Thermal Image Super-Resolution for Low-Light Object Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing object detection systems using both visible light and thermal images face challenges due to degradations in visible light images caused by weather conditions or low light, leading to slower or degraded performance.

Innovation Solution

A method and system that involve obtaining thermal images, upsampling them using super-resolution techniques, and projecting them into a visible light color space for input into a visible light object detector, allowing for effective object detection even when visible light images are degraded.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If multiple object detection networks and pipelines are employed for detecting objects in both visible light and thermal images, then detection capability is improved, but system performance becomes slower and degraded

Engineering Contradiction:

Improvedetection capabilityVSAvoidsystem performance speed

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent merges thermal image data with visible light image data into a unified detection framework. By combining the complementary information from both modalities (thermal provides temperature contrast, visible light provides color and texture), the system achieves improved detection capability while maintaining efficient single-pipeline processing, thus resolving the contradiction between reliability and productivity

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent creates a multi-functional detection system that can process both thermal and visible light images through a unified object detection network. This universal approach allows the same network to handle different image types and detection scenarios, improving overall system efficiency and performance while maintaining high detection capability across various conditions

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Reliability

If thermal images are used for object detection, then detection is possible under degraded visible light conditions, but image resolution is lower

Engineering Contradiction:

Improvedetection capability in degraded conditionsVSAvoidimage resolution

Core Design Contradiction:

ReliabilityVSManufacturing precision

Solution Approach 1:

The patent combines thermal images with visible light images to create a composite image that leverages the strengths of both modalities. The visible light image provides high-resolution color and texture information, while the thermal image provides temperature-based object differentiation. This merging allows the system to maintain high resolution while achieving reliable detection in degraded visible light conditions

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent creates a composite image representation that integrates data from both thermal and visible light sensors. This composite approach is analogous to using composite materials - combining different types of information (thermal signatures and visual appearance) to create a more robust and high-resolution detection input that overcomes the limitations of either modality alone

Inventive Principle:

Principle #40Composite materials

3Manufacturing precision

If visible light images are used for object detection, then high resolution and color information are available, but performance degrades under sun flare, rain, low light, and other weather conditions

Engineering Contradiction:

Improveimage resolution and color informationVSAvoiddetection performance in various weather conditions

Core Design Contradiction:

Manufacturing precisionVSReliability

Solution Approach 1:

The patent creates a composite imaging approach that combines visible light images (providing high resolution and color) with thermal images (providing weather-resistant temperature contrast). This composite strategy allows the system to maintain high detection reliability across various weather conditions while preserving the advantages of visible light imaging, as the thermal component compensates for weather-induced degradations

Inventive Principle:

Principle #40Composite materials

4Device complexity

If a shared object detection pipeline is used for both thermal and visible light images, then system complexity is reduced, but thermal image resolution limitations affect detection accuracy

Engineering Contradiction:

Improvedetection pipeline complexityVSAvoidobject detection accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent merges thermal and visible light image data before feeding it into a single object detection pipeline. By combining the data at the input stage, the system maintains a simple shared pipeline architecture while ensuring that the detection network receives enriched information that compensates for thermal image resolution limitations, thus achieving both low complexity and high accuracy

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS20250159317A1Thermal sensor image augmentation for object detection

Publication Date: 2025.05.15 FAURECIA IRYSTEC INC

AI summary

A system and method is provided for of detecting an object within an image. The method includes: obtaining a thermal image captured by a thermal image sensor; upsampling the thermal image using a super-resolution technique to generate an upsampled thermal image; and detecting an object within the upsampled thermal image through inputting the upsampled thermal image into a visible light object detector configured to detect objects within image data within a visible light color space.