CN102507882A - Beef quality multi-parameter compressive evaluation method - Google Patents

Abstract

The invention discloses a beef-quality multi-parameter compressive evaluation method and belongs to the range of meat quality evaluation. The evaluation method comprises the following steps of: on the basis of taking beef tenderness and beef moisture content as evaluation indexes of beef compressive quality, utilizing the beef quality multi-parameter compressive evaluation method which combines subjective evaluation and objective evaluation, and utilizing a large amount of experiments and verifications performed by a beef quality multi-parameter online detection system, (1) respectively establishing a predication model for performing online detection on the beef tenderness and the beef moisture content; (2) carrying out industrial field online detection: carrying out online grading detection on the inner quality of the beef; and (3) utilizing a compressive quality evaluation model to carry out statistic analysis on judgment of subjective evaluation staff to obtain a summary sheet of subjective evaluation results, and combining the two steps to carry out the compressive evaluation. According to the invention, a model for converting a multi-target function into a single-target function is established, so that detection results are in one-to-one correspondence with sensory evaluation descriptions and the detection is much closer to a real market so as to guide enterprise production and market consumption, therefore, the method provided by the invention has the advantage of wide practical value.

Description

The multiple parameter overall assessment method of beef quality

Technical field

The invention belongs to meat quality evaluation scope, particularly a kind of multiple parameter overall assessment method of beef quality.

Background technology

Food security and quality are two aspects that the consumer pays close attention to food the most; Under the prerequisite that health quarantine department effectively guarantees food security; Beef quality then is the main factor of the decision consumer's purchase intention and the market price; The evaluation of rational and effective beef quality all has a very important role to the whole beef production and the process of circulation.

The quality of beef mainly comprises appearance attribute (like color), mouthfeel (tender degree, water cut etc.) and local flavor (like fragrance, succulence etc.), and wherein, tender degree, water percentage are the main factors of quality-determining.According to the consumption survey that Shackelford etc. is done, in these three factors of the tender degree, succulence and the local flavor that influence consumer's purchase decision, tender degree is again wherein topmost factor.

Current, the evaluation method of beef quality mainly contains three kinds: the one, and flavor evaluation, the sense of smell through the people, the sense of taste, sense of touch etc. are carried out direct evaluation of meat quality to the tender degree of beef, mouthfeel, local flavor etc.; The 2nd, the organoleptic indicator's (like marbling, yellowish pink, fatty look etc.) to beef comes beef integrated quality (like tender degree, mouthfeel, local flavor etc.) is evaluated indirectly through manual work; The 3rd, through the technical indicator such as shearing force, the pH value etc. that measure beef quality is estimated.Method one is the quality characteristic of reflection beef the most reliably, but it to grading personnel's professional standards, estimate environment and sample is prepared to have higher requirements, process is complicated; Method two is simple to operate fast, but the correlativity of the index of surveying and beef quality be not very high, it is bigger that the result is influenced by varietal difference; Method three has objectively reflected beef quality in a certain respect, but can't reflect the integrated quality of beef accurately.

Summary of the invention

The objective of the invention is to propose a kind of multiple parameter overall assessment method of beef quality; It is characterized in that this evaluation method is to be beef integrated quality evaluation index with tenderness of beef utilizing and water percentage, adopt the multiple parameter overall assessment method of the beef quality that subjective assessment combines with objective data; The a large amount of experiments and the correction of the on-line detecting system through the beef quality multiparameter; Set up the forecast model that online detection tenderness of beef utilizing, water percentage are set up in (1) respectively, the online detection of (2) industry spot, online grading detects (3) integrated quality evaluation model to the beef interior quality; Statistical study is carried out in subjective evaluation and test personnel's judge; Obtain the summary sheet of a subjective evaluation result, comprehensive evaluation is carried out in the two combination; Specifically comprise:

1) statistical study is carried out in all subjective evaluation and test personnel's judge, obtained the summary sheet and corresponding tender degree, the measured data statistical form of water percentage of a subjective evaluation result, as shown in table 1.Table 2 is depicted as meat grade, quality zone and mouthfeel and describes correspondence table.

