Establishment method of brittleness grading standard of brittle tilapia and brittleness grading method of brittle tilapia

Abstract

The invention belongs to the technical field of tilapia quality detection, and particularly relates to an establishment method of a brittleness grading standard of crisp tilapia and a brittleness grading method of the crisp tilapia. The method comprises the following steps: determining the hardness, elasticity and resilience of fresh crisp tilapia mossambica by adopting a texture analyzer, and setting sensory evaluation crispness classification for cooked crisp tilapia mossambica; performing linear regression analysis on texture analyzer detection parameters and sensory evaluation brittleness grades, and fitting a linear regression equation; and substituting the detection parameters of the texture analyzer into the linear regression equation to obtain a predicted value of the crisp tilapia, performing ROC curve analysis, determining a threshold value corresponding to the crisp grade division, and obtaining a crisp grading standard of the crisp tilapia. Compared with a traditional detection method, based on the technical scheme, the brittleness of the crisp tilapia can be graded more objectively and accurately only according to the detection results of hardness, elasticity and resilience of the fresh crisp tilapia, and a simpler method is provided for quality detection of the crisp tilapia.

Description

Technical Field

[0001] The present invention belongs to the technical field of quality detection of tilapia, and particularly relates to a method for establishing a crispness grading standard for crispy tilapia and a method for grading the crispness of crispy tilapia. Background Art

[0002] Crispness is one of the most important indicators of crispy tilapia, which determines the quality of the taste. In order to distinguish the differences between different quality fish meats, it is necessary to conduct sensory evaluation to complete the classification of crispness levels. Traditionally, the crispness evaluation of crispy tilapia mainly relies on manual inspection, which has low efficiency and poor reliability. Coupled with the lack of a unified grading standard, the grading results of crispy tilapia are often chaotic, which is not conducive to the healthy development of the crispy tilapia industry. Moreover, for tilapia processing enterprises, manual inspection cannot meet the requirements of large-scale production. Therefore, it is necessary to adopt an evaluation method that does not rely on human senses and is convenient for automated operation to complete the grading of tilapia crispness.

[0003] Chinese Patent CN118112190A discloses a method for grading the crispness of crispy tilapia, which uses a texture analyzer to measure the crispness of raw crispy tilapia, sets up sensory evaluation crispness levels for cooked crispy tilapia, and conducts sensory evaluation at the same time. The correlation analysis is carried out between the crispness value of raw crispy tilapia and the sensory score of cooked crispy tilapia, so as to set the comparison situation between the crispness value of raw crispy tilapia and the grading of cooked crispy tilapia. Although the grading of the crispness of crispy tilapia can be achieved, since only the hardness of one side of the fish meat sample is measured by a texture analyzer, the detection index used is single, and the regression relationship between the sensory index and the instrument detection index is not established, so the accuracy needs to be improved. Summary of the Invention

[0004] The purpose of the present invention is to make up for the deficiencies of the prior art, establish a crispness grading standard for crispy tilapia, provide a more convenient method for the quality detection of crispy tilapia, and objectively and accurately grade the crispness of crispy tilapia.

[0005] In order to achieve the above purpose, the present invention provides a method for establishing a crispness grading standard for crispy tilapia, including the following steps:

[0006] (1) After stunning different crispness fresh crispy tilapia, cut off the tails to bleed, remove the heads, bones and internal organs, wash them clean, and obtain crispy tilapia fish fillets; Peel the back part of the crispy tilapia fish fillets, and after sucking the surface moisture, obtain fish meat samples;

[0007] (2) Along the direction from the head to the tail, evenly divide at least 3 parts of the dorsal muscle on one side of the fish meat sample, and cut at least 2 test samples from each part; Measure the hardness, chewiness and adhesiveness of the test samples, and take the average value respectively to obtain different detection parameters;

[0008] (3) Develop a sensory evaluation standard table for cooked crispy tilapia, and classify the grades of cooked crispy tilapia according to this standard; slice the other dorsal muscle of the fish sample, steam it to obtain steamed crispy tilapia; score the steamed crispy tilapia according to the grade of the cooked crispy tilapia to obtain the sensory evaluation crispness classification;

[0009] (4) Conduct a linear regression analysis on the different detection parameters and the sensory evaluation crispness classification, and obtain the linear regression equation of different detection parameters by fitting through a linear regression model;

[0010] The linear regression model is: Y = constant + A1*H + A2*E + A3*R; where Y is the predicted value, H represents hardness, the unit is g, E represents elasticity, R represents resilience, A1 is the coefficient corresponding to hardness, A2 is the coefficient corresponding to elasticity, and A3 is the coefficient corresponding to resilience;

[0011] (5) Substitute the different detection parameters into the linear regression equation, obtain the predicted value of the crispy tilapia, conduct an ROC curve analysis, determine the threshold corresponding to the crispness grade division, and obtain the grading standard; the grading standard is the crispness grading standard for crispy tilapia;

[0012] There is no time sequence limit between step (2) and step (3).

