CN-121978028-A - Method, equipment and medium for evaluating muscle quality of living cultured jewfish based on hyperspectral image

Abstract

The invention relates to a method, equipment and medium for evaluating the muscle quality of living cultured jewfish based on hyperspectral images, belonging to the technical field of mariculture, wherein the method comprises the steps of obtaining a body surface hyperspectral image of a target living jewfish; and inputting the body surface hyperspectral image data of the target living jewfish into a pre-constructed cultured jewfish muscle quality evaluation model to obtain the target fish muscle texture characteristic parameters and the digitized distribution diagram of the target fish muscle texture characteristic parameters. According to the invention, the rapid and nondestructive measurement of the living cultured perch sample is realized, particularly, the portable land-based hyperspectral imaging system is utilized, the working time and cost are reduced, the applicable conditions are widened, the construction of an estimation model is more convenient, and the efficient measurement of the muscle texture characteristic parameters of the target living cultured perch can be realized.

Inventors

• CHE SHUAI
• LIU SHUFANG
• LI ANG
• WANG HUAN
• AN CHANGTING

Assignees

• 中国水产科学研究院黄海水产研究所

Dates

Publication Date

20260505

Application Date

20260121

Claims (10)

1. 1. A method for evaluating the muscle quality of a living cultured jewfish based on hyperspectral images, which is characterized by comprising the following steps: Acquiring a body surface hyperspectral image of a target living jewfish; And inputting the body surface hyperspectral image data of the target living jewfish into a pre-constructed cultured jewfish muscle quality evaluation model to obtain the target fish muscle texture characteristic parameters and the digitized distribution diagram of the target fish muscle texture characteristic parameters.
2. 2. The method for evaluating the muscle quality of the living cultured jewfish based on the hyperspectral image according to claim 1, wherein the determining method of the muscle quality evaluation model of the cultured jewfish is as follows: collecting hyperspectral image data of the body surfaces of living jewfish in different jewfish culture production areas, and synchronously measuring the muscle texture characteristic parameters of the corresponding positions of the cultured jewfish; Extracting characteristic wave bands of the target living body perch body surface hyperspectral image data based on living body perch body surface hyperspectral characteristic spectrum information, wherein the characteristic wave bands comprise spectrum information in a spectrum range of 400-1000 nm; And carrying out spectrum pretreatment on the characteristic wave band to eliminate the influence of environmental noise, and constructing an evaluation model for the muscle quality of the cultured jewfish by utilizing a machine learning method and combining the pretreated hyperspectral image and actually measured muscle texture characteristic parameter data.
3. 3. The method for evaluating the muscle quality of the living cultured jewfish based on the hyperspectral image according to claim 2, wherein hyperspectral image data of the body side surface of the living cultured jewfish is collected, and the collection distance is 30-60 cm.
4. 4. The method for evaluating the muscle quality of the living cultured jewfish based on the hyperspectral image according to claim 3, wherein the method for synchronously measuring the muscle texture characteristic parameters of the corresponding part of the cultured jewfish by adopting a texture analyzer is specifically comprising the following steps: Acquiring and collecting hyperspectral data of the cultured jewfish; Removing surface skin and fascia from the obtained muscle; cutting the muscle with the surface skin and fascia removed into a plurality of muscle cube blocks; placing the plurality of muscle cube blocks on a texture measuring platform, and measuring the characteristic parameters of the muscle texture of the corresponding part of the cultured jewfish by adopting a full texture analysis mode, wherein the characteristic parameters comprise hardness, elasticity, adhesiveness, cohesiveness, gumminess, chewing property and resilience; And respectively taking average values of the texture characteristic parameters acquired by the plurality of muscle cube small blocks to represent the muscle texture characteristic parameters of the corresponding target cultured jewfish bodies.
5. 5. The method for evaluating the muscle quality of the living body cultured jewfish based on the hyperspectral image according to claim 4, wherein the extracting the characteristic wave band of the target living body jewfish body surface hyperspectral image data based on the living body cultured jewfish body surface hyperspectral characteristic spectrum information specifically comprises: Acquiring image data of a reflection reference plate; Calibrating the living body cultured perch body surface hyperspectral image data by adopting the reflection reference plate image data, and preprocessing the calibrated cultured perch body surface hyperspectral image data to obtain preprocessed hyperspectral information; According to the pretreated hyperspectral information, hyperspectral characteristic spectrum information of the cultured perch body surface is obtained.
6. 6. The method for evaluating the muscle quality of the living cultivated jewfish based on the hyperspectral image according to claim 5, wherein the preprocessing method comprises one or more combination of Savitzky-Golay smoothing process, standard normal transformation process, trending process, multiple scattering correction process, first reciprocal process, second derivative process.
7. 7. The method for evaluating the muscle quality of the living cultivated jewfish based on hyperspectral images according to claim 6, wherein the machine learning method comprises one or more of partial least squares regression, support vector machine regression, random forest regression, convolutional neural network and counter-propagating neural network.
8. 8. The method for evaluating the muscle quality of the living cultured jewfish based on the hyperspectral image according to claim 2, wherein the step of obtaining the body surface hyperspectral image of the target living cultured jewfish specifically comprises the step of obtaining the body surface hyperspectral image data of the target jewfish by using a portable land-based hyperspectral imaging system.
9. 9. A computer device comprising at least one processor and a memory communicatively coupled to the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform a hyperspectral image based method of evaluating muscle quality of a living cultivated layer of micropterus salmoides as claimed in any one of claims 1-8.
10. 10. A computer-readable storage medium, characterized in that the medium stores a computer program which, when executed by a processor, implements a method for evaluating muscle quality of a living cultivated-living lateolabrax japonicus based on hyperspectral images as claimed in any one of claims 1 to 8.

