2006 Fiscal Year Final Research Report Summary
Prediction of mono unsaturated fatty acid composition by computer image analysis method using fineness of marbling particles
| Project/Area Number |
17580230
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Zootechnical science/Grassland science
|
|---|
| Research Institution | Obihiro University of Agriculture and Veterinary Medicine |
|---|
Principal Investigator |
KUCHIDA Keigo Obihiro University of Agriculture and Veterinary Medicine, Graduate School of Animal and Food Hygiene, Professor, 大学院畜産学研究科, 教授 (50271747)
|
|---|
| Project Period (FY) |
2005 – 2006
|
| Keywords | beef cattle / image analysis / fatty acid / melting point / MUFA / 皮下脂肪 / 融点 |
|---|
| Research Abstract |
Mirror type digital camera for beef carcass cross section was developed by us collaborating with Hokkaido Prefectural Industrial Institute etc. in 2004. This equipment leads to take extreme high resolution digital image (4500x3000 pixel) of carcass cross section. The targets of our project were to evaluate marbling feature in detail, to investigate the relationship between marbling characteristics and chemical component of meat and to introduce objective meat evaluation method by image analysis to beef cattle breeding scheme.
1. The area of each marbling particle was ranked in descending order. The sum of the top five particle areas divided by the total area of marbling was used as a good indicator of marbling coarseness (Coarseness of 1〜5 particles). Least square means of the marbling area percentages of Japanese Black, Japanese Black×Japanese Brown, Japanese Brown, Japanese Black×Holstein (JBH), Holstein were 43.02%, 36.62%, 29.72%, 31.76% and 19.34%, respectively. These values differ
… More
significantly, except for those corresponding to Japanese Brown and JBH. Coarseness of 1〜5 particles of JBH (19.88) and Holstein (20.14) were significantly higher than those of other breeds (P<0.05).
2. A multiple regression analysis was performed by the REG procedure in SAS to predict the BCS number. BCS numbers evaluated by graders were used as a dependent variable, and traits obtained by the image analysis were used as independent variables which were limited to 5. Images with adjusted luminance and extracted along the outline of the rib eye had the highest determination coefficient (R^2=0.783). Ninety five point seven percent of the BCS numbers evaluated by graders were correctly predicted using BCS numbers by the image analysis, suggesting that it is possible to predict BCS numbers with high accuracy.
3. Melting point of Japanese Black steers were predicted using information from high resolution digital image of beef carcass cross section. Average of melting point was 28.22±3.24℃(19.40〜34.95). Correlation coefficients of melting point with carcass weight (0.08), BMS number (0.05), BCS number (0.18). There was no high correlation coefficient of melting point with carcass and image analysis traits. R-square value of multiple regression to predict melting point using all data including 8 investigation dates was 0.13. R-square values ranged from 0.60 to 0.86 when the multiple regression analyses were done by investigation day. Less
|
Research Products
(18 results)
