What question did this study set out to answer?

This research aims to develop an efficient framework for monitoring fish feeding states using video data.

February 28, 2026AnimalsOpen Access

Edge-Deployable Fish Feeding-State Quantification and Recognition via Frame-Pair Motion Encoding and EfficientFeedingNet

View Full Paper

Ask AI

Bookmark

Authors

Framework quantifies feeding states in fish using video analysis, suggesting efficient management strategies for aquaculture.

This research aims to develop an efficient framework for monitoring fish feeding states using video data.
Proposed a motion-driven framework for feeding-state quantification.
Utilized frame-pair dense optical-flow encoding.
Integrated a lightweight network called EfficientFeedingNet.
Constructed a perception-based dataset with reproducible binary labels.
Models trained on the Perceptual Dataset achieved >90% test accuracy.
EfficientFeedingNet attained 96.53% test accuracy.
Framework runs at 143.24 frames per second on Jetson Orin NX.

Xiao et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

February 28, 2026AnimalsOpen Access

View Full Paper

Ask AI

Bookmark

Authors

Framework quantifies feeding states in fish using video analysis, suggesting efficient management strategies for aquaculture.

This research aims to develop an efficient framework for monitoring fish feeding states using video data.
Proposed a motion-driven framework for feeding-state quantification.
Utilized frame-pair dense optical-flow encoding.
Integrated a lightweight network called EfficientFeedingNet.
Constructed a perception-based dataset with reproducible binary labels.
Models trained on the Perceptual Dataset achieved >90% test accuracy.
EfficientFeedingNet attained 96.53% test accuracy.
Framework runs at 143.24 frames per second on Jetson Orin NX.

Xiao et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark