Grain particle surface regularity detection method
A technology for grain particles and detection methods, applied in the direction of testing moving fluids/granular solids, measuring devices, optical testing flaws/defects, etc. performance and accuracy, improve detection efficiency, and achieve accurate detection results
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0094] A method for detecting the surface regularity of grain grains, comprising the following steps:
[0095] a) collecting images of the grains to be measured;
[0096] attached figure 1 , 2 An example of the captured grain image.
[0097] b) preprocessing the grain image to be measured to obtain a preprocessed grain image;
[0098] The above preprocessing specifically includes one or more of image segmentation, image registration, gray scale processing, binarization processing, and image enhancement processing.
[0099] c) extracting the feature information of the preprocessed grain image, the feature information including the central coordinates, minimum radius, and maximum radius of the grain;
[0100] Before the detection starts, first extract the characteristic information of the grain standard samples from different regions, different varieties, and different years according to the method from step a to step c, and establish a grain standard sample feature informat...
Embodiment 2
[0170] This experimental example is to study the accuracy of the test results.
experiment example 1
[0172] The grain surface regularity detection method provided by the present invention is used to detect wheat samples, including 1000 grains of regular grains, diseased grains, insect-eaten grains, raw bud grains, moldy grains, and damaged grains. The test results are shown in Table 1.
[0173] Table 1 Identification results of wheat samples
[0174]
[0175] It can be seen from Table 1 that the number of regular kernels, diseased kernels, insect-eaten kernels, sprouted kernels, moldy kernels, and damaged kernels was correctly identified to a total of 5890 kernels, and the correct recognition rate was 98.17%.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