[0013] Preferably, the length of any one of the test samples in step (2) is 1.5 - 2.0 cm, the width is 1.5 - 2.0 cm, and the thickness is 1.5 - 2.0 cm.

[0014] Preferably, in step (2), a texture analyzer is used to measure the test sample; the parameters of the texture analyzer are set as follows: the test mode is set to the full texture mode, the starting point parameter is set to 10 gf, the deformation target is set to 30 mm, the residence time is set to 3 s, the sampling rate is set to 6 pps, the number of cycles is set to 2 times, and it is automatically cycled.

[0015] Preferably, when measuring in step (2), three different parts of any one test sample are measured.

[0016] Preferably, the steaming time in step (3) is 6 - 8 min.

[0017] Preferably, the grades of the cooked crispy tilapia in step (3) include grade 0, grade 1, grade 2, grade 3, grade 4, and ≥ grade 5;

[0018] The quality elements of grade 0 are: soft meat, no elasticity, and no obvious chewing feeling;

[0019] The quality elements of grade 1 are: relatively soft meat, slightly elastic, and slightly chewy;

[0020] The quality factors for level 2 are: delicate texture, small elasticity, and average chewiness;

[0021] The quality factors for level 3 are: relatively firm texture, springy, and strong chewiness;

[0022] The quality factors for level 4 are: firm texture, strong elasticity, and strong chewiness;

[0023] The quality factors for level ≥5 are: tight texture, strong elasticity, and difficult to chew.

[0024] Preferably, the linear regression model in step (4) is: Y = 0.038 + 0.004*H - 0.018*E + 0.503*R.

[0025] Preferably, the grading standard in step (5) is: Y < 0.984, crispness is level 0; 0.984 ≤ Y < 1.51, crispness is level 1; 1.516 ≤ Y < 2.280, crispness is level 2; 2.280 ≤ Y < 3.234, crispness is level 3; 3.234 ≤ Y < 4.718, crispness is level 4; Y ≥ 4.718, crispness ≥ level 5.

[0026] The present invention also provides an application of the grading standard obtained by the establishment method described in the above technical solution in the crispness grading of crisp tilapia.

[0027] The present invention also provides a method for grading the crispness of crisp tilapia, comprising the following steps:

[0028] (1) Stun the fresh crisp tilapia to be tested, cut off the tail to drain blood, remove the head, bones, and internal organs, wash cleanly, and obtain a crisp tilapia fish fillet; Peel the back part of the crisp tilapia fish fillet, absorb the surface moisture, and obtain a fish meat sample;

[0029] (2) Along the direction from the head to the tail, evenly divide one side of the dorsal muscle of the fish meat sample into at least 3 parts, and cut at least 2 test samples from each part; Measure the hardness, chewiness, and adhesiveness of the test samples, and take the average value respectively to obtain different detection parameters;

[0030] (3) Substitute the different detection parameters into the linear regression equation to obtain a predicted value; The linear regression equation is the linear regression equation obtained by the establishment method described in the above technical solution;

[0031] (4) Correlate the predicted value with the grading standard to obtain the crispness level of the fresh crisp tilapia to be tested; The grading standard is the grading standard obtained by the establishment method described in the above technical solution.

[0032] Beneficial effects:

[0033] The present invention provides a method for establishing a crispness grading standard for crispy tilapia, which measures the hardness, elasticity, and resilience of fresh crispy tilapia to obtain different detection parameters; sets up a sensory evaluation of the crispness grading for cooked crispy tilapia; performs a linear regression analysis on different detection parameters and the sensory evaluation of the crispness grading, and obtains a linear regression equation for different detection parameters through fitting with a linear regression model; substitutes different detection parameters into the linear regression equation to obtain the predicted value of the crispy tilapia, performs a ROC curve analysis, determines the threshold corresponding to the crispness grade division, and obtains the crispness grading standard for crispy tilapia. Based on the crispness grading standard of the crispy tilapia, compared with the traditional detection method, it can more objectively and accurately grade the crispness of the crispy tilapia.

[0034] When grading the crispness of the crispy tilapia according to the present invention, only the detection results of the hardness, elasticity, and resilience of the fresh crispy tilapia need to be substituted into the linear regression equation obtained by the above establishment method to obtain the predicted value; the predicted value is corresponded to the grading standard obtained by the above establishment method, and the crispness grade of the crispy tilapia can be objectively and accurately obtained, providing a more convenient method for the quality detection of the crispy tilapia. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments.

