Method for online monitoring gel strength change in surimi hot working process based on hyperspectral imaging technology

Abstract

The invention relates to a method for on-line monitoring of gel strength change in a surimi hot processing process based on a hyperspectral imaging technology, and the method comprises the following steps: collecting original hyperspectral data of a surimi sample in a heating process, and determining the gel strength of the surimi sample; carrying out preprocessing on the original hyperspectral data; constructing a data set based on the preprocessed original hyperspectral data and gel strength; training and verifying a PLS (Partial Least Squares) model by utilizing the data set to obtain a gel strength prediction model; and monitoring the gel strength change in the surimi hot working process by using the gel strength prediction model. According to the invention, online, rapid and nondestructive monitoring of the gel strength in the two-stage heating process of the minced fillet in industry can be realized.

Description

Technical Field

[0001] The invention relates to the technical field of food quality detection, and in particular to a method for online monitoring of gel strength changes during surimi thermal processing based on hyperspectral imaging technology. Background Art

[0002] In recent years, with the rapid development of society and the guarantee of abundant fishery resources, surimi and its products are deeply loved by consumers because of their good gel properties, high nutritional value, crisp taste, ready-to-eat and in line with the dietary concept of the current public. At present, surimi raw materials are mainly used to produce surimi products such as fish balls, fish cakes, fish rolls, fish sausages, fish tofu and aquatic simulated foods (such as artificial crab meat, shrimp meat, shark fins, scallops), which have high economic value. However, since surimi is easily affected by various factors during processing, which seriously affects the yield and quality of surimi, it is very important to develop a method for rapid non-destructive detection of gel strength during online monitoring of surimi thermal processing.

[0003] In recent years, other non-destructive technologies used to monitor quality changes during food processing include infrared imaging, near-infrared spectroscopy, machine vision, and electronic noses. These technologies collect relatively simple sample information, while hyperspectral imaging technology can obtain both spectral data and image information of samples, which can greatly enrich the data set and realize quality visualization analysis. Therefore, the study of online monitoring methods for gel strength changes during the thermal processing of surimi has important practical significance. Summary of the invention

[0004] The purpose of the present invention is to propose a method for online monitoring of gel strength changes during fish paste thermal processing based on hyperspectral imaging technology, so as to solve the problems existing in the above-mentioned prior art and realize online, rapid and non-destructive monitoring of gel strength during the two-stage heating process of industrial fish paste.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A method for online monitoring of gel strength changes during surimi thermal processing based on hyperspectral imaging technology, comprising:

[0007] collecting raw hyperspectral data of the surimi sample during heating, and determining the gel strength of the surimi sample;

[0008] Preprocessing the original hyperspectral data;

[0009] Based on the preprocessed raw hyperspectral data and gel intensity, a dataset is constructed;

[0010] Using the data set to train and verify the PLS model, and obtain a gel strength prediction model;

[0011] The gel strength prediction model is used to monitor the change of gel strength during the thermal processing of surimi.

[0012] Optionally, preprocessing the original hyperspectral data includes:

[0013] Performing black-and-white correction on the original hyperspectral data; wherein the original hyperspectral data includes: VNIR and NIR dual-band hyperspectral data;

[0014] Extract the region of interest from the corrected hyperspectral data;

[0015] Obtaining the average of the spectral reflectances of all pixels in the region of interest as average spectral data of the surimi sample;

[0016] The average spectrum data is subjected to standard normal variable transformation, multivariate scattering correction, first-order derivative, and second-order derivative processing.

[0017] Optionally, the method for performing black-white correction on the original hyperspectral data is:

[0018]

[0019] Among them, I represents the corrected reflectance hyperspectral image data, I0 represents the dark image, Rb represents the original hyperspectral image, and Rw represents the white reference image.

[0020] Optionally, the expression of the PLS model is:

[0021] Y=XA+B=XW*C+B=DC+B

[0022] Z*=Z(P'Z) -1

[0023] Where A is the PLS coefficient, B is the residual matrix of Y, D is the score matrix of X, Z is the PLS weight, P and C are the loadings of X and Y respectively, Z* is the regression coefficient matrix, X is the independent variable, Y is the dependent variable, and P' is the transpose of the loading matrix of the independent variable matrix X.

