Revolutionizing Fish Farming with Computer Vision: A Survey of Techniques and Trends

Abstract

The aquaculture industry has emerged as a critical contributor to global food security, necessitating innovative technologies to enhance productivity, sustainability, and fish welfare. Computer vision, powered by advances in artificial intelligence and deep learning, is revolutionizing fish farming by enabling automated monitoring, analysis, and decision-making. This survey provides a comprehensive overview of computer vision techniques and trends applied to modern aquaculture. Key applications include fish detection and counting, species classification, growth estimation, behavior analysis, and disease detection, all of which reduce manual intervention and improve operational efficiency. The paper highlights commonly used models such as convolutional neural networks (CNNs), YOLO-based object detection frameworks, and 3D imaging techniques for underwater environments. Furthermore, emerging trends such as multimodal sensing, edge computing, real-time monitoring, and the integration of Internet of Things (IoT) devices are discussed. Challenges including underwater visibility, occlusion, limited annotated datasets, and system scalability are critically analyzed. The survey concludes by identifying future directions in precision aquaculture, emphasizing the potential of computer vision to drive sustainable fish farming practices and meet the growing global demand for aquatic products.