--- title: "Smoke and Fire Detection Videos Dataset" description: "The dataset consists of videos containing fire and smoke scenes, varying in length and content. Each video frame is annotated with bounding boxes that localize…" url: "https://unidata.pro/datasets/smoke-and-fire-detection-videos-dataset/" date_modified: "2026-04-24T20:14:27+03:00" language: "en-US" --- This smoke and fire detection dataset offers 85 high-quality RGB videos in MP4 format with JSON annotations, including frame numbers, object coordinates, and classes for fire and smoke. The dataset supports object detection, fire and smoke recognition, and computer vision tasks, enabling deep learning models for real-time monitoring, early detection, and efficient fire management under varied environmental conditions. ## Dataset Structure ### The Numbers Section **Numbered list:** - **Number:** 85 — **Text:** Videos ### Tooltips Section **Tooltip items:** - **Name:** Object Detection - **Name:** Computer Vision - **Name:** Machine learning - **Name:** Smart Cities ### Dataset Information **Table with data:** | Characteristic | Data | | --- | --- | | Description | Videos with fire and smoke | | Data types | Video | | Tasks | Object Detection, Computer Vision | | Number of videos | 85 | | Marking | Bounding Box | | Number of files in a set | part1: 19 videos x 1 min part2: 9 videos x 17 mins part3: 57 videos x 3 mins | | Labeling | Frame_num, width, height, objects with coordinates and classes | **Media Slider:** - **Video on Slayder:** **Link to the sample:** [Download sample](https://drive.google.com/drive/folders/1eN_sw4u8dxoKF_c0hXs3lEODggSInUaw?usp=sharing) ### Technical Specifications **Table with data:** | Characteristic | Data | | --- | --- | | Video extension | MP4 | | Extension of labeling file | JSON | **Source and data collection methodology:** Source and collection methodology: Data was collected by a partner of Unidata. ### Dataset Use Cases - Slider **Industry Cards:** - **Industry:** Forestry Management — **Title:** Early Wildfire Detection — **Text:** Smoke and Fire Detection Videos Dataset enables forestry authorities to detect wildfire occurrences promptly. With annotated RGB videos showing fire spreads and smoke detection, monitoring systems can track fire behavior, analyze environmental conditions, and provide real-time alerts. This supports efficient fire management, early suppression, and protection of forested areas. - **Industry:** Industrial Safety — **Title:** Monitoring Fire Hazards in Factories — **Text:** Industrial facilities can use this video dataset for fire recognition and smoke detection in manufacturing and storage areas. The dataset provides high-quality videos with bounding box annotations, enabling learning models to identify fire events quickly, reduce false alarms, and improve safety protocols while integrating real-time monitoring into existing detection systems. - **Industry:** Urban Emergency Response — **Title:** Supporting Fire and Smoke Response Teams — **Text:** Emergency services can use the dataset to train detection systems for fires in urban areas. By analyzing video surveillance and fire detection in different environments, responders can predict fire spreading, plan evacuation strategies, and optimize resource allocation. This dataset strengthens fire safety measures and enhances rapid emergency response capabilities. - **Industry:** Research and Development — **Title:** Training Deep Learning Fire Detection Models — **Text:** Researchers can utilize this dataset to develop and test advanced deep learning algorithms for fire recognition and smoke detection. With diverse video clips showing varying fire situations and environmental conditions, the dataset supports accurate model training, evaluation of detection algorithms, and improvement of video-based fire surveillance systems. ### Fact **FAQs Heading:** FAQs **List of Questions:** - **Question:** Can I evaluate a sample before purchasing the full fire surveillance dataset? — **Answer:** Yes, we provide free samples for evaluation. This allows you to assess the video quality, annotation accuracy, and the dataset's suitability for your fire recognition and smoke detection projects before committing to the full purchase. - **Question:** What is the primary application of this fire and smoke video dataset? — **Answer:** This dataset is designed for developing and improving automated fire detection systems and computer vision models. It serves as essential training data for machine learning and deep learning algorithms focused on early fire recognition in video surveillance and monitoring systems. - **Question:** What types of annotations are provided for model training? — **Answer:** Each video frame includes detailed bounding box annotations in JSON format. These labels precisely identify the location and class (fire or smoke) of objects, which is crucial for training accurate object detection and recognition systems. - **Question:** How was this video data for fire detection collected? — **Answer:** The video sequences were collected by a trusted partner of Unidata. - **Question:** How are Unidata datasets licensed? — **Answer:** Unidata datasets follow a dual-licensing model. Free samples are provided for trial and testing, while the complete datasets, including this comprehensive fire surveillance collection, are available exclusively through purchase. - **Question:** Do Unidata datasets comply with data privacy regulations? — **Answer:** Yes. All datasets are curated in compliance with GDPR and applicable data protection laws. Data is collected from legally permissible sources to ensure ethical and lawful usage in your computer vision and machine learning projects. - **Question:** How is the dataset stored and delivered? — **Answer:** Unidata stores all datasets securely on AWS cloud infrastructure, aligned with ISO 27001 and ISO 27701 standards. This guarantees a secure, reliable, and privacy-focused environment for handling all video data and annotations. - **Question:** How does the dataset distinguish between fire and smoke in its labels? — **Answer:** Each annotated frame includes object class labels that separately identify fire and smoke, alongside coordinates and frame numbers in JSON format. This lets models learn to detect and differentiate the two independently. [Full list of this site's AI-readable pages](https://unidata.pro/llms.txt)