-
Notifications
You must be signed in to change notification settings - Fork 0
Home
Welcome to the USDA-Hack-Meat-Sensor wiki!
The United States Department of Agriculture is seeking to develop and implement their own standardized meat grading system to increase honesty and reliability of meat grading.
This challenge focuses on improving the accuracy and consistency of beef grading through technology. Participants will work with a dataset comprising 3,500 ribeye images and expert-grade quality assessments. The goal is to develop image classification models with less than 1% error rate and minimal standard deviation from the mean. By participating, you'll contribute to enhancing the integrity of beef grading and potentially impact grading practices across various commodities.
Success in the AMS Beef Grading Image Classification Challenge will be assessed through several key categories including:
- overall beef grade accuracy & consistency
- grade specific accuracy & consistency
- ingenuity, simplicity, accuracy & consistency in additional variables.
Participants will be evaluated based on their performance in these categories, with awards presented to teams demonstrating excellence in one or more areas. Each path offers its own set of challenges and opportunities for recognition, encouraging diverse approaches and innovative solutions.