CN101477101A - Method for pork carcass quality prediction - Google Patents
Method for pork carcass quality prediction Download PDFInfo
- Publication number
- CN101477101A CN101477101A CNA2008100193507A CN200810019350A CN101477101A CN 101477101 A CN101477101 A CN 101477101A CN A2008100193507 A CNA2008100193507 A CN A2008100193507A CN 200810019350 A CN200810019350 A CN 200810019350A CN 101477101 A CN101477101 A CN 101477101A
- Authority
- CN
- China
- Prior art keywords
- equation
- lean meat
- carcass
- pig
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the field of meat product processing, in particular to a method used in pork carcass quality prediction. The method is characterized in that a prediction equation of lean meat rate is that y is equal to 66.914 + 0.0682x1- 0.6789x9 or y is equal to 57.742 - 0.5871x7 +0.2023x10 or y is equal to 56.806- 0.1054x1 - 0.5316x7 + 0.3029x10, wherein y is lean meat rate (percent), x1 is hot carcass weight (kg), x7 is lumbosacral fat thickness (mm), x9 is F<b> (mm) and x10 is M<b> (mm). The method establishes three regression equations capable of accurately predicting pig carcass lean meat rate, and verifies the prior prediction model through measuring the physiological characters of market pigs, thereby improving the accuracy of the prediction model.
Description
Technical field
The invention belongs to field of meat product processing, specifically is a kind of method that is used for pork carcass quality prediction.
Background technology
The hog on hook classification is meant some the related economic proterties according to hog on hook, artificial it is divided into different grades, and the market purchasing price of different grades and hog on hook is connected, thereby realizes high quality and favourable price, promotes the raising of national market pig quality.
The research of external relevant hog on hook classification technique has had very long history, and it is perfect that the carcass grade assessment technology also has been tending towards, and they have been applied to these technology in the practice, goes to instruct raising to raise on a household basis the purchase and the processing of pig and meat production processing enterprise.Yet, the hog on hook classification technique of China is just at the early-stage, do not have a hog on hook grade evaluation system comprehensive, complete, system, therefore we also just can't accomplish to fix the price according to the quality, high quality and favourable price, make to raise the family and butcher the family and all can not reasonably carry out standard to it and handle, thereby make both sides all suffer enormous economic loss, trusting degree has each other also reduced.
As far back as earlier 1900s countries in the world just successively set up oneself stage division and standard according to national conditions separately.Yet the grade scale of each countries and regions also is not quite similar, and generally all judges according to indexs such as muscle development degree, fat deposition situation, genetic ability for carcass weight.
China in 1988 has set up GB/T 9959.1-88 " belt leather aquatic foods, freeze sheet pork " national standard.And this standard was revised in 2003, but every index and content are not almost adjusted.According to half genetic ability for carcass weight, back-fat thickness and butcher processing request pig half trunk is divided into Three Estate.
Strict, this standard is not the carcass grading standard, is that the applicability or the operability of grade scale of the economic target of quoting all exists very big problem
[7]The index of this standard is that late nineteen eighties is determined, be applicable to China's pork pig kind at that time, and the today after the more than ten years, the pork pig kind of China has had very big change, the economic target type of existing standard and the concrete incompatible already present actual conditions of numerical value can't be estimated hog on hook at all more accurately.In addition, the original hierarchy system of China can not connect with the market purchasing price, thereby can not promote the high quality of the market pig of China, stimulates China's pig industry and pork to butcher the development of processing industry.Therefore, formulate adapting to that China's the present situation can accurately estimate the carcass grading standard of Chinese commodity pig and set up the perfect market pig trunk payment purchase system of a cover is the present task of top priority.
