Related Experiment Video
Updated: Aug 4, 2026
05:56
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
Published on: April 14, 2023
2.4K
CarcassFormer: an end-to-end transformer-based framework for simultaneous localization, segmentation and
Minh Tran1, Sang Truong1, Arthur F A Fernandes2
1Department of Computer Science and Computer Engineering, 1 University of Arkansas, Fayetteville, AR 72701, USA.
Poultry Science
|June 26, 2024
Summary
This study introduces CarcassFormer, an automated system using machine learning and computer vision to assess poultry carcass quality. It accurately detects and classifies defects, improving food safety and processing efficiency.
Area of Science:
- Food Science and Technology
- Computer Science
- Agricultural Engineering
Background:
- Automated quality assessment in poultry processing is essential for food safety and efficiency.
- Current methods often rely on manual inspection, which can be subjective and labor-intensive.
- Defects can arise from various factors, including animal welfare and equipment malfunctions.
Purpose of the Study:
- To develop and evaluate an automated system for poultry carcass quality assessment.
- To introduce the CarcassFormer framework for defect detection, segmentation, and classification.
- To compare CarcassFormer's performance against state-of-the-art methods.
Main Methods:
- Utilized a Transformer-based architecture for visual representation extraction.
- Developed an end-to-end framework (CarcassFormer) for defect analysis.
- Trained and benchmarked the system on a dataset of 7,321 poultry carcass images.
Main Results:
- CarcassFormer demonstrated superior performance in classification, detection, and segmentation tasks.
- Achieved significant improvements across key evaluation metrics (AP, AP@50, AP@75).
- Qualitative results showed high precision in identifying fine details and localizing defects.
Conclusions:
- CarcassFormer offers an effective, automated solution for poultry carcass quality assessment.
- The framework accurately identifies defects, enhancing food safety and processing oversight.
- The system's public availability will foster further research in automated food quality inspection.
