To enhance the efficiency and accuracy of phenotypic data acquisition in sturgeon aquaculture, this study systematically investigates intelligent measurement methods and equipment for sturgeon phenotypes. By integrating 3D vision imaging, depth-brightness mapping models, and machine vision-based segmentation algorithms, a series of models for estimating individual and group sturgeon weights were developed, including models based on body length and area, 3D surface area, and volume. Based on these methods, a portable in-situ measuring device, a high-throughput intelligent weighing platform, and an AR-based wearable measurement system were developed, featuring high throughput, portability, and high precision. Laboratory and field tests demonstrated the systems' ability to perform accurate and efficient sturgeon weight and behavior monitoring. These results provide effective tools for sturgeon germplasm evaluation, farming management, and digital twin modeling, promoting the development of intelligent aquaculture technologies.
 

Sturgeon; phenotype measurement; intelligent equipment; digital twin