Existing market mainly concentrates on lean meat percentage, tender degree, specious degree, nutritive value, homogeneity of product and " green " degree for the requirement of fresh meat based article.China is the big country that pork is produced, and produces pork per year and accounts for about 50% of world's pork total production, ranks first in the world.But the average lean meat percentage of Chinese commodity pig but has only 50%, compares with the relatively advanced country of some animal husbandry also to have very big gap.The carcass grading system that a wherein very important reason is exactly China is very backward, is that the research or the foundation of grade scale of classification technique has all lagged far behind other animal husbandry developed country.Thereby the live pig that causes China is purchased not science of the modes of payments utmost point, can not accomplish high quality and favourable price, can not promote the raising of market pig lean meat percentage.
On the other hand, the forecast model of the lean meat percentage rectification that should upgrade in time.Because divide the change of mixing along with hog on hook back-fat thickness reducing in heredity will cause tissue, thereby make the regression equation of setting up and be not suitable for existing colony in the past
[25]Once more, because the full trunk technology of ultrasound wave is ripe gradually, measures and become reality for the lose some benefits lean meat percentage of piece of each main advantage.Compare with carcass weight with simple mensuration lean meat percentage, the lean meat percentage that online estimation trunk is mainly cut apart cube meat is very valuable, so just can be for laying a good foundation based on the hierarchical approaches of the different cube meat prices that cut meat.
Summary of the invention
The purpose of this invention is to provide a kind of method that is used for pork carcass quality prediction, specifically set up the forecast model of new lean meat percentage, to be fit to existing colony.
The said method that is used for pork carcass quality prediction is characterized in that, the lean meat percentage predictive equation is y=66.914+0.0682 x
1-0.6789 x
9Or y=57.742-0.5871 x
7+ 0.2023 x
10Or y=56.806-0.1054x
1-0.5316 x
7+ 0.3029 x
10Wherein y is lean meat percentage (%), x
1Be hot carcass weight (kg); x
7Be that waist is recommended fat thickness (mm); x
9Be F
b(mm); x
10Be M
b(mm).
The said method of the present invention has been set up 3 regression equations that can accurately predict the hog on hook lean meat percentage, and passes through the measurement of the physiological character of market pig, and original forecast model is verified, has improved the accuracy of forecast model.
Description of drawings
Fig. 1 is the back fat distribution schematic diagram
Fig. 2 is the standardized residual distribution scatter diagram of equation 5
Fig. 3 is the standardized residual distribution scatter diagram of equation 6
Fig. 4 is the standardized residual distribution scatter diagram of equation 7
Fig. 5 is the predicted value and the measured value design sketch of equation 5
Fig. 6 is the predicted value and the measured value design sketch of equation 6
Fig. 7 is the predicted value and the measured value design sketch of equation 7
Embodiment
Embodiment
1, the basic statistics amount of each carcass trait of market pig
Table 1 shows that the average lean meat percentage of market pig trunk is (56.42 ± 7.03) %, and the evenly heat carcass weight is (58.49 ± 9.62) kg; The average fat thickness of market pig is between (19.47 ± 6.94) mm and (36.69 ± 8.41) mm.Fig. 1 (not containing the thinnest fat thickness of 9～13 ribs) shows back fat attenuation gradually earlier, after this begins progressive additive again to the X position, recommends first decline of regional fat thickness at X with waist and afterwards increases; The back fat of neck chest junction is the thickest in each measurement point, and thinnest part is the position between 9～13 ribs.But generally speaking, have a declining tendency along its fat thickness of vertebrae direction (first thoracic vertebrae is to recommending vertebra).To survey the thick mean value of M meat be (59.01 ± 7.50) mm.
The basic statistics amount of table 1 carcass trait
Annotate: in the table
aBe with reference to measured position in the Japanese pig half carcass grading standard;
bBe two indexs that adopted with reference to a kind of ZP technology in European Union's classification means;
cBe with reference to the measured position of Canadian hog on hook output grade scale.