Table 1, bright beef quality measured data and subjective contrast comprehensive evaluation form

Table 2, meat grade, quality zone and mouthfeel are described correspondence table

2) adopt the linear weighted function combined method with the conversion processing of multiple objective function, be about to a plurality of partial objectives for function f in the multi-objective optimization question to the single goal function ₁(X), f ₂(X) ..., f _n(X), distribute its corresponding weighting factor ω according to its magnitude and the significance level in global design ₁, ω ₂..., ω _n, get ω then _if _i(X) linear combination constitutes a unified goal function, that is:

$f\left(X\right)=\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\ω}}_j{f}_j\left(X\right),$ Wherein $\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\ω}}_j=1$

With the former multiple objective function f of single goal function f (X) comprehensive characterization _j(X) characteristic.

Utilize the linear weighted function combined method to construct the inherent integrated quality objective function of following beef for this patent:

F (x)=ω ₁J+ ω ₂W formula (1)

In the formula

J---tender degree; W---water percentage; ω ₁, ω ₂---weighting factor.

According to table 1, formulate the corresponding relation of the inherent integrated quality objective function f of beef (x) value and subjective appreciation grade.Wherein, the O-water-injected meat, G-is first-class, and F-is good, and E-is medium, and D-is general, and C-is poor, and B-is very poor, the A-extreme difference;

Comprehensive tenderness of beef utilizing, water cut value confirm that to the influence of beef quality grade the weight coefficient of each regional tenderness of beef utilizing and water cut value is as shown in table 3:

The weight coefficient of table 3 zones of different

3) the visible near-infrared on-line detecting system of hyperchannel of the inherent integrated quality of structure beef; The spectral information of collected specimens under visible near-infrared all band; And the actual value that obtains the tender degree of sample, water percentage with National Standard Method is as reference point; Set up the forecast model of shear power of beef, water percentage respectively, and model is proofreaied and correct through a large amount of experiments.Utilize the shear force value and the water cut value of this model on-line prediction sample.

4) with the predicted value of the above-mentioned tenderness of beef utilizing that obtains, water percentage,, obtain the respective value of tender degree j and water percentage W, consult the existing figure that confirms the hierarchical region of beef quality again, confirm the weight coefficient ω that sample is corresponding through look-up table 1 ₁, ω ₂, be updated to objective function f (x)=ω ₁J+ ω ₂Among the W, obtain the functional value of the inherent integrated quality of beef, and then obtain quality grade that the consumer is familiar with and corresponding mouthfeel description.

The invention has the beneficial effects as follows the beef quality online measuring technique; The on-line prediction that not only comprises each quality index values; Also comprise the problem how to use each index to come concentrated expression beef integrated quality; The present invention through making up the transformation model of multiple objective function to the single goal function, has realized the Comprehensive Evaluation Problem of the many index of quality of beef with the interior quality index digitizing of beef; The present invention describes testing result and sensory evaluation corresponding one by one, makes the detection reality that is close to the market more, can better guide enterprise production and market comsupton; Method provided by the invention, for the detection of beef integrated quality provides a kind of important realization approach, grading has practical value to the beef interior quality.

Description of drawings

Fig. 1 is a this patent method principle schematic.

Fig. 2 is for confirming the hierarchical region figure of beef quality.

Fig. 3 is the inherent integrated quality on-line detecting system of a beef synoptic diagram.

Embodiment

The present invention proposes a kind of multiple parameter overall assessment method that adopts subjective assessment to combine with objective data to estimate a kind of beef quality of beef quality.Explain below in conjunction with accompanying drawing and embodiment.

Principle schematic as shown in Figure 1, online detection forecast model is set up in (1) among the figure, the online detection of (2) industry spot, (3) integrated quality evaluation model; The multiple parameter overall assessment method of said beef quality obtains the summary sheet and corresponding tender degree, the measured data statistical form of water percentage of a subjective evaluation result at first need statistical study be carried out in all subjective evaluation and test personnel's judge, and is as shown in table 1.