[0036] Figure 1 It is a schematic diagram of the sampling part of the crispy tilapia;

[0037] Figure 2 It is a schematic diagram of cutting and sampling the crispy tilapia;

[0038] Figure 3 It is a schematic diagram of the placement when testing the fish meat sample with a texture analyzer;

[0039] Figure 4 It is the ROC curve of the Y value of adjacent crispness grades; among them, the grades from A to C are 1 - 2 grades, 2 - 3 grades, and 3 - 4 grades in sequence. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] The present invention provides a method for establishing a crispness grading standard for crispy tilapia, including the following steps:

[0041] (1) Stun the fresh crispy tilapia to be tested, cut off its tail to bleed, remove its head, bones, and internal organs, wash it clean to obtain a crispy tilapia fish fillet; peel the back part of the crispy tilapia fish fillet, and after absorbing the surface moisture, obtain a fish meat sample;

[0042] (2) Along the head-to-tail direction, evenly divide one side of the dorsal muscle of the fish meat sample into at least 3 parts, and cut at least 2 test samples from each part; measure the hardness, chewiness, and adhesiveness of the test samples, and take the average values respectively to obtain different detection parameters;

[0043] (3) Develop a sensory evaluation standard table for cooked crispy tilapia, and classify the grades of cooked crispy tilapia according to this standard; slice the other side of the dorsal muscle of the fish meat sample and steam it to obtain steamed crispy tilapia; score the steamed crispy tilapia according to the grades of cooked crispy tilapia to obtain the sensory evaluation crispness grading;

[0044] (4) Conduct a linear regression analysis on the different detection parameters and the sensory evaluation crispness grading, and obtain the linear regression equation of the different detection parameters by fitting through a linear regression model;

[0045] The linear regression model is: Y = constant + A1*H + A2*E + A3*R; where Y is the predicted value, H represents hardness, the unit is g, E represents elasticity, R represents resilience, A1 is the coefficient corresponding to hardness, A2 is the coefficient corresponding to elasticity, and A3 is the coefficient corresponding to resilience;

[0046] (5) Substitute the different detection parameters into the linear regression equation to obtain the predicted value of the crispy tilapia, conduct a ROC curve analysis, determine the threshold corresponding to the crispness grade classification, and obtain the grading standard; the grading standard is the crispness grading standard for crispy tilapia;

[0047] There is no time sequence limit between step (2) and step (3).

[0048] In the present invention, fresh crispy tilapia with different crispness is stunned, then the tail is cut to bleed, the head, bones, and internal organs are removed, and it is cleaned to obtain a crispy tilapia fillet; the skin of the fish back part of the crispy tilapia fillet is removed, and after absorbing the surface moisture, a fish meat sample is obtained.

[0049] As an implementation method, the sample quantity of the fresh crispy tilapia with different crispness in the present invention ≥ 34; as another implementation method, the sample quantity of the fresh crispy tilapia with different crispness in the present invention ≥ 36; as an implementation method, the sample quantity of the fresh crispy tilapia with different crispness in the present invention ≥ 44; as an implementation method, the sample quantity of the fresh crispy tilapia with different crispness in the present invention ≥ 52.

[0050] After obtaining the fish meat sample, in the present invention, along the head-to-tail direction, evenly divide one side of the dorsal muscle of the fish meat sample into at least 3 parts, and cut at least 2 test samples from each part; measure the hardness, chewiness, and adhesiveness of the test samples, and take the average values respectively to obtain different detection parameters.

[0051] As an implementation method, along the head-to-tail direction, one side of the dorsal muscle of the fish meat sample is evenly divided into 3 parts, and 2 test samples are cut and prepared from each part. By cutting the dorsal muscle (i.e., the muscle on the back of tilapia), the accuracy of crispness measurement can be improved compared with the muscles on the belly and tail of the fish.

[0052] As an implementation method, the cutting method of the present invention is longitudinal cutting perpendicular to one side of the dorsal muscle of the fish meat sample. By defining the cutting method, the deviation can be further avoided and the accuracy of crispness measurement can be improved.

[0053] As an implementation method, the length of any one of the test samples is 1.5 - 2.0 cm; as another implementation method, the length of any one of the test samples is 1.5 cm. As an implementation method, the width of any one of the test samples is 1.5 - 2.0 cm; as another implementation method, the width of any one of the test samples is 1.5 cm. As an implementation method, the thickness of any one of the test samples is 1.5 - 2.0 cm; as another implementation method, the thickness of any one of the test samples is 1.5 cm. By defining the specifications of the test samples, the accuracy of crispness measurement can be further improved.