[0024] Optionally, before training the PLS model using the data set, the method includes:

[0025] The PLS model was simplified using VCPA, VCPA-IRIV, VCPA-GA, and IRIV algorithms.

[0026] Optionally, simplifying the PLS model using VCPA, VCPA-IRIV, VCPA-GA, IRIV algorithm includes:

[0027] Based on the VCPA algorithm, an exponential decreasing function is used to gradually reduce the number of variables; and binary matrix sampling is used to create subsets;

[0028] Based on the IRIV algorithm, the usefulness of variables is evaluated by observing the changes in RMSECV after adding or deleting variables;

[0029] Based on the VCPA-GA algorithm, the variable space is narrowed down and the selected variables are further optimized by simulating the process of natural selection, crossover and mutation to find the optimal variable combination;

[0030] Based on the VCPA-IRIV algorithm, VCPA is first used to reduce the variable space, and then IRIV is used to classify and iterate the variables to optimize the selection.

[0031] Optionally, collecting raw hyperspectral data of the surimi sample during the heating process includes:

[0032] placing a container containing surimi on a mobile platform and aligning it with a spectrum collection device;

[0033] When the platform moves, the complete spectral information of the surimi is obtained through the spectral acquisition device.

[0034] The beneficial effects of the present invention are:

[0035] The invention pre-treats the minced fish, and then heats it in a water bath in two stages. During the heat treatment process, hyperspectral data in the VNIR (400-1000nm) and NIR (900-1700nm) bands are collected, and the gel strength is measured. The obtained hyperspectral data is corrected in black and white, and the spectra of all pixels in the region of interest are collected by ENVI, and the average represents a sample. The spectral data is pre-processed, and the optimal pre-processing method is determined. A partial least squares regression model is constructed for the spectral data and the gel strength. The variable screening algorithm is used to obtain the optimal model effect in the VNIR and NIR bands during the heat treatment process, and the change of the gel strength of the minced fish during the heat treatment process is visualized and analyzed, so that the change of the gel strength during the heat treatment process of the minced fish can be monitored online in practical applications, and the visualization of the gel strength is realized, so that non-professionals can also clearly understand the specific situation during the heat treatment process. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0037] Figure 1A schematic flow chart of a method for online monitoring of gel strength changes during surimi thermal processing based on hyperspectral imaging technology according to an embodiment of the present invention;

[0038] Figure 2 It is a result diagram of characteristic variables extracted by four variable screening algorithms of the embodiment of the present invention;

[0039] Figure 3 Schematic diagram of optimal model parameters of an embodiment of the present invention; wherein (a) is a principal component score diagram of the optimal PLS model interactive verification, and (b) is a result regression diagram of the optimal model;

[0040] Figure 4 A schematic diagram for visual analysis of the optimal model of an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0043] like Figure 1 As shown, this embodiment proposes a method for online monitoring of gel strength changes during surimi thermal processing based on hyperspectral imaging technology, comprising:

[0044] collecting raw hyperspectral data of the surimi sample during heating, and determining the gel strength of the surimi sample;

[0045] Preprocessing the original hyperspectral data;

[0046] Based on the preprocessed raw hyperspectral data and gel intensity, a dataset is constructed;

[0047] Using the data set to train and verify the PLS model, and obtain a gel strength prediction model;

[0048] The gel strength prediction model is used to monitor the change of gel strength during the thermal processing of surimi.

[0049] Specifically, in this embodiment, a method for online monitoring of gel strength changes during surimi thermal processing based on hyperspectral imaging technology includes: obtaining 100 kg of live dry frozen silver carp surimi (AA grade) from Xiamen Anjing, thawing the surimi and putting it into a meat grinder to crush it, adding 2.5% surimi and water to adjust the moisture content to 80%, and putting it into an aluminum box; heating at 40°C for 30 minutes, heating at 90°C for 20 minutes, and taking 30 samples every 5 minutes, for a total of 300 samples; collecting hyperspectral data of the surimi samples and measuring the gel strength of the surimi samples; extracting spectral data of the sample hyperspectral region of interest, and constructing a PLS quantitative model with the spectral data and the surimi gel strength; using VCPA, VCPA-IRIV, VCPA-GA, and IRIV variable screening algorithms to simplify the model and improve the model robustness; this embodiment can realize online, rapid and non-destructive monitoring of gel strength during the two-stage heating process of surimi in industry.

