Thermal Image Retrieval Using Feature-Based Geometric Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional thermography systems face inefficiencies and high risks of error in retrieving and processing thermal images, particularly when repeatedly analyzing the same scene or object, due to manual handling and poor alignment between images.
Innovation Solution
A method and system for feature-based retrieval of thermal images, using a reference image to determine corresponding locations in retrieved images, applying a transformation function for geometric alignment, and placing thermographic tools on these locations, enabling efficient and accurate scene comparison over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual handling and alignment of thermal images is performed, then flexibility in processing is maintained, but time consumption and error risk increase significantly
Solution Approach 1:
The system performs automatic image retrieval and alignment without manual intervention. The processor automatically retrieves thermal images based on similarity measures and aligns them using transformation functions, making the system self-sufficient and eliminating the need for manual handling while reducing time consumption
Solution Approach 2:
The patent replaces manual mechanical alignment operations with automated computational methods. The processor uses transformation functions and similarity measures to automatically align thermal images, substituting the mechanical manual process with an automated digital system that reduces both time and error risk
2Ease of operation
If manual alignment of thermal images is performed, then image processing can be done, but error risk increases
Solution Approach 1:
The system automatically retrieves and aligns thermal images without manual intervention, eliminating human error. The processor independently performs image retrieval based on similarity measures and applies transformation functions for alignment, making the system self-sufficient and highly reliable
Solution Approach 2:
The system uses similarity measures as feedback to automatically retrieve relevant thermal images and adjust alignment parameters. This closed-loop approach ensures accurate matching and alignment while eliminating manual errors, as the system self-corrects based on quantitative similarity metrics
3Measurement precision
If feature-based retrieval with transformation functions is implemented, then alignment accuracy improves, but system complexity increases
Solution Approach 1:
The patent replaces complex manual alignment procedures with automated transformation functions. The processor automatically applies geometric transformations based on detected features, achieving high alignment accuracy without requiring complex manual operations. The complexity is shifted from manual skill to automated computational algorithms
Solution Approach 2:
The system automatically performs feature detection, image retrieval, and alignment transformation without manual intervention. The processor self-manages the entire alignment process using transformation functions, achieving high precision while keeping the user interface simple. The complexity is handled internally by the system rather than being exposed to the user
4Productivity
If automated image retrieval is implemented, then processing speed increases, but complexity of implementation increases
Solution Approach 1:
The patent replaces manual image selection and processing with automated retrieval systems. The processor automatically retrieves thermal images using similarity measures, eliminating time-consuming manual search and selection processes. The implementation complexity is encapsulated in the automated algorithms, providing fast processing speed to the user with minimal effort
Solution Approach 2:
The system implements a universal retrieval mechanism using similarity measures that can handle various thermal image retrieval scenarios. This multi-functional approach allows the system to automatically retrieve and align images across different conditions and applications, achieving high productivity through a single versatile automated system rather than multiple specialized manual processes
Data Source
AI summary
A method and an image processing system of processing thermal images captured using a thermal imaging system. The processing comprises determining a reference image depicting a scene and retrieving, from a number of thermal images comprised in a data storage, a set of one or more thermal images based on a predetermined similarity measure. A measurement location in the reference image is determined, and for each of the one or more retrieved thermal images: a corresponding measurement location in the respective retrieved thermal image is determined; and a thermographic tool is placed on the corresponding measurement location in the respective retrieved thermal image.