RGB-D Herd Animal Mass Prediction in Semi-Open Environments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining animal body mass in large herds are laborious, invasive, and imprecise due to environmental uncertainties and complex interactions, especially in semi-open and poorly structured breeding environments, limiting their application and increasing uncertainty.
Innovation Solution
A method using RGB-D imaging and supervised machine learning to predict body mass by capturing top-view videos, extracting geometric characteristics, and applying filtering techniques to separate animals from their environment, followed by computer modeling to build robust prediction algorithms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If visual inspection or weighing with scales is used to determine body mass, then measurement can be obtained, but the process becomes laborious, invasive, and stressful for animals requiring separation and individual containment
Solution Approach 1:
The patent replaces mechanical weighing systems with an optical sensing system using depth cameras and computer vision algorithms. The system captures depth images of animals and uses image processing to estimate body mass without physical contact, eliminating the need for mechanical scales and manual handling while maintaining measurement capability
Solution Approach 2:
The patent creates a digital copy of the animal's physical form through depth imaging and 3D reconstruction. By generating a virtual model from optical data, the system estimates body mass from the digital representation rather than requiring physical measurement, enabling non-contact assessment of multiple animals simultaneously
2Ease of operation
If conventional computer vision is used in semi-open breeding environments, then body mass measurement can be attempted, but environmental uncertainties lead to high degrees of measurement uncertainty
Solution Approach 1:
The patent transitions from 2D RGB imaging to 3D depth imaging, adding the depth dimension to capture spatial information. This dimensional enhancement allows the system to measure body volume and estimate mass more accurately by incorporating three-dimensional geometric data that is invariant to lighting and environmental conditions
Solution Approach 2:
The patent changes the measurement parameter from 2D image intensity to 3D depth values. By using depth information (distance from camera to object surface) as the primary measurement parameter, the system becomes insensitive to variations in luminosity, temperature, and other environmental factors that affect optical intensity
3Extent of automation
If mathematical models with multivariable regression are used to predict body mass, then indirect measurement can be achieved, but the models become very complex and inflexible to handle environmental uncertainties
Solution Approach 1:
The patent extracts only the essential geometric features needed for mass estimation from the complex environmental data. By focusing on 3D body volume and shape parameters from depth images rather than attempting to model all environmental variables, the system achieves automated prediction with simplified, more flexible models that directly relate observable geometry to mass
Data Source
AI summary
A method for predicting body mass of herd animal that comprises the steps of: collecting data from the livestock environment; detect and select the animal; determining geometric characteristics of each RGB-D depth image; and obtaining the prediction of herd animal body mass. Furthermore, the present invention relates to a system for predicting body mass of herd, comprising: at least one RGB-D capture set including at least one RGB-D imaging sensor; at least one storage module; and at least one remote monitoring module; wherein the storage module stores a set of instructions that, when executed by a processor, carries out the method for predicting body mass of herd animal.