Pig Weight Estimation via Camera and Scale Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current camera-based systems for pig weight estimation in pig breeding are inaccurate and resource-intensive, failing to efficiently identify pigs ready for removal due to variations in animal behavior, density, and additional criteria beyond weight, such as health and nutrition requirements.
Innovation Solution
A device comprising a detection system, like an optical camera or electronic scale, paired with a database and evaluation module that uses pig-related data to generate a rejection indicator, considering weight and additional criteria like health, nutrition, and movement, to accurately identify pigs ready for removal.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If camera-based systems are used for weight estimation, then cost is reduced compared to mechanical scales, but measurement precision deteriorates due to inaccuracies in weight estimation
Solution Approach 1:
The patent combines multiple detection methods (camera-based estimation, mechanical scale measurements, RFID identification) into a hybrid system. The camera system provides continuous monitoring and identification, while mechanical scales provide periodic accurate weight measurements. This merging allows the system to maintain low overall cost while achieving high measurement precision through data fusion from multiple sources.
Solution Approach 2:
The patent introduces an intermediary data fusion layer that processes information from both camera-based estimation and mechanical scale measurements. This intermediary component reconciles the imprecise continuous data from cameras with the accurate but intermittent data from scales, producing reliable weight estimates even when mechanical scales are not actively measuring.
2Measurement precision
If mechanical routing control systems are used for pig selection, then sorting accuracy is improved, but device complexity and maintenance requirements increase
Solution Approach 1:
The patent replaces complex mechanical routing control systems with a simplified system based on RFID identification and camera-based monitoring. Pigs are identified individually through RFID tags, and their selection for removal is determined through automated image processing and data analysis, eliminating the need for complex mechanical routing mechanisms while maintaining sorting accuracy.
Solution Approach 2:
The patent creates a virtual model of each pig through RFID identification and camera imaging, storing their characteristics in a database. This virtual copy allows the system to track and select pigs for removal based on their digital representation rather than requiring complex physical routing mechanisms, significantly reducing device complexity.
3Measurement precision
If individual animal monitoring with radio units is implemented, then measurement precision is improved, but cost and device complexity increase significantly
Solution Approach 1:
The patent makes the camera system multi-functional by using it for both identification (through image recognition of individual pigs) and weight estimation (through image processing). This eliminates the need for separate radio monitoring units, reducing device complexity while maintaining the ability to track and monitor individual animals throughout the facility.
Solution Approach 2:
The system uses the pigs' own visual characteristics (color, shape, markings) and RFID tags as natural identifiers, eliminating the need for additional monitoring equipment attached to each animal. The camera system automatically captures and processes images to identify and track individual pigs without requiring active cooperation or additional devices on the animals themselves.
4Productivity
If camera-based systems operate continuously to track all pigs, then productivity is improved, but use of energy and computational resources increases
Solution Approach 1:
The patent implements periodic activation of the camera system and mechanical scales based on triggered events rather than continuous operation. The system activates cameras and scales when pigs approach weighing stations or when removal decisions need to be made, significantly reducing energy consumption while maintaining the ability to efficiently track and select pigs for removal when needed.
Solution Approach 2:
The system performs preliminary identification and tracking of pigs using RFID tags and camera images before weighing or removal decisions are required. This preliminary action allows the system to pre-process and store pig identification data, so that when a removal decision is needed, the system can quickly retrieve and act on the information without requiring continuous high-power operation of all components.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Device for the automated marking of a pig ready for slaughter, comprising: a detection device for determining a weight value of at least one pig, a database with pig-related data, wherein the pig-related data includes information about at least one specific or identifiable pig, and an evaluation device for generating a slaughter indicator, wherein the evaluation device is configured for data exchange with the detection device and with the database, characterized in that the evaluation device generates the slaughter indicator depending on the determined weight value and depending on the pig-related data in order to automatically mark the pig as ready for slaughter with the slaughter indicator.