the image distortion. noise;
[0026] Step 4: Take the 3*3 area in the upper...
Embodiment 2
[0034] Second-class fruit (with surface defects and the total area is not greater than 1cm 2 Apple) as the measured object, including the following steps:
[0035] Step 1, obtain the main view and two left and right side views of the tested apple through the CCD camera;
[0036] Step 2, respectively extracting the R component images corresponding to the RGB images of the left and right side views, and adopting a color image space mean filter to smooth and filter the R component images of the left and right views respectively, so as to reduce the noise of the image;
[0037] Step 3: Transform the left and right side views from the RGB model space to the HIS model space, extract the corresponding H component images, and use the color image space mean filter to smooth and filter the H component images of the left and right views respectively to reduce the image distortion. noise;
[0038] Step 4: Take the 3*3 area in the upper left corner of the R component image after the filt...
Embodiment 3
[0046] With the third-class fruit (the total area of surface defects is greater than 1cm 2 Apple) as the measured object, including the following steps:
[0047] Step 1, obtain the main view and two left and right side views of the tested apple through the CCD camera;
[0048] Step 2, respectively extracting the R component images corresponding to the RGB images of the left and right side views, and adopting a color image space mean filter to smooth and filter the R component images of the left and right views respectively, so as to reduce the noise of the image;
[0049] Step 3: Transform the left and right side views from the RGB model space to the HIS model space, extract the corresponding H component images, and use the color image space mean filter to smooth and filter the H component images of the left and right views respectively to reduce the image distortion. noise;
[0050] Step 4: Take the 3*3 area in the upper left corner of the R component image after the filt...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 