AI Feed Pellet Tracking for Precise Land-Based Fish Farm Feeding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current land-based aquaculture feeding systems lack precision, leading to inefficiencies such as overfeeding or underfeeding, which results in increased costs, environmental impact, and fluctuating water quality, due to inaccurate waste feed estimation methods.
Innovation Solution
An AI-based system using a feed camera and object enhancer in an effluent pipe to capture and classify uneaten feed pellets, determining pellet trajectories and generating real-time counts to adjust feeding amounts accurately.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If automated feeding systems are implemented, then feeding precision is improved, but device complexity increases
Solution Approach 1:
The patent replaces mechanical feeding systems with an AI-based vision system that uses image processing and machine learning algorithms to detect, count, and monitor feed pellets. The system substitutes physical measurement devices with computational methods, using cameras and software to achieve precise feeding control without complex mechanical sensors or actuators.
Solution Approach 2:
The system creates a digital copy of the physical feeding environment by capturing images of the tank and feed pellets. This virtual representation is then processed through AI algorithms to extract feeding information, replacing the need for direct physical measurement and control mechanisms while maintaining high precision.
2Measurement precision
If computational resources are increased to improve precision, then measurement accuracy is improved, but cost increases
Solution Approach 1:
The patent divides the image processing task into distinct segments: background subtraction to isolate moving objects, contour detection to identify pellet boundaries, and classification to distinguish feed pellets from other objects. This segmentation allows each computational step to be optimized independently, reducing overall resource requirements while maintaining high measurement accuracy.
Solution Approach 2:
The system processes only the necessary portions of the image data required for feed monitoring, rather than analyzing the entire image in full detail. By focusing computational resources on relevant regions (where feed pellets are located) and using efficient algorithms, the system achieves accurate waste measurement without requiring excessive computational power.
Data Source
Figure 1
Figure 2A~2B
Figure 3
AI summary
Embodiments for an AI-based system and method for feed monitoring in a land-based fish farm are described where the system may include: a feed camera mounted to an effluent pipe for capturing a video feed having images of objects that traverse a field of view of the feed camera when each image is acquired; and a special-purpose computer that executes a pellet-tracking algorithm that employs a region of interest (ROI) proposal module, an ROI classification module, an ROI tracking module, and a trajectory classification module for at least counting uneaten feed pellets. In at least one embodiment, an object enhancer may be mounted to the effluent pipe opposite to and in the field of view of the feed camera to provide a background so that the objects in the acquired images have a definable perimeter.