Weighted Feature Fusion of CNNs Utilizing Magnitude and Phase Information for Chicken Health Detection

Recent studies have shown the potential of phase features in audio processing, providing distinct insights into signal structure beyond what magnitude features can capture. In chicken health detection, however, existing models often rely heavily on magnitude-based features, neglecting phase-related information that could enhance detection performance. This paper introduces a Weighted Feature Fusion of Convolutional Neural Networks (WFF-CNNs), which integrate magnitude features, namely Mel Frequency Cepstral Coefficient (MFCC), with phase features, specifically Modified Group Delay Cepstral Coefficient (MGDCC), to enhance the identification of chicken-health issues. By integrating phase information, the proposed WFF-CNNs address an acknowledged limitations in current single-stream approaches that predominantly utilize magnitude-only features, hence augmenting the capacity of the model for capturing acoustic signals in chicken vocalizations. Compared with conventional CNNs using only MFCC, our approach overcomes the benchmark challenge of insufficient phase-based feature representation, leading to more robust detection of chicken health states. The efficacy of the proposed method is validated using audio signals from a publicly available database. Experimental results demonstrate that CNNs using MGDCC outperform conventional CNNs using MFCC, highlighting the benefits of using phase information. Furthermore, the WFF-CNNs utilizing MFCC and MGDCC outperform single-feature CNNs and existing standard models, confirming the efficacy of integrating both magnitude and phase information for accurate chicken health assessment. We anticipate that these findings will be useful in the formulation of more complete guidelines for monitoring poultry health, considering both size and phase characteristics. Note to Practitioners—Poultry farming represents a crucial sector in global food security and economics, with significant impact on both large-scale industrial operations and small-scale farmers. Disease outbreaks in chicken populations can lead to devastating economic losses - not just from chicken mortality, but also from decreased productivity, increased treatment costs, and potential trade restrictions. For instance, a single outbreak of avian influenza can result in the loss of entire flocks and disrupt local economies, particularly affecting small-scale farmers who may lack advanced monitoring systems. Our research addresses this critical need by developing an automated health monitoring approach that analyzes chicken vocalizations to detect potential health issues before they become severe outbreaks. Using advanced sound analysis techniques that examine multiple aspects of chicken vocalizations simultaneously, our method achieves 96.92% accuracy in health detection, surpassing traditional analysis methods. This analysis approach has the potential to be developed into an automated detection system suitable for small and medium-sized farms, requiring only basic microphones installed in chicken coops or farming areas, making it a cost-effective and easily implementable solution. However, system performance may be affected by environmental noise in real-world conditions. Future developments will focus on improving robustness across different farm environments and testing with various chicken breeds to validate real-world effectiveness. This development could enhance sustainable healthcare standards in poultry farming.