Determining method for quality grade of Chinese mitten crab based on information fusion

Abstract

The invention belongs to the technical field of nondestructive testing, and relates to a determining method for quality grade of Chinese mitten crab based on information fusion. The method comprises the following steps: collecting a hyperspectral image of visible near infrared band of the crab shell of Chinese mitten crab and conducting a correction; then dividing the region of interest, extracting RGB information, also extracting the spectral information of visible near infrared band in the region of interest and screening characteristic wavelength; filtering processing the corrected hyperspectral image in step 1 and extracting texture information of the region of interest after obtaining a principal component image; and finally fusing the characteristic wavelength, the texture information and the RGB information and through an algorithm establishing a hierarchical model based on fusion information. Information fusion technology is utilized, the internal and external attribute information of Chinese mitten crab are combined, and the quality grades of Chinese mitten crab are embodied more three-dimensionally and comprehensively. The determining method has the advantages that the accuracy of classification is high, and the performance is more stable; and the defects of the prior art with single index, complicated process and low efficiency are solved.

Description

Technical field [0001] The invention belongs to the technical field of non-destructive testing, and specifically relates to a method for judging the quality of Chinese mitten crab based on information fusion. Background technique [0002] Eriocheir sinensis is an important and precious aquatic product in my country. It is not only delicious, but also rich in protein, vitamins, calcium, phosphorus, iron and other trace elements, and has high nutritional value. With the continuous improvement of people's quality of life, the demand for crabs has also increased year by year. As a leader in the aquaculture industry, the annual output has reached tens of thousands of tons. In order to meet the increasing demand of the market, the production of crabs is also increasing year by year. [0003] The characteristics of the back of the crab shell are the most significant signs of maturity of the Chinese mitten crab. The characteristics of the crab shell will directly affect the consumers' fir...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the problems and deficiencies of single index, cumbersome process and low efficiency in the grading of Chinese mitten crabs by the above-mentioned artificial and spectral image technology, the present invention aims to solve one of the problems, adopting a hyperspectral system to collect crab shell images and extract spectral information and screen out the characteristic wavelengths, extract the crab shell texture information and color (RGB) information, and then fuse the characteristic spectrum, texture information and color (RGB) information to establish a grading model based on the fusion information to achieve fast and non-destructive classification of Chinese mitten crabs. detection, and can meet the needs of online detection

Images