# High-Accuracy Recognition Method for Diseased Chicken Feces Based on Image and Text Information Fusion

## TL;DR

This paper introduces a new method that combines images and text to more accurately detect sick chickens by analyzing their feces, improving reliability and efficiency in farming.

## Contribution

The novel MMCD model uses multimodal fusion with MASA, DSconv, and GCA modules to enhance diagnostic accuracy and reduce computational costs.

## Key findings

- MMCD outperforms single-modal methods with significant improvements in Accuracy (+8.69%), Recall, Precision, and F1 score.
- The model reduces parameters by 7.5M and computations by 1.62 GFLOPs compared to ResNet50.
- Multimodal fusion proves effective for pathological fecal detection in complex farm environments.

## Abstract

The health of chickens can be judged by observing their feces. Traditional methods rely on manual checks or image-only analysis, which can be inaccurate due to lighting or image quality issues. In this study, we propose a new method that combines images with text descriptions to better identify whether a chicken is sick. The system can both “see” the image and “understand” the written description, making the judgment more complete. Test results show that this approach is more reliable than using images alone, and it performs well even in complex farm environments. It helps farmers detect problems earlier, reduce medication use, and improve overall farming efficiency.

Poultry feces, a critical biomarker for health assessment, requires timely and accurate pathological identification for food safety. Conventional visual-only methods face limitations due to environmental sensitivity and high visual similarity among feces from different diseases. To address this, we propose MMCD (Multimodal Chicken-feces Diagnosis), a ResNet50-based multimodal fusion model leveraging semantic complementarity between images and descriptive text to enhance diagnostic precision. Key innovations include the following: (1) Integrating MASA(Manhattan self-attention)and DSconv (Depthwise Separable convolution) into the backbone network to mitigate feature confusion. (2) Utilizing a pre-trained BERT to extract textual semantic features, reducing annotation dependency and cost. (3) Designing a lightweight Gated Cross-Attention (GCA) module for dynamic multimodal fusion, achieving a 41% parameter reduction versus cross-modal transformers. Experiments demonstrate that MMCD significantly outperforms single-modal baselines in Accuracy (+8.69%), Recall (+8.72%), Precision (+8.67%), and F1 score (+8.72%). It surpasses simple feature concatenation by 2.51–2.82% and reduces parameters by 7.5M and computations by 1.62 GFLOPs versus the base ResNet50. This work validates multimodal fusion’s efficacy in pathological fecal detection, providing a theoretical and technical foundation for agricultural health monitoring systems.