2, correlation analysis between the market pig carcass trait
Table 2 shows, survey between 9 fat thickness and the carcass lean meat percentage and have extremely strong negative correlation (P＜0.0001).Wherein, with the carcass lean meat percentage correlativity the strongest be F position fat thickness (reaching-0.90353), secondly be that waist is recommended fat thickness and 3～4 rib fat thickness (being respectively-0.90045 ,-0.84692).The thinnest fat thickness of 9～13 ribs and X position fat thickness
[4]And the correlativity between the carcass lean meat percentage is respectively-0.8151 and-0.789; Correlativity remarkable (P＜0.05) between hot carcass weight and the lean meat percentage, and exist extremely strong positive correlation (P＜0.0001, wherein the correlativity with the M place is the strongest, reaches 0.59817), M place thickness and lean meat percentage to be extremely significantly positive correlation (P＜0.0001) between other proterties.There is extremely strong positive correlation (P＜0.0001) in each independent variable between back fat and the back fat; Related coefficient between M and the back fat is less, except that and last rib fat thickness between correlativity significantly, and all have conspicuousness relevant (P＜0.05) between other fat thickness.
Phenotypic correlation coefficient between table 2 market pig carcass trait,
3, the comparison of market pig carcass lean meat percentage predictive equation
Adopt the R in the SAS linear regression
2Back-and-forth method (R-square) determines that the subclass of the different independent variable numbers of optimum prediction dependent variable (considers actual operability, comprise 3 independents variable that need ruler to measure in the subclass at most), use total regression model (Full model) then different independents variable is carried out regretional analysis, therefrom select and have statistical significance (model and partial regression coefficient all have statistical significance) and representational model compare (table 3).With 6～7 fat thickness (x
4) be degree of fitting (the coefficient of determination R of the simple equation of dependent variable
2) be 0.6877, RMSE is 3.93548 (equations 1); Best model is an equation 4 in the monobasic linear model, its R
2Be 0.8164, RMSE is 2.75751; Only need ruler measure in the bivariate regression model of some thickness of backfat gained best model be equation 5 (R2=0.8259, RMSE=2.68943); The best bivariate regression model that needs ruler to measure 2 gained is equation 6 (R
2=0.8651, RMSE=2.36706); The best model that contains hot carcass weight variable on the basis of 2 of ruler measurements in the equation is an equation 7, its R
2Be 0.8753, RMSE is 2.27971.In all regression equations of table 3, the variable number in the equation is many more, its R
2Big more, RMSE is more little; The prediction effect of equation 10,11,12,13 is better than above-mentioned forecast model, and the accuracy of its model and degree of accuracy are all very high, but can increase operation easier during practical application.
If according to estimating formula R about lean meat percentage in European Community's hog on hook grade scale
20.64 and the requirement of RMSE＜2.5
[5], equation 1,2,3,4,5 does not satisfy its requirement, and other equation all meets.Though the forecast model of 3 point measurement gained (equation 10,11,12,13) is better than the forecast model of 2 point measurement gained, can additionally increase labor capacity when considering practical application, relatively waste time and energy, therefore when selecting better model, put aside the equation of measuring 3 indexs with ruler.In conjunction with China present pig industry present situation and the actual conditions that each slaughterhouse scale, equipment, technical merit etc. differ greatly, when practical application, advise adopting 5,6,7 three models of equation that carcass lean meat percentage is predicted and divided rank.
The comparison of each regression equation of table 3 estimation hog on hook lean meat percentage
aThe square root of square error is the estimation of error term standard deviation.
