Auditing Beef and Carcass Quality Every day

Background

Canada’s beef industry has conducted Beef Quality Audits every few years starting in the mid-1990’s. The Audits provide very valuable information about what our industry is doing well in terms of producing quality beef carcasses (e.g. marbling, tenderness), and where it can do better in terms of avoiding some of the estimated $200 million annual cost of quality defects such as bruising, tag, liver condemnations and injection site lesions. However, doing these audits on a five-year basis means that we are often learning about problems that started quite a while ago. However, new technologies may provide the opportunity to track carcass and offal quality (and defects) on an ongoing basis and allow us to recognize and respond to emerging quality challenges and opportunities in a timely manner.  

Objectives

1. Carcass cameras: 

• Hide-on carcass camera: Use images from a hide-on camera to develop initial algorithms to recognize hide color, score tag (mud), detect horns, and potentially read brands.

• Hide-off carcass camera: Use images from a hide-off camera to detect critical bruising

2. Ribeye camera: Develop and/or incorporate equations measuring or predicting carcass/primal yield, ribeye marbling texture and blood splash, dark cutting, tenderness and yellow fat into a new Carcass Information System (CIS) Module

3. Offal cameras:

• Develop an automated liver abscess camera to classify livers as edible or inedible and share this through the CIS mobile app along with benchmarking information.
• Examine if X-ray imaging of skulls can automate dentition checks and potentially predict animal age better.

4. Quality Verification Module: test whether the new CIS Quality Verification Module can enable branded beef programs to verify specific beef quality attributes

5. Quality Management Module: use the new CIS Quality Management Module to help plants and feedlots exchange carcass and offal related information that they can discuss with their nutritionist, veterinarian and other production specialists

What they Did

The AAFC Lacombe team further developed grading and photographic imaging to automatically collect hide color, tag, horn, brand, bruise, injection site lesion, muscling, hump height, liver abscess, and age data. They added new programs evaluating marbling texture, blood splash, dark cutter, yellow fat and tenderness into the e+v ribeye camera. However, COVID19 prevented access to commercial beef processing facilities and significantly impaired progress on the implementation objectives.

1. Carcass cameras: Personnel from e+v in Germany visited a commercial beef processing facility to determine suitable locations and to capture test images. However, the camera could not be installed due to COVID-19 restrictions on travel and facility access.
2. Ribeye camera: A software system to link the ribeye camera carcass yield predictions, box scale weights in the fabrication room and reporting templates for actual versus expected yields per cut were developed. The dark cutting, tenderness and initial yellow fat detection algorithms were programmed into the ribeye camera. However, COVID19 impeded further testing and development of ribeye algorithms in commercial facilities.
3. Offal cameras: a. A mobile app to report liver evaluation data to feedlots was designed in collaboration with Elanco Knowledge Solutions. However, programming of the liver camera system was halted, as its installation in the commercial beef processing facility was prevented by COVID-19.
4. Quality Verification Module: The Quality Verification Module was demonstrated to Certified Angus Beef personnel. An initial plan to develop brand-specific algorithms was discussed, but further work was halted because the COVID19 pandemic prevented access to commercial beef facilities.
5. Quality Management Module: An iOS and Android app was created to document quality-related parameters for feedlot shipments, including the ability to attach photos of records and/or images of cattle. A web-based interface to allow packing plants to receive shipment information and an email-based shipment notification/verification system for producers and Quality Assurance personnel was completed. However, system testing and training of feedlot producers was not possible because of COVID-19.

The AAFC Lacombe team was able to investigate whether x-ray images of the skull can predict animal age at the slaughter plant (objective 3b).

What They Learned

Current iDXA equipment does not provide sufficient quality images for age identification.

A portable x-ray camera could be utilized to achieve quality images for automatic image analyses and age verification within a packing plant environment. However, a mechanism to ensure repeatable placement of the camera in relation to the skull needs to be developed.

What it Means

Accurate Convolutional Neural Networking models for age classification are possible. In a plant environment, continuous improvement of the models can be achieved as the number of images increases using an ongoing learning process.