[0054] As an implementation method, the present invention uses a texture analyzer to measure the test samples. The parameters of the texture analyzer are set as follows: the test mode is set to the full texture mode, the starting point parameter is set to 10 gf, the deformation target is set to 30 mm, the dwell time is set to 3 s, the sampling rate is set to 6 pps, the number of cycles is set to 2 times, and it is automatically cycled. As an implementation method, when measuring with the texture analyzer, 3 different parts of any one test sample are measured. As an implementation method, the present invention places the test samples on ice covered with plastic wrap, with the peeled surface perpendicular to the horizontal plane and the cut longitudinal surface facing up, waiting for measurement. By defining the placement position of the test samples, the deviation can be further reduced and the accuracy of crispness measurement can be improved.

[0055] After obtaining the fish meat sample, the present invention formulates a sensory evaluation standard table for cooked crispy tilapia and divides the grades of cooked crispy tilapia according to this standard; slices the other side of the dorsal muscle of the fish meat sample and steams it to obtain steamed crispy tilapia; scores the steamed crispy tilapia according to the grades of the cooked crispy tilapia to obtain the sensory evaluation crispness grading.

[0056] As an implementation method, the steaming time of the present invention is 6 - 8 min; as another implementation method, the steaming time of the present invention is 7 min. By defining the steaming time, the best crispness of the edible taste can be guaranteed.

[0057] As an implementation, the grades of the crispy and cooked tilapia of the present invention include grade 0, grade 1, grade 2, grade 3, grade 4, and ≥ grade 5; the quality elements of grade 0 are: soft meat, no elasticity, and no obvious chewing feeling; the quality elements of grade 1 are: relatively soft meat, slightly elastic, and slightly chewing feeling; the quality elements of grade 2 are: delicate meat, small elasticity, and general chewing feeling; the quality elements of grade 3 are: relatively firm meat, springy, and strong chewing feeling; the quality elements of grade 4 are: firm meat, strong elasticity, and strong chewing feeling; the quality elements of ≥ grade 5 are: tight meat, strong elasticity, and difficult to chew.

[0058] As an implementation, the present invention forms an evaluation group of personnel and conducts multiple sensory evaluation trainings according to national standards GB / T 14195-1993 and GB / T 16860-1997. Since the taste deviations of different evaluators are not the same, in order to avoid the prejudice of evaluators against the product due to factors such as their respective different preferences, 8 people are used for sensory evaluation in each batch. Each evaluator conducts independent tasting and gives a result evaluation.

[0059] As an implementation, after slaughtering the crispy tilapia, it is stored under the condition of 4°C and the determination is completed within 0-24 hours; as another implementation, after slaughtering the crispy tilapia, it is stored under the condition of 4°C and the determination is completed within 4 hours. By limiting the time point of the determination, the present invention can further improve the accuracy of the determination of the crispness grade of the crispy tilapia fish meat.

[0060] After obtaining the different detection parameters and the sensory evaluation crispness grading, the present invention conducts a linear regression analysis on the different detection parameters and the sensory evaluation crispness grading, and obtains a linear regression equation of different detection parameters through fitting by a linear regression model; the linear regression model is: Y = constant + A1*H + A2*E + A3*R; where Y is the predicted value, H represents hardness, the unit is g, E represents elasticity, R represents resilience, A1 is the coefficient corresponding to hardness, A2 is the coefficient corresponding to elasticity, and A3 is the coefficient corresponding to resilience.

[0061] As an implementation, the linear regression model of the present invention is: Y = 0.038 + 0.004*H - 0.018*E + 0.503*R.

[0062] After obtaining the linear regression equation, the present invention substitutes the different detection parameters into the linear regression equation, obtains the predicted value of the crispy tilapia, conducts an ROC curve analysis, determines the threshold corresponding to the crispness grade division, and obtains a grading standard; the grading standard is the crispness grading standard of the crispy tilapia.

[0063] As an implementation manner, the grading standard of the present invention is as follows: when Y < 0.984, the brittleness is grade 0; when 0.984 ≤ Y < 1.51, the brittleness is grade 1; when 1.516 ≤ Y < 2.280, the brittleness is grade 2; when 2.280 ≤ Y < 3.234, the brittleness is grade 3; when 3.234 ≤ Y < 4.718, the brittleness is grade 4; when Y ≥ 4.718, the brittleness ≥ grade 5.

[0064] The present invention also provides an application of the grading standard obtained by the establishment method described in the above technical solution in the brittleness grading of crispy tilapia.

[0065] The present invention also provides a method for grading the brittleness of crispy tilapia, which includes the following steps:

[0066] (1) After knocking out the fresh crispy tilapia to be tested, cut off its tail to bleed, remove the head, bones and internal organs, wash it clean, and obtain the crispy tilapia fish fillet; peel the back part of the crispy tilapia fish fillet, and after sucking dry the surface moisture, obtain the fish meat sample;

[0067] (2) Along the direction from the head to the tail, evenly divide one side of the back muscle of the fish meat sample into at least 3 parts, and cut and prepare at least 2 test samples for each part; measure the hardness, chewiness and adhesiveness of the test samples, and take the average value respectively to obtain different detection parameters;

[0068] (3) Substitute the different detection parameters into the linear regression equation to obtain the predicted value; the linear regression equation is the linear regression equation obtained by the establishment method described in the above technical solution;

[0069] (4) Correlate the predicted value with the grading standard to obtain the brittleness grade of the fresh crispy tilapia to be tested; the grading standard is the grading standard obtained by the establishment method described in the above technical solution.

