Multi-Scene Fire and Smoke Detection Benchmark (MS-FSDB)
收藏资源简介:
Multi-Scene Fire and Smoke Detection Benchmark (MS-FSDB) 是由浙江大学创建的一个全面且精细的火灾与烟雾检测基准数据集。该数据集包含12,586张图像,描绘了2,731个场景,涵盖了3,603个正样本和8,983个负样本。数据集的创建过程包括从公开资源中系统地收集多样化资源,并进行场景扩展和重新标注,以确保数据的准确性和一致性。MS-FSDB不仅包括火焰检测,还涵盖了烟雾检测任务,适用于各种室内外复杂场景,旨在为火灾检测技术的突破和发展提供强有力的支持。
Multi-Scene Fire and Smoke Detection Benchmark (MS-FSDB) is a comprehensive and refined fire and smoke detection benchmark dataset created by Zhejiang University. This dataset contains 12,586 images depicting 2,731 scenes, covering 3,603 positive samples and 8,983 negative samples. The construction of the dataset involves systematically collecting diverse resources from public sources, followed by scene expansion and re-annotation to ensure data accuracy and consistency. MS-FSDB covers not only flame detection but also smoke detection tasks, and is applicable to various complex indoor and outdoor scenarios. It aims to provide strong support for the breakthroughs and development of fire detection technologies.
多场景火灾和烟雾检测基准
数据集概述
- 数据集名称: Multi-Scene Fire and Smoke Detection (FSD) Benchmark
- 数据集类型: 包含五个先前的FSD数据集和一个新的测试集
- 数据集处理: 重新处理了先前的FSD数据集,统一了所有数据集的标签格式
数据集用途
- 用途: 用于多场景火灾和烟雾检测的基准测试
数据集来源
- 来源: 由Xiaoyi Han, Nan Pu, Zunlei Feng, Yijun Bei, Qifei Zhang, Lechao Cheng, Liang Xue等人创建
- 机构: 浙江大学、特伦托大学、合肥工业大学、苏州城市大学
数据集发布
- 发布状态: 论文已被PRCV 2024接受,具体代码发布时间为7月10日之后

- 1Benchmarking Multi-Scene Fire and Smoke Detection浙江大学 · 2024年