Through tabling look-up, obtain the corresponding tender degree of beef quality (j), the detected value of water percentage (W).

When the detected value of using tender degree and these two indexs of water percentage is evaluated beef interior quality grade, need carry out one with the conversion processing of multiple objective function to the single goal function, promptly adopt the linear weighted function combined method.The basic thought of linear weighted function combined method is with a plurality of partial objectives for function f in the multi-objective optimization question ₁(X), f ₂(X) ..., f _n(X), distribute its corresponding weighting factor ω according to its magnitude and the significance level in global design ₁, ω ₂..., ω _n, get ω then _if _i(X) linear combination constitutes a unified goal function, that is:

$f\left(X\right)=\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\ω}}_j{f}_j\left(X\right),$ Wherein $\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\ω}}_j=1$

With the former multiple objective function f of single goal function f (X) comprehensive characterization _j(X) characteristic.

This patent utilizes the linear weighted function combined method to construct following beef interior quality objective function:

f(x)＝ω ₁·j+ω ₂·W

In the formula

J---tender degree; W---water percentage; ω ₁, ω ₂---weighting factor.

According to table 1, table 2, formulate the corresponding relation of beef interior quality objective function f (x) value and subjective appreciation grade, wherein, and the O-water-injected meat, G-is first-class, and F-is good, and E-is medium, and D-is general, and C-is poor, and B-is very poor, the A-extreme difference; Weight Determination is as shown in table 3; Each grade and beef interior quality objective function f (x) value corresponding relation are as shown in table 4.

Bright beef quality measured data of table 1 and subjective contrast comprehensive evaluation form

Table 2, meat grade, quality zone and mouthfeel are described correspondence table

The weight coefficient of table 3, zones of different

Table 4, each grade and beef interior quality objective function f (x) value mapping table

Secondly; The visible near-infrared on-line detecting system of hyperchannel of structure beef interior quality as shown in Figure 3; This system is removable visible and near infrared spectrum detection system, and the spectral detection system architecture is that to connect wavelength respectively be that first spectrometer 1, the wavelength of 400-950nm is second spectrometer 2, high power halogen tungsten lamp light source 3 and 4 optical fiber detection probe 7 of 900-2600nm to 2 * 8 road sonet multiplexers 4; Two spectrometers link to each other with computing machine 5 and display screen 6 through the usb data line; The spectral information of 4 optical fiber detection probe 7 collected specimens under visible near-infrared all band; And obtain the reference point of the tender degree of sample, water percentage with National Standard Method; Set up the forecast model of shear power of beef, water percentage respectively; And through a large amount of experiments model is proofreaied and correct, utilize the shear force value and the water cut value of this model on-line prediction sample.

Online testing process:

Detection system is embedded into carcass grading detects the operation place; Eye muscle cross section between the 12-13 sternal rib is detected; Operating personnel ask on average the characteristic spectrum of equal spectrum as sample of making even through the start and stop of touch switch control spectra collection process to four curves of spectrum of each sample;

Characteristic spectrum to sample is carried out pre-service, and the method that pretreated method is carried out when setting up tender degree, water cut prediction is consistent;

Pretreated characteristic spectrum is updated in tender degree, the water cut prediction, obtains corresponding predicted value;

The predicted value of tender degree, water percentage is updated in the beef interior quality evaluation model, and obtaining with tender degree, water percentage is the beef inherent quality grade of evaluation index

At last,,, obtain the respective value of tender degree j and water percentage W, consult Fig. 2 again, confirm the hierarchical region of beef quality, confirm the weight coefficient ω that sample is corresponding through look-up table 1 with the tenderness of beef utilizing that said method obtained, the predicted value of water percentage ₁, ω ₂, be updated to objective function f (x)=ω ₁J+ ω ₂Among the W, obtain the functional value of beef integrated quality, according to the functional value of the resulting beef quality of said method, can draw the quality grade that the consumer is familiar with again.

