Smartphone transillumination classifies woody breast in chicken fillets
Original title: Smartphone-Enabled Depth-Aware Transillumination Imaging for Detection of Chicken Breast Fillet Myopathies in Chicken Fillets
On 267 fillets, a multimodal model reached about 90% test accuracy versus about 82% for 2D-only models.
Researchers describe a smartphone method that classifies woody breast severity in broiler chicken fillets using transillumination images and 3D point clouds from an iPhone 16 Pro. They evaluated 267 fillets labeled normal, mild, or severe and extracted 91 optical, geometric, and physics-informed features. Severe tissue showed about 26% greater light attenuation per unit thickness than normal tissue. A multimodal ensemble of Fusion-DGCNN, XGBoost, and physics-guided logistic regression reached about 90% test accuracy.
Unconfirmed: Accuracy and attenuation figures are the authors' reported experimental results.
- The set included 87 normal, 85 mild, and 95 severe fillets.
- Partial correlation of image features with severity was r = 0.413 after controlling for thickness.
- 2D-only models reached about 82% test accuracy.
- The multimodal ensemble reached about 90% test accuracy.
From the original article
1 min read at arxiv.orgWoody Breast (WB) myopathy is a major quality defect in broiler chickens characterized by fibrosis, increased tissue hardness, and altered structural properties that affect product quality. Conventional palpation is subjective and labor-intensive, while many non-destructive methods require specialized equipment or substantial computational resources.
By Chaitanya Pallerla · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.