IJmond Industrial Smoke Segmentation Dataset

Yen-Chia Hsu; Despoina Touska

cs.CVMar 24, 2026

IJmond Industrial Smoke Segmentation Dataset

Authors: Yen-Chia Hsu, Despoina Touska

Organizations: Informatics Institute, University of Amsterdam, Netherlands

Abstract

This report describes a dataset for industrial smoke segmentation, published on a figshare repository (https://doi.org/10.21942/uva.31847188). The dataset is licensed under CC BY 4.0.

Figures & tables

Panel of Figure 1 — Figure 1: Example of a raw image that is scrapped from the "kooks_1" monitoring camera, as well as its annotated mask. White pixels (with value 255) indicate high opacity smoke, and gray pixels (with value 155) indicate low opacity smoke. Background pixels have black color (value 0).

Panel of Figure 2 — Figure 2: Examples of raw images from "kooks_2", and "hoogovens_6_7".

Panel of Figure 3 — Figure 3: Examples of cropped images from the raw image in Figure 1 and their corresponding masks.

Image Segmentation

Image segmentation assigns pixels or voxels to objects and regions, supporting measurement, scene understanding, and decisions across image domains.

See all 399 papers

AusSmoke meets MultiNatSmoke: a fully-labelled diverse smoke segmentation dataset
Apr 26, 2026Weihao Li, Hongjin Zhao, Gao Zhu +4Wildfire Detection Remote Sensing
Bushfire Research Centre of Excellence, Australian National University
A Sentinel-2 benchmark dataset for deep-learning active-fire segmentation across 25 California wildfires, July 2020-August 2026
Sep 14, 2026Shreyan Mitra, Mohammadreza Narimani, Parastoo FarajpoorImage Segmentation Remote Sensing Image Segmentation
California High School, San Ramon, CA, 94583, USA · Department of Biological and Agricultural Engineering, University of California, Davis, Davis, CA, 95616, USA
Wildfire Spread Scenarios: Increasing Sample Diversity of Segmentation Diffusion Models with Training-Free Methods
Mar 20, 2026Sebastian Gerard, Josephine SullivanDiffusion Sampling Diffusion Model Guidance
KTH Royal Institute of Technology, Stockholm, Sweden

Apr 26, 2026J/K move · Enter open · S save

AusSmoke meets MultiNatSmoke: a fully-labelled diverse smoke segmentation dataset

Weihao Li, Hongjin Zhao, Gao Zhu, Ge-Peng Ji+3

Bushfire Research Centre of Excellence, Australian National University

Wildfire Detection Remote Sensing Semantic Segmentation

Open paper AskPDF arXiv

Wildfires are an escalating global concern due to the devastating impacts on the environment, economy, and human health, with notable incidents such as the 2019-2020 Australian bushfires and the 2025 California wildfires underscoring the severity of these events. AI-enabled camera-based smoke detection has emerged as a promising approach for the rapid detection of wildfires. However, existing wildfire smoke segmentation datasets that are used for training detection and segmentation models are limited in scale, geographically constrained, and often rely on synthetic imagery, which hinders effective training and generalization. To overcome these limitations, we present AusSmoke, a new smoke segmentation dataset collected from Australia to address the data scarcity in this region. Furthermore, we introduce a MultiNational geographically diverse and substantially larger fully-labelled benchmark, called MultiNatSmoke, that consolidates publicly available international datasets with the newly collected Australian imagery, expanding the scale by an order of magnitude over previous collections. Finally, we benchmark smoke segmentation models, demonstrating improved performance and enhanced generalization across diverse geographical contexts. The project is available at Github.

Abstract

Image Segmentation

Explore similar work

AusSmoke meets MultiNatSmoke: a fully-labelled diverse smoke segmentation dataset

A Sentinel-2 benchmark dataset for deep-learning active-fire segmentation across 25 California wildfires, July 2020-August 2026

Wildfire Spread Scenarios: Increasing Sample Diversity of Segmentation Diffusion Models with Training-Free Methods