How AI Tracks Fish in 3D Underwater! 🐟 #Sciencefather #researchawards #artificialintelligence

Indoor recirculating aquaculture systems (RAS) are advanced setups designed to improve aquaculture productivity by integrating components such as water circulation, filtration, oxygen supply, and microbial filters. These systems support high-density fish farming while maintaining water quality. Given their complexity, automated monitoring technologies like target detection and tracking are essential for observing fish behavior. Behavior such as reduced swimming or surface gathering can indicate stress, illness, or environmental issues like low dissolved oxygen, highlighting the need for continuous monitoring.

Among monitoring techniques, 3D target tracking stands out for its ability to accurately capture fish movements and behavior in three-dimensional space. This enables more detailed behavioral metrics such as swimming speed, spatial distribution, and depth. While 2D tracking is limited by the lack of depth data and is commonly used for animals on flat surfaces, 3D tracking is more suitable for fish that swim freely in all directions. Of the available 3D tracking systems, underwater parallel stereo vision offers the most promise for aquaculture due to its cost-effectiveness, single imaging medium, and accurate depth perception without the complications of air-water refraction.

To address the limitations of current 3D tracking methods—such as high computational costs and accuracy loss in noisy underwater environments—a two-stage 3D multi-fish tracking (TMT) model has been proposed. In the first stage, it uses YOLOv8x and DeepSORT to extract fish patches from stereo images. In the second stage, it applies patch-based stereo matching, improved Semi-global Matching (SGM), and point cloud filtering to calculate 3D positions. By focusing only on fish-containing patches, the TMT model improves tracking accuracy, reduces computational load, and streamlines the 3D fish behavior monitoring process in RAS environments.

International Conference on Computer Vision

The International Research Awards on Computer Vision recognize groundbreaking contributions in the field of computer vision, honoring researchers, scientists and innovators whose work has significantly advanced the domain. This prestigious award highlights excellence in fundamental theories, novel algorithms and real-world applications, fostering progress in artificial intelligence, image processing and deep learning.

Visit Our Website : computer.scifat.com 

Nominate now : https://computer-vision-conferences.scifat.com/award-nomination/?ecategory=Awards&rcategory=Awardee 

Contact us : computersupport@scifat.com 

#researchawards #shorts #technology #researchers #conference #awards #professors #teachers #lecturers #biologybiologiest #OpenCV #ArtificialIntelligence #MachineLearning #DeepLearning #NeuralNetworks #DataScience #physicist #coordinator #business #genetics #medicirne #bestreseracher #bestpape 

Get Connected Here: 

================== 

Twitter :   x.com/sarkar23498

Youtube : youtube.com/channel/UCUytaCzHX00QdGbrFvHv8zA

Pinterest : pinterest.com/computervision69/

Instagram : instagram.com/saisha.leo/?next=%2F

Tumblr : tumblr.com/blog/computer-vision-research