Work

Condition-Embedded Visible-Infrared image dataset for typical ground features

Year: 2026

Type: data-paper

Abstract: Infrared imaging is significantly influenced by weather, temperature and humidity, indicating notable differences in infrared characteristics for the same scene under varying weather conditions. Therefore, studying translation from visible light images to infrared images must consider the conditional information. However, existing visible-light-infrared image datasets do not account for conditional information. This dataset rigorously records imaging condition information. From 2022 to 2024, it has accumulated 22,435 pair data with condition information from multiple regions, covering five kinds of weather including cloudy, sunny, rainy, snowy, and foggy. The scenes include eight typical ground features such as water, forest, grassland, beach, farmland, buildings, roads, and bare land. It accurately records information such as temperature, humidity etc. at the moment of data collection. This dataset provides data support for studying the precise translation between visible light and infrared images.

Source: Scientific Data

DOI: 10.1038/s41597-026-06847-7

Authors: Huang Zhijian, Hui Bingwei, Ma Yanxin, Yang Zhichao, Zhang Shengxian

Institutions: Changsha University, National University of Defense Technology

Language: English

Topic: Advanced Image Fusion Techniques

Subfield: Media Technology

Field: Engineering

Domain: Physical Sciences

Open Access status: gold

Estimated APC paid: $2,490

Funders: Education Department of Hunan Province, National Key Laboratory of Shockwave and Detonation Physics