Applied research note: Predicting carcass portion weights for purebred turkey (Meleagris gallopavo) lines using a 2D imaging system
Summary: Improving carcass portion yields (e.g., breast meat) is a major goal of modern turkey breeding and traditionally requires manual collection of portion weights. This can be a labor-intensive process considering the large amount of data needed to be useful for breeding companies. Recently, th...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-03-01
|
| Series: | Journal of Applied Poultry Research |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1056617124001041 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Summary: Improving carcass portion yields (e.g., breast meat) is a major goal of modern turkey breeding and traditionally requires manual collection of portion weights. This can be a labor-intensive process considering the large amount of data needed to be useful for breeding companies. Recently, there has been increasing interest in using computer vision systems to assess parameters such as size, weight, volume, and grade of poultry meat. The present study developed mathematical equations to predict turkeys’ (4,000) meat yield using a non-invasive real-time 2D carcass imaging system. Although our breast meat models proved to be good, the thigh and drum models did not demonstrate a high correlation between observed and predicted weights probably due to the orientation of the image and any potential shifts made during image capture. These results represent a first step in developing prediction models for valuable turkey carcass portions using practical imaging systems. Further investigations need to take place to demonstrate this system can be more fruitful than simply predicting portion weight off live weight and help the industry to better collect phenotypes in a cost-effective manner. |
|---|---|
| ISSN: | 1056-6171 |