Description

Method, equipment and medium for evaluating muscle quality of living cultured jewfish based on hyperspectral image Technical Field The invention belongs to the technical field of mariculture, and particularly relates to a method, equipment and medium for evaluating the muscle quality of living cultured jewfish based on hyperspectral images. Background The jewfish is an important sea water cultured fish in China, the annual total yield of the jewfish accounts for about 12 percent of the national sea water fish cultured yield, and the jewfish has important economic value. With the improvement of the culture yield of the jewfish, consumers increasingly pay attention to the muscle quality of the jewfish. Among these, the texture characteristics of muscle encompass the structural, mechanical and surface characteristics perceived by muscles and muscle products during human chewing, and are considered to be sensory factors of muscle quality that directly affect consumer satisfaction. The muscle texture characteristics are characterized by hardness, elasticity, adhesiveness, cohesiveness, gumminess, chewiness, and recovery. However, the current method for evaluating the texture characteristics of muscles mainly relies on a texture analyzer to quantitatively describe, and the method needs complex pretreatment of samples, consumes a great deal of labor and time, is destructive to the samples, and affects production while failing to meet the real-time detection of the samples. Furthermore, the heterogeneity of fish muscle also makes conventional assay methods subject to representative errors. The evaluation of the muscle texture characteristics of the jewfish still lacks a nondestructive high-flux digital method, and the healthy and sustainable development of the jewfish culture industry is severely restricted. Therefore, the nondestructive high-throughput assessment of the muscle quality of the cultured jewfish by utilizing the hyperspectral imaging technology model is a key technology which needs to be solved urgently. The hyperspectral imaging technology can simultaneously capture high-resolution space and spectrum data, has the advantages of being rapid, accurate, lossless and capable of continuous monitoring, and has been widely applied to the fields of meat product authenticity, color, flavor, freshness, nutrition parameters, microbial pollution evaluation and the like. However, the related applications are concentrated on the muscle after processing as a sample, and the application of the muscle quality evaluation in living marine cultured fish has not been reported yet. Especially, the jewfish is cultivated in coastal provinces of China in a mode of pond cultivation, off-shore cage cultivation and industrial circulating water cultivation. Different culture regions and culture modes create different muscle texture characteristics of the jewfish. At present, a living Lateolabrax muscle texture characteristic inversion model suitable for a main culture area of the Lateolabrax is not available. Disclosure of Invention The invention aims to provide a method for evaluating the muscle quality of the living cultured jewfish based on hyperspectral images, so that a rapid, accurate, nondestructive and high-throughput method for measuring the muscle texture characteristic parameters is constructed aiming at the living cultured jewfish, is a basis for evaluating the muscle quality of the cultured jewfish and screening germplasm resources, and has challenges and innovativeness. In order to achieve the above object, the present invention provides a method for evaluating muscle quality of a living cultured jewfish based on hyperspectral images, comprising: Acquiring a body surface hyperspectral image of a target living jewfish; And inputting the body surface hyperspectral image data of the target living jewfish into a pre-constructed cultured jewfish muscle quality evaluation model to obtain the target fish muscle texture characteristic parameters and the digitized distribution diagram of the target fish muscle texture characteristic parameters. Further, the determination method of the cultured jewfish muscle quality evaluation model comprises the following steps: collecting hyperspectral image data of the body surfaces of living jewfish in different jewfish culture production areas, and synchronously measuring the muscle texture characteristic parameters of the corresponding positions of the cultured jewfish; Extracting characteristic wave bands of the target living body perch body surface hyperspectral image data based on living body perch body surface hyperspectral characteristic spectrum information, wherein the characteristic wave bands comprise spectrum information in a spectrum range of 400-1000 nm; And carrying out spectrum pretreatment on the characteristic wave band to eliminate the influence of environmental noise, and constructing an evaluation model for the muscle quality of the cultured jewfish by utilizing a m