ML Object Detection for Multi-Emissivity Temperature Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional techniques fail to provide accurate temperature measurements of multiple objects in an image due to differences in emissivity, which complicates the analysis of thermal radiation data.
Innovation Solution
A pyrometric apparatus employing a digital camera and machine learning to identify objects within an image, referencing an emissivity database to determine the emissivity of each object, and subsequently using pyrometric techniques to calculate their temperatures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional temperature measurement techniques are used on multiple objects with different emissivities, then the measurement process is simple, but the temperature measurement accuracy deteriorates
Solution Approach 1:
The patent segments the image into multiple regions corresponding to different objects, and assigns different emissivity values to each region based on object identification. This allows accurate temperature measurement for each object despite varying emissivities, resolving the contradiction between measurement precision and device complexity by applying complexity only where needed (per-object emissivity assignment) rather than uniformly across the entire measurement system.
Solution Approach 2:
The patent dynamically changes the emissivity parameter for different objects in the image based on their identified types. By retrieving emissivity values from a database according to the detected object class, the system adapts the measurement parameters to match the specific thermal radiation characteristics of each object, thereby improving temperature measurement accuracy without requiring a completely complex reconfiguration of the measurement system.
2Measurement precision
If object-specific emissivity detection is implemented for each object in the image, then temperature measurement accuracy improves, but the processing time and computational complexity increase
Solution Approach 1:
The patent performs preliminary object identification and emissivity assignment before temperature measurement. By detecting objects and retrieving their emissivity values in advance, the system prepares the necessary parameters beforehand, which streamlines the subsequent temperature measurement process and reduces overall processing time despite the added step of object-specific analysis.
Solution Approach 2:
The patent uses a database that stores pre-characterized emissivity information for various object types. Instead of measuring emissivity directly for each object in real-time, the system copies the appropriate emissivity value from the database based on object identification, significantly reducing the computational time and complexity required compared to direct emissivity measurement while maintaining measurement accuracy.
3Measurement precision
If machine learning algorithms are used to identify objects and their emissivity, then measurement accuracy for multiple objects improves, but the device complexity and computational requirements increase
Solution Approach 1:
The patent introduces an intermediary database that stores emissivity information for various object types. The machine learning algorithm identifies objects and queries this database to retrieve appropriate emissivity values, rather than requiring the system to directly measure or calculate emissivity for each object. This intermediary layer simplifies the overall system architecture by separating object identification from emissivity determination, reducing computational complexity while maintaining measurement precision.
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
Enables accurate temperature distribution measurement across multiple objects with different emissivities, improving measurement accuracy and reducing false alerts by automatic object detection and classification.
Implementation Method 1
the emissivity of a surface of an object corresponds to the effectiveness of the surface in emitting energy as thermal radiation
Data Source
AI summary
A method includes capturing an image, and performing region detection on the captured image. The region detection includes identifying an object represented in the captured image. The method further includes detecting emissivity of the identified object and determining the temperature of the object based on the detected emissivity of the object.