Contamination detection in chicken carcasses using computer vision and RGB images

• Grant number: 24/09497-0
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: September 01, 2024
• End date: August 31, 2025
• Field of knowledge: Agronomical Sciences - Animal Husbandry - Animal Production
• Principal Investigator: Murilo Mesquita Baesso
• Grantee: Bianca Martim Tinti
• Host Institution: Faculdade de Zootecnia e Engenharia de Alimentos (FZEA). Universidade de São Paulo (USP). Pirassununga , SP, Brazil

Abstract

Brazil, one of the largest producers and exporters of chicken meat, faces challenges in ensuring the quality and safety of its chicken carcasses. To improve efficiency and accuracy in contamination detection, this project aims to develop an automated system for detecting contamination in chicken carcasses using computer vision, aiming to increase the accuracy and efficiency of inspections in the poultry industry. Specifically, the project seeks to capture and process RGB images of carcasses to create a comprehensive database, apply image processing techniques such as histogram equalization and median and sharpening filters to identify and isolate contamination spots, and implement computer vision algorithms (SVM, KNN and Random Forest) to classify regions of interest in images. The methodology involves capturing images, followed by defining regions of interest in the images to focus on areas of contamination. RGB images will be processed to separate color components, and principal component analysis (PCA) will be applied to reduce the dimensionality of the data, facilitating pattern detection. Image processing techniques will be used to improve contrast and reduce noise in images. The results will be analyzed using computer vision algorithms. The performance of the algorithms will be evaluated based on accuracy, sensitivity, specificity and F1-score. The system is expected to increase food safety, reduce inspection time and operational costs, providing a practical and efficient solution for detecting contamination in chicken carcasses.