Artificial Intelligence Applications for Automating Fish Age

Tatyana Valentinovna Snytnikova; Marina Viktorovna Solovey; Danil Litvishenko

Tatyana Valentinovna Snytnikova KSTU https://orcid.org/0000-0003-2750-4041
Marina Viktorovna Solovey KSTU https://orcid.org/0000-0001-5784-4077
Danil Litvishenko KSTU

Keywords: Automation, Otolith, Fish Age, Neural Network, Convolutional Network, Machine Learning, Ichthyology, Aquaculture, Bio resources, Digital image processing

Abstract

Ensuring access to high-quality, protein-rich foods like fish is vital for human nutrition, making the development of fisheries and aquaculture a global priority. Fishing faces sustainability challenges due to declining aquatic resources, while aquaculture offers scalable solutions but requires optimized growth processes. Accurate fish age assessment is crucial for both sectors: in aquaculture, it informs feeding strategies, breeding selection, and harvest planning, while in fisheries, it supports sustainable quota setting by protecting juvenile populations. Traditional methods for assessing fish age - such as analyzing scales, otoliths, or fin rays - rely on counting annual growth rings, but these methods are labor-intensive and prone to human error. Digital technologies, particularly artificial intelligence and computer vision, are transforming this process. Researchers have demonstrated that neural networks can analyze otolith and scale images with over 90% accuracy, significantly outperforming manual methods. Several software programs have been developed to address this task, but many are complex to use, limiting their accessibility. To solve this problem, Kaliningrad State Technical University has developed user-friendly software that uses a hybrid convolutional-regression neural network to automate age assessment from otolith images. The model achieved 92% accuracy on a test dataset, making it a practical alternative to manual counting. Future plans include adapting the tool for local fish species, creating a mobile app for field use, and integrating it with catch reporting systems. This innovation enhances the efficiency of fisheries and aquaculture.