[0050] Among them, simplifying the PLS model means reducing the variables in constructing the PLS model, which reduces the time for model calculation. These algorithms screen important variables based on the contribution rate and importance of the variables to the model.

[0051] Furthermore, the original hyperspectral data of the surimi sample collected during the heating process include:

[0052] placing a container containing surimi on a mobile platform and aligning it with a spectrum collection device;

[0053] When the platform moves, the complete spectral information of the surimi is obtained through the spectral acquisition device.

[0054] Specifically, in this embodiment, first, the surimi is pretreated and placed in an aluminum box for two-stage heating;

[0055] The frozen surimi used in this embodiment is AA grade, including: thawing the surimi, cutting it into small pieces, putting it into a meat grinder and stirring it for 3 minutes, adding 2% NaCl, adjusting the moisture content to 80%, continuing to stir for 3 minutes, putting it into a cylindrical aluminum box with a diameter of 2 cm and a height of 2 cm for heating, heating at 40°C for 30 minutes and heating at 90°C for 20 minutes.

[0056] Secondly, we collected hyperspectral data of the surimi during the heating process and measured the gel strength of each surimi, for a total of 300 data sets.

[0057] In this embodiment, VNIR and NIR dual-band hyperspectral data are collected every 5 minutes during the heating process. After the collection is completed, the gel strength of the surimi is tested immediately. The test speed is 1 mm / s, the compression distance is 12 mm, and the trigger force is 15N.

[0058] Gel strength = breaking force (g) × breaking distance (cm)

[0059] When performing the hyperspectral image acquisition operation, the hyperspectral imaging system used in this embodiment is a hyperspectral imager, a platform control system, a computer, etc., wherein the NIR-HSI (900-1700nm) system consists of an imaging spectrometer with a spectral resolution of 4nm (N17E, Specim, Finland), a camera with a resolution of 640×512, a camera lens (640miniRaptor, Northern Ireland), two tungsten lamps (LS-150, Wuling Optics, Taiwan, China), a mobile platform driven by a stepper motor (HSIM-800, Wuling Optics, Taiwan, China) and a computer with software.

[0060] The specific method of hyperspectral image acquisition is: place the aluminum box containing surimi on a mobile platform and align it with the camera. When the platform moves, the spectral information of a line of the spatial position of the surimi in the entire spectral region is obtained by the hyperspectral imager, and then the aluminum box is driven by the platform to move to obtain the spectral information of the surimi at other positions until the complete spectral information of the sample is obtained. The moving speed of the sample stage in the VNIR spectral imaging system is 8.3mm / s, the exposure time is set to 90ms, and the image resolution is 1604×605pixel; because the spectral scanning range is 678 effective wavelengths in the 400-1000nm band, the size of the three-dimensional data module finally obtained is 1604×605×678. The spectral resolution of the NIR spectral imager is 4nm, and the resolution of the camera and camera lens is 640×512, and the image resolution is 640×459pixel. Because the spectral scanning range is 678 effective wavelengths in the 400-1000nm band, the size of the three-dimensional data module finally obtained is 640×459×512. The entire acquisition process is carried out in a dark box, and whiteboard and blackboard information is collected for subsequent correction to prevent the light in the environment from affecting the collected hyperspectral images.

[0061] Furthermore, preprocessing the original hyperspectral data includes:

[0062] Performing black-and-white correction on the original hyperspectral data; wherein the original hyperspectral data includes: VNIR and NIR dual-band hyperspectral data;

[0063] Extract the region of interest from the corrected hyperspectral data;

[0064] Obtaining the average of the spectral reflectances of all pixels in the region of interest as average spectral data of the surimi sample;

[0065] The average spectrum data is subjected to standard normal variable transformation, multivariate scattering correction, first-order derivative, and second-order derivative processing.

