Multi-Angle Underwater Camera System for Fish Weight Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current fish farming technologies lack a completely automated, efficient, and maintenance-friendly solution for simultaneously measuring fish biomass and controlling feeding in netpens, as existing systems are either ineffective, manual, or prone to errors due to limitations in image capture and processing.
Innovation Solution
A submerged camera system with multiple underwater cameras, processing images from different angles to identify fish, pellets, and faeces based on shape, size, contrast, and color, calculating fish volume and weight, and controlling feed distribution automatically.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single camera is used to measure fish weight, then the device complexity is reduced, but the measurement precision deteriorates due to inability to capture fish from multiple angles
Solution Approach 1:
The patent combines multiple cameras (at least two cameras) into a single integrated measurement system with a common processing unit. This merging approach maintains measurement precision by capturing fish from multiple angles simultaneously, while reducing device complexity compared to having separate measurement systems for each camera angle.
Solution Approach 2:
The patent transitions from single-angle (2D) camera measurement to multi-angle (3D) measurement by positioning cameras at different angles around the fish. This dimensional change enables accurate volume calculation and weight determination without requiring complex mechanical manipulation of the fish or measurement apparatus.
2Productivity
If manual image capture and analysis is used, then the equipment cost is reduced, but the productivity deteriorates due to time-consuming manual processes
Solution Approach 1:
The system implements automated image capture and processing where the cameras continuously capture fish images and the processing unit automatically analyzes them to determine weight. This self-service automation eliminates manual intervention, significantly improving productivity while the integrated design keeps the automation complexity manageable.
Solution Approach 2:
The patent replaces manual mechanical image analysis with automated computer vision processing. The processing unit uses algorithmic image analysis instead of human operators, substituting mechanical/manual processes with automated electronic processing to enhance measurement speed and productivity.
3Measurement precision
If fish are forced through a transfer channel for measurement, then the measurement precision is improved, but the fish welfare deteriorates causing stress and altered behavior
Solution Approach 1:
Instead of forcing fish through a confined transfer channel (traditional approach), the patent inverts the approach by bringing the measurement system (cameras) to where the fish naturally swim in the netpen. This reversal eliminates the need for fish manipulation and transfer, maintaining both measurement precision and fish welfare.
Solution Approach 2:
The patent introduces an intermediary measurement approach where cameras capture images of fish in their natural environment without physical contact or confinement. This intermediary optical measurement method achieves precise volume and weight measurement without the harmful effects of forced transfer channels on fish stress and behavior.
4Device complexity
If pellet detection is performed near the bottom of the netpen, then the pellet counting is simplified, but the measurement reliability deteriorates due to inability to distinguish pellets from faeces
Solution Approach 1:
The patent applies local quality by positioning the camera specifically in the feeding zone where feed is distributed, rather than uniformly throughout or at the bottom of the netpen. This localized detection approach ensures that the camera captures images of both pellets and faeces in the same field of view, enabling reliable differentiation and accurate feed consumption measurement while maintaining system simplicity.
5Reliability
If a cage of net is used to encapsulate the sensor, then the measurement reliability is improved by reducing fish interference, but the ease of operation deteriorates due to increased maintenance and cleaning requirements
Solution Approach 1:
The patent extracts the camera sensor from the problematic cage-of-net enclosure and positions it freely in the feeding zone. This extraction eliminates the maintenance and cleaning issues associated with net cages while maintaining measurement reliability through proper camera positioning and automated image analysis that can distinguish pellets from faeces without physical barriers.
Data Source
Figure 1a
Figure 1b
Figure 2~3a
AI summary
A method and system for registration of predominately freely moving objects in a fish farm netpen are described, in which a camera housing (2) with a number of underwater cameras, is submerged in the netpen suspended from a cable (20). Each camera (1a, 1 b, 1c) takes individual images, at respective angles, of objects that freely pass the camera housing and sends said images to a computer processing unit (5) and the computer processing unit (5) analyses the images taken at different angles to determine the main pattern of the objects, by grouping each pattern element based on relations between shape, size, contrast and/or colour, and to determine mutual distance between given pattern elements for identification of the main pattern and determine, based on the main pattern, whether the objects are fish (10), pellets (11), faeces (12) or other unknown elements.