Vision-Based Poultry Counting on Conveyor Belts for Cage Fill Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing poultry caging systems rely on indirect weight-based methods to determine the number of animals in each cage, leading to inaccuracies and difficulties in meeting specific animal count requirements for slaughter and compliance checks, as well as payment discrepancies for farmers and operators.
Innovation Solution
A system and method using an image-based counting approach with a neural network model to identify and track poultry on a conveyor belt, calculating the exact number of birds entering cages by processing a sequence of images, ensuring accurate counting and control of the caging process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If weight-based indirect determination is used to count poultry in cages, then the system is simple to operate, but the counting precision is poor due to animals bunched together and weight variations
Solution Approach 1:
The patent replaces the mechanical weighing system with an optical vision system. Instead of using weight-based determination, the system uses image acquisition devices (cameras) to capture images of poultry on the conveyor belt, processes these images through neural networks to identify and count individual animals, and determines when the cage is filled based on visual detection rather than weight measurement. This substitution resolves the contradiction by providing precise individual animal counting without the limitations of weight-based methods.
2Reliability
If weight-based counting is used, then the device complexity is low, but the reliability is poor for meeting specific animal count requirements
Solution Approach 1:
The system replaces the simple but unreliable weight-based mechanism with a sophisticated vision-based counting system that provides reliable individual animal identification. The neural network processing enables accurate counting of each poultry animal, ensuring compliance with specific animal count requirements for slaughter and regulatory standards, while the automated image processing and counting logic provide consistent and reliable operation.
Solution Approach 2:
The system implements feedback control by continuously monitoring the count of poultry animals in the cage through image processing and comparing it against the target number. The system adjusts the caging process in real-time, stopping the conveyor belt when the correct number of animals is detected, ensuring reliable compliance with required animal counts.
3Productivity
If indirect weight-based counting is used, then the operation is simple, but the productivity is reduced due to delays in determining filled state
Solution Approach 1:
The patent replaces the delayed weight-based determination system with an immediate visual detection system. Image acquisition devices continuously capture images of poultry on the conveyor belt, and the neural network processes these images in real-time to immediately identify when the cage is filled. This eliminates the delay inherent in weight-based systems, allowing the conveyor belt to stop precisely when the correct number of animals is present, thereby improving caging speed and productivity.
Solution Approach 2:
The system performs preliminary identification and tracking of poultry animals on the conveyor belt before they enter the cage. By pre-identifying individual animals and tracking their movement, the system can determine the filled state immediately when the target number is reached, without waiting for weight accumulation, thus improving productivity.
Data Source
AI summary
A description is provided of a counting system for counting poultry in a caging machine configured to convey poultry on a conveyor belt in order to introduce them into a cage. The system comprises:an acquisition group for acquiring a sequence of images of a same portion of the conveyor belt during successive time instants;a processing apparatus comprising:an identification unit for processing the sequence of images in order to identify the presence of the poultry in the each image;a tracking unit for calculating poultry tracks based on the identified poultry presence, each poultry track indicating the path followed by a respective bird along the conveyor belt;a counting unit configured to calculate a count indicating a number of birds that have reached the cage on the basis of the calculated tracks;an output unit for outputting the calculated count for control of the caging machine.