[0066] Specifically, in this embodiment, the region of interest (ROI) of the corrected hyperspectral image is extracted, and the mean of the spectral reflectance of all pixels in the ROI is taken as the average spectrum of the surimi sample;

[0067] The hyperspectral data black and white correction in this embodiment uses the following formula:

[0068]

[0069] Among them, the corrected reflectance hyperspectral image I is expressed in units of relative reflectance (%); Rb represents the original hyperspectral image; I0 is a dark image (0% reflectance), and Rw is a white reference image (100% reflectance).

[0070] In a preferred embodiment, a circular area with a radius of 60 pixels is used as the region of interest. The range of the region of interest is determined based on the circular aluminum box containing the minced fish. In order to ensure that the region of interest can be a complete circular area when the minced fish container is directly used for online monitoring, the radius is set to 60 pixels in this embodiment to achieve online monitoring of the change in gel strength during the thermal processing of minced fish. A total of 600 minced fish spectral data are obtained under dual bands.

[0071] The obtained average spectral data were preprocessed, and the preprocessed average spectral data and gel strength were divided into training set and prediction set in a ratio of 7:3 for subsequent modeling;

[0072] In this embodiment, the average spectrum is subjected to preprocessing methods such as standard normal variable transformation, multivariate scattering correction, first-order derivative, second-order derivative, etc., to obtain hyperspectral data with smooth spectral information after noise removal.

[0073] Due to the light scattering characteristics of the surimi sample surface and the noise and interference caused by the instrument, the above four preprocessing methods were used to obtain the preprocessed spectral data. The KS grouping method was used to divide the data set into a training set and a prediction set in a ratio of 7:3.

[0074] Specifically, in this embodiment, before using the data set to train the PLS model, the process further includes:

[0075] The PLS model was constructed with gel strength and spectral data, and the VCPA, VCPA-IRIV (variable combination population analysis-iterative retention of information variables), VCPA-GA (variable combination population analysis-genetic algorithm), and IRIV (iterative retention of information variables) algorithms were used to simplify the PLS model, and the optimal model was used for visual analysis.

[0076] VCPA is inspired by the "survival of the fittest" principle observed in natural evolution, and uses an exponentially decreasing function (EDF) to gradually reduce the number of variables. Binary matrix sampling is then used to ensure that each variable has an equal chance of selection, thereby creating various subsets. Model population analysis is used to identify the subset with the lowest root mean square error of cross validation (RMSECV).

[0077] IRIV is a novel variable selection technique based on a binary matrix transformation filtering method that evaluates the usefulness of variables by observing the change in RMSECV after adding or removing variables. After multiple iterations, IRIV retains both strong and weak information variables while eliminating uninformative and interfering variables. The retained variables are then inverted to determine the best subset.

[0078] VCPA-GA After VCPA reduces the variable space and sets the final EDF residual variable to 100, GA further optimizes the selected variables by simulating the process of natural selection, crossover, and mutation to find the optimal variable combination.

[0079] The hybrid VCPA-IRIV strategy approach combines VCPA and IRIV to process high-dimensional spectral data. VCPA first reduces the variable space, and IRIV optimizes the selection by classifying and iterating the variables to remove less informative variables. This hybrid approach utilizes the systematic reduction of VCPA and the iterative filtering capabilities of IRIV to optimize the variable selection process.

[0080] The task of simplifying the PLS model is to reduce the dimension of the X matrix by screening out important variables (features), so that the PLS model is more concise, accurate and robust. When VCPA and its hybrid strategy algorithm perform stepwise regression, the model will optimize the prediction performance by gradually adding or removing independent variables. Cross-validation technology can help evaluate the impact of different feature subsets on the generalization ability of the model and select the most appropriate variable subset to simplify the PLS model. In the final analysis, the point of variable screening is to reduce the variables in the data set, so that the input data set becomes smaller and the model runs faster.