[0070] As an implementation manner, the cutting method of the present invention is longitudinal cutting perpendicular to one side of the back muscle of the fish meat sample. By defining the cutting method, the present invention can further avoid deviation and improve the accuracy of brittleness measurement.

[0071] As an implementation manner, the present invention uses a texture analyzer to measure the test samples, and the parameters of the texture analyzer are set as follows: the test mode is set to the whole texture mode, the starting point parameter is set to 10 gf, the deformation target is set to 30 mm, the residence time is set to 3 s, the sampling rate is set to 6 pps, the number of cycles is set to 2 times, and it is automatically cycled.

[0072] As an implementation manner, the method for measuring the hardness, chewiness and adhesiveness of the fish meat sample of the present invention includes: along the head-to-tail direction, evenly divide one side of the dorsal muscle of the fish meat sample into at least 3 parts, and cut at least 2 test samples from each part; use a texture analyzer to measure the hardness, chewiness and adhesiveness of the test samples, and take the average values respectively to obtain different detection parameters; As another implementation manner, the present invention evenly divides one side of the dorsal muscle of the fish meat sample into 3 parts along the head-to-tail direction, and cuts 2 test samples from each part. As an implementation manner, when the present invention conducts the measurement, 3 different parts of any one test sample are measured.

[0073] As an implementation manner, the present invention substitutes the different detection parameters into the linear regression equation Y = 0.038 + 0.004*H - 0.018*E + 0.503*R; where Y is the predicted value, H represents hardness, the unit is g, E represents elasticity, and R represents resilience.

[0074] As an implementation manner, the present invention corresponds the predicted value to the grading standard to obtain the crispness grade of the to-be-tested fresh crispy tilapia; the grading standard is: Y < 0.984, the crispness is grade 0; 0.984 ≤ Y < 1.51, the crispness is grade 1; 1.516 ≤ Y < 2.280, the crispness is grade 2; 2.280 ≤ Y < 3.234, the crispness is grade 3; 3.234 ≤ Y < 4.718, the crispness is grade 4; Y ≥ 4.718, the crispness ≥ grade 5.

[0075] In order to further illustrate the present invention, the following combines the drawings and embodiments to describe in detail a method for establishing a crispness grading standard for crispy tilapia and a method for grading the crispness of crispy tilapia provided by the present invention, but they cannot be understood as limiting the protection scope of the present invention.

[0076] Example 1

[0077] A method for grading the crispness of crispy tilapia consists of the following steps:

[0078] 1. Sample acquisition

[0079] Provided by the tilapia farm of Guangdong Youpei Supply Chain Co., Ltd., with an average mass of 1 - 2 kg. Stun the live crispy tilapia (provided by the tilapia farm of Guangdong Youpei Supply Chain Management Co., Ltd., with an average mass of 1 - 2 kg), cut off the tail to bleed, remove the head, bones and internal organs, and wash it clean with running water, seal it in a bag, and transport it to the laboratory with ice bag preservation.

[0080] 2. Measurement of texture analyzer parameters of crispy tilapia

[0081] (1) Select the dorsal muscle of the crispy tilapia (from the middle of the dorsal fin to the beginning of the caudal fin, such asFigure 1 the marked B area), according to Figure 2 For each fish sample, 6 test samples with a size of 1.5 cm × 1.5 cm × 1.5 cm (length × width × thickness) were cut. Specifically: At 4°C under chilled conditions, the skin was removed from the back of the crispy tilapia. The moisture on the surface of the skinned fish meat sample was blotted dry with kitchen absorbent paper. The unilateral dorsal muscle of the fish meat sample was cut for sample preparation with a knife. The unilateral dorsal muscle of the fish meat sample was evenly cut into three large pieces, named 1, 2, and 3 in sequence from the head to the tail; 2 test samples were cut from each sample block respectively.

[0082] (2) According to Figure 3 the placement method, the skinned surface of the test sample faced left and the longitudinal surface faced up, and it was placed on ice covered with plastic wrap and waited to be measured.

