Feed Bunk Segmentation for Reachable Feed Volume Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing feed bunk management systems fail to provide accurate and timely information on feed availability, leading to uneven distribution and consumption, which can result in sub-optimal animal feeding, health issues, and inaccurate intake measurements due to manual and error-prone refusal recording.
Innovation Solution
An automated feed bunk image analysis method that segments the feed bunk into reachable and unreachable areas, providing real-time alerts and adjustments to redistribute feed, and calculates accurate intake by distinguishing between consumed and unreachable feed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual refusal recording is used, then operation simplicity is maintained, but measurement precision of intake deteriorates
Solution Approach 1:
The patent replaces manual mechanical recording of refusal with an automated image analysis system using cameras and machine learning algorithms to detect and measure feed refusal, thereby improving measurement precision while accepting increased system complexity
Solution Approach 2:
The system creates visual copies (images) of the feed bunk and refusal areas, then analyzes these copies through machine learning to determine intake measurements, eliminating the need for direct physical measurement while improving accuracy
2Ease of operation
If feed is allowed to accumulate in unreachable areas, then loss of substance is reduced, but ease of operation deteriorates
Solution Approach 1:
The system continuously monitors feed levels and refusal in both reachable and unreachable areas, providing feedback that triggers automated redistribution operations when refusal exceeds thresholds, balancing operational ease with reduction of feed waste
Solution Approach 2:
The system performs preliminary detection of refusal accumulation in unreachable areas and initiates redistribution before significant waste occurs, maintaining ease of operation by acting only when necessary while preventing substance loss
3Productivity
If real-time feed monitoring is implemented, then productivity is improved, but device complexity increases
Solution Approach 1:
The system enables automated self-monitoring of feed levels and refusal through image capture and machine learning analysis, improving productivity by eliminating manual monitoring while accepting the complexity of the automated system
Solution Approach 2:
The image analysis system performs multiple functions including detecting refusal, measuring feed levels, identifying unreachable areas, and triggering redistribution operations, improving productivity through multi-functionality while managing overall system complexity
Data Source
AI summary
Disclosed are embodiments for segmenting an animal feed bunk. The feed bunk is segmented, in some embodiments, into segments that include reachable animal feed, and other segments that include unreachable animal feed. Volume estimates of feed in each of the segments are generated, in some embodiments, via imaging data of the feed bunk captured by a LIDAR sensor or passive optical sensor. The volume estimates trigger alerts in some instances indicating feed shortages or a need for a push up operation to move feed from an unreachable segment to a reachable segment. Demarcation of feeding events, coupled with volume estimations of feed remaining provide for determination of refusal volumes, which are useful in determining animal appetite.