[0081] This implementation uses the spectral data and gel strength of surimi after pretreatment for PLS modeling. The optimal pretreatment method is determined, and the specific results are shown in Table 1. Then, based on the optimal pretreatment PLS model, the VCPA, VCPA-IRIV, VCPA-GA, and IRIV algorithms are used to simplify the PLS model. The results are shown in Table 2. The selected wavelengths are as follows Figure 2 As shown, the optimal model is Figure 3 As shown, Figure 3 (a) is the principal component score of the best PLS model interaction validation, Figure 3 (b) is the result regression of the best model, and the best model is used for visual analysis, such as Figure 4 shown.

[0082] PLS projects the independent variable X and the dependent variable Y into the latent variable space to build a linear regression model. This process takes into account the covariance information between X and Y to maximize the correlation between X and Y. PLS decomposes X and Y into several X scores (D) and constructs the PLS model using the following equation:

[0083] Y=XA+B=XW*C+B=DC+B

[0084] Z*=Z(P'Z) -1

[0085] Where A is the PLS coefficients, B is the Y residual matrix, D is the score matrix of X, Z is the PLS weights, P and C are the loadings of X and Y respectively, and Z* is the regression coefficient matrix. The set of data projected from the spectral data is called the orthogonal factors of the "latent variables". The optimal number of orthogonal factors depends on the prediction error and is usually achieved by using the lowest value of the prediction residual error sum of squares (PRESS).

[0086] Table 1 PLS model results in two bands based on different preprocessing methods

[0087]

[0088] Variable combination population analysis (VCPA), this strategy includes two key processes. First, the exponential decreasing function (EDF) is a simple and effective "survival of the fittest" principle in Darwin's natural evolution theory, which is used to determine the number of variables to be maintained and continuously shrink the variable space. Assume that the run of EDF is set to N, which means that it will go through N runs to iteratively filter variables. In other words, the variable space is reduced in units of N times. In the i-th run of EDF, the ratio calculation formula of the remaining variables is shown in (1). Secondly, in each EDF run, the binary matrix sampling (BMS) strategy is used to provide each variable with the same selection opportunity and generate different variable combinations to generate a subset population to construct a sub-model population. Then, the model population analysis (MPA) is used to find a variable subset with a lower cross-validation root mean square error (RMSECV). Calculate the frequency of each variable appearing in the best 10% sub-model. The higher the frequency, the more important the variable.

[0089] r i =e -θi (1)

[0090] Where θ is a constant parameter that controls the EDF curve. θ is related to the curvature of the EDF and is positively correlated with the speed of the falling curve. It can be calculated under the following conditions: (1) Initially, i is equal to 0 and all p variables are used for modeling, resulting in r 0=1; (2) at the rth N In the iterations, there are still ω variables left; then r N is equal to ω / p. ω is the number of variables remaining after running N times. Under the above conditions, θ can be determined as: Under the above conditions, θ can be determined as:

[0091]

[0092] Hybrid strategy based on VCPA,VCPA continuously shrinks the variable space based on EDF, and finally makes the large variable space smaller and optimized. To solve the existing limitation of GA and IRIV for a large number of variables, VCPA is modified and coupled with GA and IRIV to produce a hybrid strategy for variable selection. In addition, this hybrid strategy can help VCPA make up for its defect of tending to select too few variables. The hybrid strategy based on VCPA includes the following two steps: Step 1: Perform VCPA to shrink the variable space. For the modified VCPA, in this work, ω is set to 100 in the EDF step, which means that there are still 100 variables that need to be further optimized by GA and IRIV. Step 2: Perform GA and IRIV to further optimize the remaining 100 variables. These 100 variables are gradually retained by eliminating other variables that do not contribute much to the EDF for N iterations. The space spanned by the remaining variables is small and optimized, making it easier and better for GA and IRIV to select the best subset of variables.

[0093] Table 2 PLS model results under dual bands based on four variable screening algorithms

[0094]

[0095] The spectral values ​​corresponding to the pixels in the prepared surimi sample hyperspectral image are extracted, and the selected spectral values ​​are substituted into the optimal model to predict the gel strength information at each pixel in the surimi hyperspectral image. Finally, the information fusion of the hyperspectral imaging technology is used to reconstruct the distribution map of the information in the surimi thermal processing process on the plane according to the pixel coordinate information and its corresponding adulteration concentration, such as Figure 4 As shown. In the gel strength distribution diagram, the colors range from blue to red. The redder the color, the higher the adulteration concentration, and the bluer the color, the lower the adulteration concentration.