[0083] (3) Using a texture analyzer (Shanghai Tengba RapidTA practical domestic texture analyzer), the longitudinal section of the test sample of each fish was detected, and three parts of each sample were measured, and six parameters of the crispy tilapia were measured, namely: hardness, elasticity, resilience, chewiness, adhesiveness, and cohesiveness. When detecting, the test mode was set to the full texture mode, the starting point (contact force) parameter was set to 10 gf, the deformation target was set to 30 mm, the dwell time was set to 3 seconds, the sampling rate was set to 6 pps, the number of cycles was set to 2 times, and it was automatically cycled.

[0084] 3. Sensory evaluation criteria and grade classification of cooked crispy tilapia:

[0085] The unilateral dorsal muscle of the fish meat sample on the other side was cut into fish slices, randomly numbered, and steamed in boiling water for 6 - 8 min to obtain cooked crispy tilapia, and it was rated according to the sensory evaluation criteria. Among them, the sensory evaluation method and criteria: The evaluation team members received multiple sensory evaluation trainings according to GB / T 14195 - 1993 and GB / T 16860 - 1997 of the national standard, and finally formulated a sensory evaluation standard table, and at the same time carried out the grade classification of cooked crispy tilapia. The specific content is shown in Table 1.

[0086] Table 1 Rating criteria for crispy tilapia

[0087] Description Rating The meat is soft, inelastic, and has no obvious chewing feeling Grade 0 The meat is relatively soft, slightly elastic, and has a slight chewing feeling Grade 1 The meat is delicate, with little elasticity, and has an average chewing feeling Grade 2 The meat is relatively firm, chewy, and has a strong chewing feeling Grade 3 The meat is firm, with strong elasticity, and has a strong chewing feeling Grade 4 The meat is tight, with strong elasticity, and is difficult to chew ≥ Grade 5

[0088] Since the taste deviations of different evaluators are not exactly the same, in order to avoid the bias of the evaluators towards the product due to factors such as their respective preferences and sensory threshold differences, 8 people participated in the sensory evaluation for each batch of this experiment. Each evaluator tasted independently and gave the result evaluation.

[0089] 4. Data entry and processing

[0090] (1) Enter the results of the texture analyzer parameter measurement and the sensory evaluation grades into the data using Excel, and perform statistical analysis on the data using software such as SPSS 27.0, the car package in R language, and GraphPad Prism 10.0. Use GraphPad Prism 10.0 for data visualization. The results are shown in Table 2.

[0091] Table 2 Results of the collinearity analysis of the linear regression model

[0092]

[0093] As can be seen from Table 2, the tolerances of hardness, chewiness, and adhesiveness are less than or equal to 0.1, and the variance inflation factor (VIF) of the expansion coefficient is much greater than 5, indicating obvious collinearity among these three variables.

[0094] (2) Based on the conclusion of step (1), use SPSS 27.0 software or the Hmisc package in R language to perform Spearman correlation analysis on the texture analyzer parameters (hardness, chewiness, and adhesiveness) and the sensory evaluation crispness. The results are shown in Table 3.

[0095] Table 3 Results of the Spearman correlation analysis of the texture analyzer parameters and the sensory evaluation indicators

[0096] Hardness (g) Elasticity Resilience Brittleness grade Hardness 1.000 <![CDATA[0.241 ** > <![CDATA[0.415 ** > <![CDATA[0.939 ** > Elasticity <![CDATA[0.241 ** > 1.000 <![CDATA[0.798 ** > <![CDATA[0.219 ** > Resilience <![CDATA[0.415 ** > <![CDATA[0.798 ** > 1.000 <![CDATA[0.382 ** > Brittleness grade <![CDATA[0.939 ** > <![CDATA[0.219 ** > <![CDATA[0.382 ** > 1.000

[0097] Note: ** indicates P < 0.01.

[0098] As can be seen from Table 3, the correlations between hardness, elasticity, resilience, and the crispness grade are statistically significant (P < 0.01), and a linear regression equation can be established using a linear regression model.

[0099] (3) According to the conclusion of step (2), use SPSS 27.0 software or R language to perform linear regression analysis on the texture analyzer parameters (hardness, elasticity, and resilience) and the sensory evaluation crispness, and establish a linear regression equation. Among them, the summary of the linear regression model and the results of the ANOVA analysis are shown in Table 4 and Table 5 in turn, and the results of the linear regression analysis are shown in Table 6. The results of the collinearity analysis show that there is no collinearity among hardness, elasticity, and resilience, which can be used to construct a regression equation.

[0100] Table 4 Summary of the linear regression model

[0101] R <![CDATA[R 2 > <![CDATA[Adjusted R 2 > Standard error Durbin-Watson 0.909 0.827 0.824 0.441 0.241

[0102] Table 5 Results of the ANOVA analysis

[0103] Sum of squares Degree of freedom Mean square F Significance Regression 153.935 3 51.312 263.892 0.000 Residual 32.277 166 0.194 / / Total 186.212 169 / / /

[0104] Table 6 Results of the collinearity analysis of the linear regression model

[0105]

[0106] Taking the crispness grade as the dependent variable, (constant), resilience, hardness, and elasticity as the predictive variables, model establishment and variance analysis were carried out, and the obtained linear regression equation is shown in Equation Ⅰ.