4, the diagnosis of regression model
With test collected respectively test the pig corresponding data respectively substitution regression equation 5,6,7 carry out regression diagnostics.Equation 5 shows, has only the standardization of 16 pigs to estimate residual error greater than 2 in 305 test pig, and the overall average error item standard deviation of test pig be estimated as 2.68943, prove that this regression model has higher accuracy; Fig. 2 illustrates that it is completely random that residual error distributes, the regression equation of determining is not violated the equation homogeneous, the incoherent hypothesis of error term is theoretical, and 87.2% testing site drops on (1.5 in the distribution plan, 1.5) between, the binary linear regression model that gained is described is reliably, and the result is comparatively accurate in its estimation; Find 22 influential testing sites by Cook D check, explanation may also exist can better these data of match model, also may be unusual observation station, may need to collect the relation of more data to confirm to be provided by influential observation; The collinearity diagnosis shows that variance inflation factor (VIF) illustrates there is not collinearity between the independent variable that the gained model is stable less than 10.
Equation 6 shows, has only the standardization of 15 pigs to estimate residual error greater than 2 in 305 test pig, and the overall average error item standard deviation of test pig be estimated as 2.36706, prove that this regression model has higher accuracy; Fig. 3 illustrates that it is completely random that residual error distributes, the regression equation of determining is not violated the equation homogeneous, the incoherent hypothesis of error term is theoretical, and 86.2% testing site drops on (1.5 in the distribution plan, 1.5) between, the binary linear regression model that gained is described is reliably, and the result is comparatively accurate in its estimation; Find 17 influential testing sites by CookD check, explanation may also exist can better these data of match model, also may be unusual observation station, may need to collect the relation of more data to confirm to be provided by influential observation; The collinearity diagnosis shows that variance inflation factor (VIF) illustrates there is not collinearity between the independent variable that the gained model is stable less than 10.
Equation 7 shows, has only the standardization of 13 pigs to estimate residual error greater than 2 in 305 test pig, and the overall average error item standard deviation of test pig be estimated as 2.27971, prove that this regression model has higher accuracy; It is completely random that Fig. 2-4 explanation residual error distributes, the regression equation of determining is not violated the equation homogeneous, the incoherent hypothesis of error term is theoretical, and 88.5% testing site drops on (1.5 in the distribution plan, 1.5) between, the multiple linear regression model that gained is described is reliably, and the result is comparatively accurate in its estimation; Find 19 influential testing sites by Cook D check, explanation may also exist can better these data of match model, also may be unusual observation station, may need to collect the relation of more data to confirm to be provided by influential observation; The collinearity diagnosis shows that variance inflation factor (VIF) illustrates there is not collinearity between the independent variable that the model of gained is stable less than 10.
5, the accuracy of regression model check
In order to verify that can above-mentioned 3 regression models of gained accurately predict the lean meat percentage of Jiangsu and surrounding area market pig, measure and cut apart at production and processing base, Huaian picked at random 110 bull hogs on hook.Table 4 is basic statistics amounts of measured value and estimated value (estimated value of equation 5,6,7).
Fig. 5,6,7 has reflected the design sketch of the predicted value of three models to measured value more intuitively.
The basic statistics amount of table 4 measured value and estimated value
Claims (1)
1, a kind of method that is used for pork carcass quality prediction is characterized in that, the lean meat percentage predictive equation is y=66.914+0.0682 x
1-0.6789 x
9Or y=57.742-0.5871 x
7+ 0.2023 x
10Or y=56.806-0.1054x
1-0.5316 x
7+ 0.3029 x
10Wherein y is lean meat percentage (%), x
1Be hot carcass weight (kg); x
7Be that waist is recommended fat thickness (mm); x
9Be F
b(mm); x
10Be M
