Multisensory Imaging for Controlled Environment Horticulture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing agricultural imaging techniques are limited to outdoor environments and struggle to effectively adapt to Controlled Environment Horticulture (CEH), where crops are not accessible for aerial imaging and rely on artificial lighting that may not provide sufficient light for imaging.
Innovation Solution
The development of a multisensory imaging system that integrates multispectral imaging and sensing, using a combination of fluid-cooled LED-based lighting fixtures and imagers/sensors to acquire imagery and spectra over a broad wavelength range, allowing for comprehensive crop monitoring from large fields of view to individual plants.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional aerial imaging techniques are used for outdoor crops, then crop monitoring precision and early detection capability are improved, but the system cannot be applied to controlled environment horticulture where crops are inaccessible to aerial imaging
Solution Approach 1:
The patent inverts the conventional aerial imaging approach by placing imaging sensors and illuminators directly within the controlled environment growth chambers, allowing close-range imaging of crops that were previously inaccessible to aerial techniques
Solution Approach 2:
The imaging system is designed to be universally applicable to both outdoor and controlled environment horticulture by integrating multiple imaging modes (multispectral, hyperspectral, thermal) and adaptable illuminators that can function in various lighting conditions
2Illumination intensity
If artificial lighting is used in controlled environment horticulture, then crop growth is enabled, but the lighting may not provide sufficient light for imaging
Solution Approach 1:
The lighting system is segmented into separate functional components: grow lights for crop photosynthesis and illuminators specifically designed for imaging, allowing each to be optimized for its respective purpose without compromising the other
Solution Approach 2:
The imaging illuminators operate periodically or on-demand rather than continuously, providing intense light bursts during imaging operations while the grow lights maintain steady operation for crop growth, thus balancing imaging needs with productivity
3Measurement precision
If multiple imaging parameters (pixel resolution, image bandwidth, radiometry resolution, positional accuracy) are optimized, then image quality and crop monitoring effectiveness are improved, but the cost of imaging equipment increases
Solution Approach 1:
The patent combines multiple imaging functions (multispectral imaging, hyperspectral imaging, thermal imaging) and illuminators into integrated imaging systems that can be deployed in controlled environment facilities, achieving high measurement precision through functional integration rather than multiple separate expensive systems
Solution Approach 2:
The system allows dynamic adjustment of imaging parameters (spectral range, resolution, temporal frequency) based on specific monitoring needs, enabling optimization of image quality for different crop conditions while managing equipment costs through selective parameter deployment
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This solution reduces costs, improves access to agricultural imaging for farmers, and enhances image quality and information derived from images, providing holistic control and monitoring solutions for CEH environments.
Implementation Method 1
fluid-cooled LED-based lighting fixtures
Implementation Method 2
spectral reflectance measurement bands
Data Source
AI summary
A one-dimensional (1D), two-dimensional (2D) or three-dimensional (3D) array of sensor nodes having a field of view and configured to measure multiple conditions within the field of view. A processor coupled to the array of sensor nodes generates multiple mono-sensory images respectively corresponding to the multiple measured conditions. Each image includes multiple pixels collectively representing a unique measured condition, and each pixel is digitally represented by coordinates for a spatial position in the field of view, and a measurement value for the unique measured condition at the spatial position. The processor processes the mono-sensory images using machine learning (ML) techniques based on a reference condition library of labeled feature sets to estimate or determine one or more environmental conditions (e.g., a condition of a plant, identification of substances or compounds present in a plant or part of a plant, ambient conditions proximate to a plant) at respective spatial positions in the field of view.