[0107] Y = 0.038 + 0.004 * H - 0.018 * E + 0.503 * R Equation Ⅰ;

[0108] Among them, H represents hardness (g), E represents elasticity, and R represents resilience.

[0109] (4) According to the regression equation in step (3), calculate the mean and standard deviation of the Y values (predicted values) of the crispy tilapia with 6 crispness grades, and the results are shown in Table 7.

[0110] Table 7 Mean and standard deviation of predicted values of 6 crispness grade indicators

[0111] Grade Hardness (g) Elasticity Resilience Predicted value Y Grade 0 - - - <0.984 Grade 1 (n = 36) 284.61±52.92 0.62±0.06 0.33±0.06 <![CDATA[1.33±0.22 d > Grade 2 (n = 52) 425.11±65.44 0.63±0.09 0.36±0.07 <![CDATA[1.91±0.27 c > Grade 3 (n = 44) 596.70±83.56 0.64±0.06 0.38±0.07 <![CDATA[2.60±0.33 b > Grade 4 (n = 34) 872.14±136.38 0.66±0.08 0.42±0.07 <![CDATA[3.73±0.54 a > ≥ Grade 5 - - - ≥4.718

[0112] Note: n represents the number of samples, a, b, c, d are classification markers, the same marker indicates no significant difference, and different markers indicate significant differences.

[0113] It can be seen from Table 7 that there are significant differences in the predicted values of the 6 crispness grades of the crispy tilapia.

[0114] 5. Using Prism software, with sensitivity (Sensitivity) as the vertical axis and 1 - specificity (1 - Specificity) as the horizontal axis, perform ROC curve analysis on the Y values (predicted values) of the crispness data of adjacent two grades to determine the optimal threshold of the Y value corresponding to the crispness grade division; the evaluation basis is to compare the area under the ROC curve (AUC). When AUC is greater than 0.5, the closer AUC is to 1, the better the model performance, indicating better diagnostic effect. If it is less than 0.5, it means the accuracy of the model is poor. The evaluation results are as Figure 4 shown.

[0115] According to Figure 4 the ROC curve prediction results, the optimal thresholds of the Y values corresponding to the crispness grade division are as follows: for grades 1 - 2, the Y value threshold is 1.516, AUC = 0.9465; for grades 2 - 3, the Y value threshold is 2.280, AUC = 0.9665; for grades 3 - 4, the Y value threshold is 3.234, AUC = 0.9629.

[0116] In summary, based on the distribution range of the Y value (predicted value) of the dataset in Table 7 and the ROC analysis results, the following brittleness grading criteria are constructed: Y < 0.984, brittleness level 0; 0.984 ≤ Y < 1.516, brittleness level 1; 1.516 ≤ Y < 2.280, brittleness level 2; 2.280 ≤ Y < 3.234, brittleness level 3; 3.234 ≤ Y < 4.718, brittleness level 4; Y ≥ 4.718, brittleness ≥ level 5.

[0117] Example 2

[0118] Another 15 crispy tilapia (average mass 1 - 2 kg) were randomly selected from the tilapia farm of Guangdong Province Youpei Supply Chain Co., Ltd. and processed in the same manner as in Example 1. The hardness, elasticity, and resilience of the crispy tilapia were measured using a texture analyzer, and the Y value was calculated by substituting into the regression equation (Equation Ⅰ). According to the brittleness grading criteria based on the Y value; moreover, the sensory brittleness of the cooked crispy tilapia was evaluated according to the scoring criteria in Table 1, and the results are shown in Table 8.

[0119] Table 8 Texture analyzer data and brittleness grade test results of crispy tilapia

[0120]

[0121]

[0122] As can be seen from Table 8, compared with the brittleness grading results obtained by discriminating with the Y value (predicted value), the accuracy rate of the brittleness grading results by sensory evaluation is as high as 86.7%.

[0123] From the above content, it can be seen that the brittleness grading method for crispy tilapia proposed by the present invention has high accuracy and can be used as the determination basis for tilapia brittleness grading.

[0124] Although the above embodiments have described the present invention in detail, they are only a part of the embodiments of the present invention, not all embodiments. People can also obtain other embodiments based on this embodiment without creative efforts, and these embodiments all fall within the protection scope of the present invention.

