Camera-Based Shrimp Weight Scale for Automated Aquaculture Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the weight of aquatic organisms in aquaculture are inaccurate and require manual intervention, leading to inefficiencies in growth control and resource management.
Innovation Solution
A scale equipped with a camera and a controller using trained organism detection and weight determination engines, which utilize supervised or unsupervised learning to automatically detect and determine the weight of organisms in images, eliminating the need for manual annotation and improving accuracy and speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing methods for determining organism weight are used, then the process can be performed, but the accuracy of weight determination is insufficient
Solution Approach 1:
The patent replaces manual mechanical weighing and annotation processes with an automated computer vision system. The scale includes a camera that captures images of organisms, and a controller with trained machine learning engines that automatically detect organisms and determine their weights, eliminating the need for manual intervention and improving both accuracy and productivity.
Solution Approach 2:
The system enables self-service weight determination by using trained machine learning engines that automatically process images and extract weight information without human assistance. The organism detection engine and weight determination engine work autonomously to identify organisms and calculate their weights based on image analysis.
2Productivity
If manual annotation methods are used, then organism weight can be determined, but the process is time-consuming and inefficient
Solution Approach 1:
The patent replaces time-consuming manual annotation with automated image processing. The controller captures images using a camera and processes them through trained machine learning engines that rapidly detect organisms and determine weights, dramatically reducing processing time and increasing productivity.
Solution Approach 2:
The system performs preliminary action by pre-training the machine learning engines with annotated data before deployment. Once trained, the engines can automatically and rapidly process new images without requiring manual annotation during operation, achieving high-speed processing while maintaining accuracy.
3Loss of information
If simple image analysis is used, then processing is fast, but hidden features in images cannot be identified
Solution Approach 1:
The patent replaces simple image analysis with advanced machine learning-based image analysis. The trained organism detection engine and weight determination engine can identify hidden features and complex patterns in images that simple algorithms cannot detect, providing more comprehensive and accurate weight determination despite increased system complexity.
Data Source
AI summary
There is described a scale for determining weight of one or more organisms contained in a sample. The scale generally has: a camera having a field of view orientable towards the sample and being configured for acquiring an image of the one or more organisms of the sample; a controller having a memory and a processor configured to perform the steps of: accessing the acquired image; using an organism detection engine being stored on the memory and being trained, detecting one or more organism representations in one or more corresponding portions of the accessed image and generating detection data concerning the one or more detected organism representations; and using an organism weight determination engine being stored on the memory and being trained, determining weight data concerning weight associated to the one or more detected organism representations based on the detection data.