🤖 AI Summary
To address the challenge of early disease detection in high-density flatfish (e.g., turbot and flounder) aquaculture, this paper proposes a fine-grained disease recognition framework guided by anatomical region segmentation (head, fins, body). We introduce FlatIMG, the first large-scale flatfish disease image dataset. Our method employs a multi-branch CNN detector supervised by region-level semantic segmentation, augmented with GAN-based synthesis and image harmonization techniques to generate high-fidelity pathological samples, mitigating data scarcity. Crucially, we pioneer anatomical structure–aware modular modeling, enabling precise lesion localization and classification. On FlatIMG, our approach achieves a 12% absolute improvement in detection accuracy over baseline methods. Moreover, it demonstrates strong cross-species generalizability, validated on salmonid (chum salmon) data. The framework supports real-time automated diagnosis, providing an implementable technical solution for smart aquaculture.
📝 Abstract
The flatfish is a major farmed species consumed globally in large quantities. However, due to the densely populated farming environment, flatfish are susceptible to injuries and diseases, making early disease detection crucial. Traditionally, diseases were detected through visual inspection, but observing large numbers of fish is challenging. Automated approaches based on deep learning technologies have been widely used, to address this problem, but accurate detection remains difficult due to the diversity of the fish and the lack of the fish disease dataset. In this study, augments fish disease images using generative adversarial networks and image harmonization methods. Next, disease detectors are trained separately for three body parts (head, fins, and body) to address individual diseases properly. In addition, a flatfish disease image dataset called exttt{FlatIMG} is created and verified on the dataset using the proposed methods. A flash salmon disease dataset is also tested to validate the generalizability of the proposed methods. The results achieved 12% higher performance than the baseline framework. This study is the first attempt to create a large-scale flatfish disease image dataset and propose an effective disease detection framework. Automatic disease monitoring could be achieved in farming environments based on the proposed methods and dataset.