Claims

1. A method for establishing a crispness grading standard for crispy tilapia, characterized in that: The steps include: (1) knocking out fresh crispy tilapia with different crispness, cutting the tail and bleeding, removing the head, bones and internal organs, and cleaning to obtain crispy tilapia fillets; peeling the back of the crispy tilapia fillets, absorbing surface moisture to obtain fish meat samples; (2) dividing the back muscle of one side of the fish meat sample into at least three parts along the direction from the head to the tail, cutting each part to prepare at least two samples to be tested; measuring the hardness, chewiness and adhesiveness of the samples to be tested, taking average values ​​respectively, and obtaining different test parameters; (3) formulating a sensory evaluation standard table for cooked crispy tilapia, and classifying the cooked crispy tilapia according to the standard; slicing the back muscle of the other side of the fish sample, steaming the slices, and obtaining the steamed crispy tilapia; scoring the steamed crispy tilapia according to the grade of cooked crispy tilapia, and obtaining a sensory evaluation crispness grade; (4) performing linear regression analysis on the different detection parameters and the sensory evaluation crispness classification, and obtaining linear regression equations of the different detection parameters by linear regression model fitting; The linear regression model is: Y=constant+A1*H+A2*E+A3*R; wherein Y is the predicted value, H represents hardness in g, E represents elasticity, R represents resilience, A1 is the coefficient corresponding to hardness, A2 is the coefficient corresponding to elasticity, and A3 is the coefficient corresponding to resilience; (5) bringing the different detection parameters into the linear regression equation to obtain the predicted value of the crispy tilapia, performing ROC curve analysis, determining the threshold value corresponding to the crispness grade division, and obtaining a grading standard; the grading standard is the crispy tilapia crispness grading standard; There is no time sequence limitation between step (2) and step (3).

2. The establishment method according to claim 1, characterized in that: In step (2), any one of the samples to be tested has a length of 1.5 to 2.0 cm, a width of 1.5 to 2.0 cm, and a thickness of 1.5 to 2.0 cm.

3. The establishment method according to claim 1, characterized in that: Step (2) using a texture analyzer to measure the sample to be tested; the texture analyzer parameters are set as follows: the test mode is set to the full texture mode, the starting point parameter is set to 10gf, the deformation target is set to 30mm, the dwell time is set to 3s, the number of points is set to 6pps, the number of cycles is set to 2 times, and automatic cycle is set.

4. The establishment method according to claim 3, characterized in that: When performing the measurement in step (2), three different parts of any sample to be measured are measured.

5. The establishment method according to claim 1, characterized in that: The steaming time in step (3) is 6 to 8 minutes.

6. The establishment method according to claim 1, characterized in that: The grades of the cooked crispy tilapia in step (3) include grade 0, grade 1, grade 2, grade 3, grade 4 and grade ≥5; The quality factors of grade 0 are: soft meat, no elasticity, and no obvious chewing feeling; The quality factors of level 1 are: soft meat, slightly elastic, and slightly chewy; The quality factors of grade 2 are: fine texture, low elasticity, and average chewing feeling; The quality factors of the three grades are: firm meat, springy, and chewy; The quality factors of the four grades are: firm meat, strong elasticity, and strong chewiness; The quality factors of grade ≥5 are: firm meat, strong elasticity and difficult to chew.

7. The establishment method according to any one of claims 1 to 6, characterized in that: The linear regression model in step (4) is: Y=0.038+0.004*H-0.018*E+0.503*R.

8. The establishment method according to any one of claims 1 to 6, characterized in that: The grading standard in step (5) is: Y value < 0.984, brittleness is level 0; 0.984 ≤ Y < 1.516, brittleness is level 1; 1.516 ≤ Y < 2.280, brittleness is level 2; 2.280 ≤ Y < 3.234, brittleness is level 3; 3.234 ≤ Y < 4.718, brittleness is level 4; Y ≥ 4.718, brittleness ≥ level 5.

9. Application of the grading standard obtained by the establishment method according to any one of claims 1 to 8 in the crispness grading of crispy tilapia.

10. A method for grading the crispness of crispy tilapia, characterized in that: The steps include: (1) knocking out a fresh crispy tilapia to be tested, cutting off the tail and bleeding, removing the head, bones and internal organs, and cleaning to obtain a crispy tilapia fillet; peeling the back of the crispy tilapia fillet, and drying the surface moisture to obtain a fish meat sample; (2) dividing the back muscle of one side of the fish meat sample into at least three parts along the direction from the head to the tail, cutting each part to prepare at least two samples to be tested; measuring the hardness, chewiness and adhesiveness of the samples to be tested, taking average values ​​respectively, and obtaining different test parameters; (3) Substituting the different detection parameters into a linear regression equation to obtain a predicted value; the linear regression equation is a linear regression equation obtained by the establishment method according to any one of claims 1 to 8; (4) The predicted value is matched with a grading standard to obtain the crispness grade of the fresh crispy tilapia to be tested; the grading standard is the grading standard obtained by the establishment method according to any one of claims 1 to 8.

Images