b(mm).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008100193507A CN101477101A (en) | 2008-01-04 | 2008-01-04 | Method for pork carcass quality prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008100193507A CN101477101A (en) | 2008-01-04 | 2008-01-04 | Method for pork carcass quality prediction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101477101A true CN101477101A (en) | 2009-07-08 |
Family
ID=40837846
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2008100193507A Pending CN101477101A (en) | 2008-01-04 | 2008-01-04 | Method for pork carcass quality prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101477101A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103344567A (en) * | 2013-06-25 | 2013-10-09 | 中国农业大学 | Raw fresh meat non-destructive inspection device |
| CN104101692A (en) * | 2014-07-04 | 2014-10-15 | 中国农业科学院农产品加工研究所 | Identification method for pork of black pigs and landraces |
-
2008
- 2008-01-04 CN CNA2008100193507A patent/CN101477101A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103344567A (en) * | 2013-06-25 | 2013-10-09 | 中国农业大学 | Raw fresh meat non-destructive inspection device |
| CN104101692A (en) * | 2014-07-04 | 2014-10-15 | 中国农业科学院农产品加工研究所 | Identification method for pork of black pigs and landraces |
| CN104101692B (en) * | 2014-07-04 | 2016-06-01 | 中国农业科学院农产品加工研究所 | The discrimination method of a kind of aardvark and landrace meat |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Craigie et al. | A review of the development and use of video image analysis (VIA) for beef carcass evaluation as an alternative to the current EUROP system and other subjective systems | |
| Doeschl-Wilson et al. | The relationship between body dimensions of living pigs and their carcass composition | |
| Gispert et al. | Relationships between carcass quality parameters and genetic types | |
| Boligon et al. | Principal component analysis of breeding values for growth and reproductive traits and genetic association with adult size in beef cattle | |
| De Paula et al. | Predicting carcass and body fat composition using biometric measurements of grazing beef cattle | |
| Leeds et al. | B-mode, real-time ultrasound for estimating carcass measures in live sheep: Accuracy of ultrasound measures and their relationships with carcass yield and value | |
| Newman et al. | Purebred-crossbred performance and genetic evaluation of postweaning growth and carcass traits in Bos indicus× Bos taurus crosses in Australia | |
| Font-i-Furnols et al. | Comparison of national ZP equations for lean meat percentage assessment in SEUROP pig classification | |
| CN104616204B (en) | A kind of polynary, fine, intelligent trunk meat stage division on automatic pig slaughtering line | |
| Janiszewski et al. | Prediction of primal cuts by using an automatic ultrasonic device as a new method for estimating a pig-carcass slaughter and commercial value | |
| Lisiak et al. | A new simple method for estimating the pork carcass mass of primal cuts and lean meat content of the carcass | |
| Tyra et al. | Analysis of the Possibility of Improving the Indicators of Pork Quality Through Selection with Particular Consideration of Intramuscular Fat (MF) Content/Analiza możliwości poprawy wskaźników jakości wieprzowiny na drodze selekcji ze szczególnym uwzględnieniem zawartości tłuszczu śródmięśniowego (IMF) | |
| Aass et al. | Ultrasound prediction of intramuscular fat content in lean cattle | |
| Marimuthu et al. | Prediction of lamb carcase C-site fat depth and GR tissue depth using a non-invasive portable microwave system | |
| Knecht et al. | Accuracy of estimating the technological and economic value of pig carcass primal cuts with an optical-needle device | |
| CN101477101A (en) | Method for pork carcass quality prediction | |
| Aass et al. | Prediction of intramuscular fat by ultrasound in lean cattle | |
| Savescu et al. | Multivariate regression analysis applied to the calibration of equipment used in pig meat classification in Romania | |
| Hicks et al. | Biases associated with genotype and sex in prediction of fat-free lean mass and carcass value in hogs | |
| Navajas et al. | Assessing beef carcass tissue weights using computed tomography spirals of primal cuts | |
| Gonzàlez et al. | Evaluation of the sustainability of contrasted pig farming systems: development of a market conformity tool for pork products based on technological quality traits | |
| CN101361677A (en) | Estimation method of lean meat percentage of live pig | |
| CN102435574B (en) | Nondestructive grading method for lamb carcass output | |
| Jansons et al. | Development of new pig carcasses classification formulas and changes in the lean meat content in Latvian pig population. | |
| Knecht et al. | Variability of fresh pork belly quality evaluation results depends on measurement locations |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20090708 |