[0096] This embodiment proposes a method for online monitoring of gel strength changes during the thermal processing of surimi, constructs a PLS regression model with the spectral data of surimi during the thermal processing and the gel strength, and uses a feature screening method to simplify the model, and determines the optimal model for visual analysis. In actual processing applications, there is no need to sample and test the surimi, and no pollution is generated, so as to achieve the effect of online monitoring.

[0097] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for online monitoring of gel strength changes during surimi thermal processing based on hyperspectral imaging technology, characterized in that: include: collecting raw hyperspectral data of the surimi sample during heating, and determining the gel strength of the surimi sample; Preprocessing the original hyperspectral data; Based on the preprocessed raw hyperspectral data and gel intensity, a dataset is constructed; Using the data set to train and verify the PLS model, and obtain a gel strength prediction model; The gel strength prediction model is used to monitor the change of gel strength during the thermal processing of surimi.

2. The method for online monitoring of gel strength changes during surimi thermal processing based on hyperspectral imaging technology according to claim 1, characterized in that: Preprocessing the original hyperspectral data includes: Performing black-and-white correction on the original hyperspectral data; wherein the original hyperspectral data includes: VNIR and NIR dual-band hyperspectral data; Extract the region of interest from the corrected hyperspectral data; Obtaining the average of the spectral reflectances of all pixels in the region of interest as average spectral data of the surimi sample; The average spectrum data is subjected to standard normal variable transformation, multivariate scattering correction, first-order derivative, and second-order derivative processing.

3. The method for online monitoring of gel strength changes during surimi thermal processing based on hyperspectral imaging technology according to claim 2, characterized in that: The method for performing black and white correction on the original hyperspectral data is: Among them, I represents the corrected reflectance hyperspectral image data, I0 represents the dark image, Rb represents the original hyperspectral image, and Rw represents the white reference image.

4. The method for online monitoring of gel strength changes during surimi thermal processing based on hyperspectral imaging technology according to claim 1, characterized in that: The expression of the PLS model is: Y=XA+B=XW*C+B=DC+B Z*=Z(P'Z) -1 Where A is the PLS coefficient, B is the residual matrix of Y, D is the score matrix of X, Z is the PLS weight, P and C are the loadings of X and Y respectively, Z* is the regression coefficient matrix, X is the independent variable, Y is the dependent variable, and P' is the transpose of the loading matrix of the independent variable matrix X.

5. The method for online monitoring of gel strength changes during surimi thermal processing based on hyperspectral imaging technology according to claim 2, characterized in that: Before training the PLS model using the data set, the following steps are also included: The PLS model was simplified using VCPA, VCPA-IRIV, VCPA-GA, and IRIV algorithms.

6. The method for online monitoring of gel strength changes during surimi thermal processing based on hyperspectral imaging technology according to claim 5, characterized in that: Simplifying the PLS model using VCPA, VCPA-IRIV, VCPA-GA, IRIV algorithms includes: Based on the VCPA algorithm, an exponential decreasing function is used to gradually reduce the number of variables; and binary matrix sampling is used to create subsets; Based on the IRIV algorithm, the usefulness of variables is evaluated by observing the changes in RMSECV after adding or deleting variables; Based on the VCPA-GA algorithm, the variable space is narrowed down and the selected variables are further optimized by simulating the process of natural selection, crossover and mutation to find the optimal variable combination; Based on the VCPA-IRIV algorithm, VCPA is first used to reduce the variable space, and then IRIV is used to classify and iterate the variables to optimize the selection.

7. The method for online monitoring of gel strength changes during surimi thermal processing based on hyperspectral imaging technology according to claim 1, characterized in that: The raw hyperspectral data collected from the surimi sample during the heating process include: placing a container containing surimi on a mobile platform and aligning it with a spectrum collection device; When the platform moves, the complete spectral information of the surimi is obtained through the spectral acquisition device.

Images