3D Carcass Imaging System with Multi-View Occlusion Correction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for determining the volume and slaughter yield of carcasses are inaccurate due to occlusions and sparse representation of inner surfaces, leading to poor quality 3D models, especially when carcasses are cut open or partially visible, which affects grading and cutting machine settings.

Innovation Solution

A multi-view imaging system with cameras on both sides of a passageway captures images simultaneously to generate a 3D carcass model, identifying occluded regions and manipulating the model based on weight and suspension position data to improve accuracy, ensuring accurate representation of volume and weight.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If a single camera or limited camera system is used to capture carcass images, then the device complexity is reduced, but the measurement precision and completeness of 3D model data deteriorate due to occlusions and sparse representation of inner surfaces

Engineering Contradiction:
Improvecamera system configurationVSAvoid3D model accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The imaging system is segmented into multiple camera units positioned at different locations (first camera at first location, second camera at second location, etc.). Each camera captures images of different portions of the carcass, including inner surfaces that would be occluded from a single viewpoint. This segmentation of the imaging function across multiple spatial positions resolves the contradiction by achieving complete surface coverage without requiring a single complex camera system.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The solution transitions from a single-point observation to multi-point spatial observation by positioning cameras at multiple locations around the passageway. This dimensional expansion in the spatial arrangement of cameras enables capture of previously occluded surfaces, improving measurement precision while maintaining relatively simple individual camera configurations.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If multiple cameras are used to capture images from different positions, then the measurement precision and coverage of carcass surfaces improve, but the device complexity and synchronization requirements increase

Engineering Contradiction:
Improvecompleteness of surface representationVSAvoidmulti-camera coordination
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system employs periodic or sequential image capture where cameras are triggered at specific intervals or sequences. The control unit coordinates the timing of each camera to capture images at appropriate moments as the carcass moves through the passageway, ensuring that all cameras capture their respective views during the same operational cycle without requiring continuous complex synchronization.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The moving carcass itself serves as the timing mechanism for image capture. As the carcass passes through the passageway, each camera automatically captures images when the carcass is in the appropriate position relative to that camera, eliminating the need for complex external synchronization systems. The system leverages the natural motion and positioning of the object being measured.

Inventive Principle:
Principle #25Self-service

3Device complexity

If images are captured at different time points, then the device complexity is reduced, but the reliability of the 3D model deteriorates due to carcass movement between captures

Engineering Contradiction:
Improvesynchronization mechanismVSAvoidconsistency of 3D model
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The system performs preliminary coordination of all camera triggers before actual image capture begins. The control unit pre-synchronizes the timing signals to all cameras based on the known motion characteristics of the carcass, ensuring that when capture begins, all cameras are ready to record simultaneously. This preliminary timing arrangement ensures temporal consistency without requiring complex real-time synchronization during the actual capture process.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3595453B13D imaging system and method of imaging carcasses
Publication Date: 2021.08.25 FRONTMATEC SMOERUM AS
  • EP3595453B1 patent drawingFigure 1~2
  • EP3595453B1 patent drawingFigure 3~5
  • EP3595453B1 patent drawingFigure 6

AI summary

A method and a system of imaging carcasses, comprising: acquiring, using a first set of cameras (101), a first set of multi-view images from respective multiple positions at a first side of a passageway (115), and acquiring, using a second set of cameras (102), a second set of multi-view images from respective multiple positions at a second side of the passageway (115); wherein the multi-view images are acquired at a first point in time upon detection of a carcass being present at a predetermined position at the passageway; computing a 3D carcass model of the carcass or a portion thereof from the first set of multi-view images and the second set of multi-view images; wherein the 3D carcass model is based on one or both of a 3D point cloud and a polygon surface; locating a first region of the 3D carcass model by locating an occluded or spatially sparsely represented region of the 3D carcass model; and manipulating the 3D carcass model at least at the first region to change the volume of the 3D carcass model or the distribution of volume of the 3D carcass model in accordance with one or both of a weight registered by a load cell and a representation of a suspension position at which the carcass is suspended.