For example, detection system to certain sample predict the outcome for: shear force value 37.6N, water cut value is 72%.Table look-up 1, can know tender degree value j=5, water cut value W=4, according to j, W, comparison diagram 2 can know that sample belongs to E class meat, checks table 3, can know that weighting coefficient ω 1, ω 2 are respectively 0.5,0.5, with j=5, W=4, ω ₁=0.5, ω ₂In=0.5 substitution formula 1, obtain beef integrated quality functional value f (x)=4.5, table look-up and 4 can know that meat belongs to E level meat, promptly medium meat, sense organ is described as tender and blurts out.

Claims (1)

1. the multiple parameter overall assessment method of a beef quality; It is characterized in that this evaluation method is to be beef integrated quality evaluation index with tenderness of beef utilizing and water percentage, adopt the multiple parameter overall assessment method of the beef quality that subjective assessment combines with objective data; The a large amount of experiments and the correction of the on-line detecting system through the beef quality multiparameter; Set up the forecast model that online detection tenderness of beef utilizing, water percentage are set up in (1) respectively, the online detection of (2) industry spot, online grading detects (3) integrated quality evaluation model to the beef interior quality; Statistical study is carried out in subjective evaluation and test personnel's judge; Obtain the summary sheet of a subjective evaluation result, comprehensive evaluation is carried out in the two combination; Specifically comprise:

1) statistical study is carried out in all subjective evaluation and test personnel's judge, obtained the summary sheet and corresponding tender degree, the measured data statistical form of water percentage of a subjective evaluation result, as shown in table 1; Table 2 is depicted as meat grade, quality zone and mouthfeel and describes correspondence table;

Table 1, bright beef quality measured data and subjective contrast comprehensive evaluation form

Table 2, meat grade, quality zone and mouthfeel are described correspondence table

2) adopt the linear weighted function combined method with the conversion processing of multiple objective function, be about to a plurality of partial objectives for function f in the multi-objective optimization question to the single goal function ₁(X), f ₂(X) ..., f _n(X), distribute its corresponding weighting factor ω according to its magnitude and the significance level in global design ₁, ω ₂..., ω _n, get ω then _if _i(X) linear combination constitutes a unified goal function, that is:

$f\left(X\right)=\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\ω}}_j{f}_j\left(X\right),$ Wherein $\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\ω}}_j=1$

With the former multiple objective function f of single goal function f (X) comprehensive characterization _j(X) characteristic;

Utilize the linear weighted function combined method to construct the inherent integrated quality objective function of following beef for this patent:

F (x)=ω ₁J+ ω ₂W formula (1)

In the formula

J---tender degree; W---water percentage; ω ₁, ω ₂---weighting factor.

According to table 1, formulate the corresponding relation of the inherent integrated quality objective function f of beef (x) value and subjective appreciation grade.Wherein, the O-water-injected meat, G-is first-class, and F-is good, and E-is medium, and D-is general, and C-is poor, and B-is very poor, the A-extreme difference;

Comprehensive tenderness of beef utilizing, water cut value confirm that to the influence of beef quality grade the weight coefficient of each regional tenderness of beef utilizing and water cut value is as shown in table 3:

The weight coefficient of table 3 zones of different

3) the visible near-infrared on-line detecting system of hyperchannel of the inherent integrated quality of structure beef; The spectral information of collected specimens under visible near-infrared all band; And the actual value that obtains the tender degree of sample, water percentage with National Standard Method is as reference point; Set up the forecast model of shear power of beef, water percentage respectively, and model is proofreaied and correct through a large amount of experiments; Utilize the shear force value and the water cut value of this model on-line prediction sample;

4) with the predicted value of the above-mentioned tenderness of beef utilizing that obtains, water percentage,, obtain the respective value of tender degree j and water percentage W, consult the existing figure that confirms the hierarchical region of beef quality again, confirm the weight coefficient ω that sample is corresponding through look-up table 1 ₁, ω ₂, be updated to objective function f (x)=ω ₁J+ ω ₂Among the W, obtain the functional value of the inherent integrated quality of beef, and then obtain quality grade that the consumer is familiar with and corresponding mouthfeel description.

Images