Contactless Temperature Measurement Using Visible and Thermal Image Registration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current contactless monitoring systems for body temperature and medical indicators require either expensive high-accuracy sensors or compromise on performance with low-cost sensors, making them unsuitable for adapting to changing environmental conditions.
Innovation Solution
A system and method utilizing a visible spectrum camera, a thermal image sensor, and processing circuitry to measure the temperature of subjects within a scene, including a reference object with unknown emissivity, by registering visible and thermal images onto a common coordinate system, identifying regions of interest, determining temperatures, and correcting for ambient temperature differences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If expensive high-accuracy sensors are used for contactless temperature measurement, then measurement precision is improved, but device cost increases
Solution Approach 1:
The patent combines a visible spectrum camera and a thermal image sensor into a single monitoring system. The visible camera captures images for identifying regions of interest and reference objects, while the thermal sensor measures temperature. By merging these two sensor types, the system achieves accurate temperature measurement without requiring expensive specialized sensors, as the thermal sensor can be a lower-cost component compensated through image processing and reference object calibration.
Solution Approach 2:
The patent introduces a reference object with known emissivity as an intermediary element in the measurement process. This reference object is captured in the visible spectrum image and used to calibrate the thermal image sensor by providing a known temperature reference point. The system uses this intermediary reference to compensate for environmental conditions and sensor variations, enabling accurate temperature measurement with lower-cost thermal sensors.
2Ease of manufacture
If low-cost sensors are used for contactless monitoring, then device cost decreases, but measurement precision and adaptability to environmental conditions deteriorate
Solution Approach 1:
The patent implements a feedback mechanism where the system continuously captures visible spectrum images and thermal images, compares the reference object's apparent temperature with its known temperature, and adjusts the temperature measurement accordingly. The processing circuitry uses this feedback loop to compensate for drift and environmental variations, maintaining measurement precision with low-cost sensors by constantly recalibrating against the reference object.
Solution Approach 2:
The reference object serves as a mediator that bridges the gap between low-cost sensor capabilities and high-accuracy requirements. By placing an object with known emissivity and temperature characteristics in the scene, the system can use it as a calibration reference to correct thermal measurements, enabling low-cost sensors to achieve high measurement precision through computational compensation.
3Ease of manufacture
If current contactless systems use low-cost sensors to reduce cost, then device cost decreases, but adaptability to changing environmental conditions worsens
Solution Approach 1:
The patent performs preliminary calibration by capturing a visible spectrum image to identify the reference object and its characteristics before taking thermal measurements. The system pre-processes the visible image to locate and characterize the reference object, then uses this pre-acquired information to guide the thermal measurement and compensation process. This preliminary action enables the low-cost system to adapt to environmental conditions by having the reference information ready for calibration.
Solution Approach 2:
The reference object acts as an environmental adapter, mediating between the thermal sensor and changing environmental conditions. By providing a stable reference point that is captured in both visible and thermal images, the system can distinguish between actual temperature changes and environmental artifacts, enabling adaptability to varying lighting, temperature, and humidity conditions despite using low-cost sensors.
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
The system achieves high-accuracy temperature measurement of subjects while adapting to changing environmental conditions, without the need for expensive sensors, by utilizing a reference object to correct for ambient temperature variations.
Implementation Method 1
a thermal image sensor configured to acquire images of the scene comprising (a) at least the RoI of each of the subjects, and (b) the reference object
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system for measuring temperature of one or more subjects within a scene including a reference object having an unknown emissivity, having an ambient temperature, the system comprising: a visible spectrum camera capable of acquiring images of the scene comprising (a) at least a Region of Interest (Rol) of each of the subjects, and (b) the reference object; a thermal image sensor capable of acquiring images of the scene comprising (a) at least the Rol of each of the subjects, and (b) the reference object; and a processing circuitry configured to: obtain (a) a visible spectrum image captured by the visible spectrum camera, and (b) a thermal image captured by the thermal image sensor, and (c) an indication of a scene ambient temperature within the scene; register the visible spectrum image and the thermal image onto a common coordinate system; identify (a) Rol pixels, on the common coordinate system, of the Rols of the subjects within the visible spectrum image, (b) reference object pixels, on the common coordinate system, of the reference object within the visible spectrum image and (c) a parameter correlated to an emissivity of the reference object, based on the reference temperature and on the indication of the scene ambient temperature; determine (a) Rol temperatures by analyzing respective Rols pixels on the thermal image, and (b) a reference temperature by analyzing the reference object pixels on the thermal image; and upon existence of a difference between the reference temperature and the scene ambient temperature, correct the Rol temperatures, based on the difference and utilizing the parameter, to compensate for the difference, giving rise to corrected Rol temperatures.