Machine learning applications for postharvest poultry processing: a review

Abstract

Poultry meat plays a vital role in global food security due to its affordability and high-quality protein content. Its production keeps growing worldwide. This highlights the need for poultry processing facilities to operate efficiently while maintaining high throughput and ensuring products meet regulatory standards and satisfy consumer expectations. Machine learning (ML), as a powerful tool, opens a new path for the poultry industry. It has been implemented across diverse stages of postharvest poultry meat processing. In this review, we summarize and elucidate the applications of ML within poultry meat processing, covering smart processing operations, quality control, safety monitoring, risk assessment, foreign material detection, and adulteration identification. Among them, ML-powered imaging, spectroscopic, sensors, and genomic sequencing techniques in poultry meat processing are also explored. In addition, we also analyzed the challenges and barriers of adopting ML in real-world processing and discussed the potential solutions and future research directions. By identifying the advancements and capabilities of ML, this review highlights the potential of ML in building a more sustainable, smart, resilient, and more digitalized poultry industry and food supply chain.