Arlovic Matej, Hrzic Franko, Patel Mitesh, Bednarz Tomasz, Balen Josip
University of J.J. Strossmayer Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology, 31000, Osijek, Croatia.
Department of Orthopaedic Surgery and Sports Medicine, Boston Children's Hospital, Harvard Medical School, Boston, USA.
Sci Rep. 2025 May 14;15(1):16759. doi: 10.1038/s41598-025-01571-5.
Timely fire detection in industrial environments is crucial to safeguarding people and property. Deep neural networks have shown effectiveness in fire detection over traditional methods. However, they require high-quality datasets, which are costly and time-intensive to gather. To overcome this issue, we created the SYN-FIRE dataset, consisting of 2000 labeled images of simulated indoor industrial fires using NVIDIA Omniverse. By using U-Net++ as the baseline, this study explores the impact of the new SYN-FIRE dataset on models' performance when combined with four publicly available datasets. Two ablation studies were conducted: one replacing portions of real data from publicly available datasets with synthetic data and the other adding various amounts of synthetic data. With over 200 models trained across three resolutions, the results indicate that incorporating additional synthetic data improved DiceScore by [Formula: see text] to [Formula: see text] (FireBot and BowFire datasets, respectively) while substituting real data with synthetic data generally enhanced performance but with exceptions. Furthermore, tests on challenging real-life fire images confirmed that synthetic data boosts model generalization, supported by GRAD-CAM saliency maps. Finally, we provide key takeaways that point out the main findings of our research. The SYN-FIRE dataset is publicly available to encourage further research in fire detection and prevention.
在工业环境中及时进行火灾探测对于保护人员和财产至关重要。与传统方法相比，深度神经网络在火灾探测中已显示出有效性。然而，它们需要高质量的数据集，而收集这些数据集成本高昂且耗时。为了克服这个问题，我们创建了SYN-FIRE数据集，该数据集由使用NVIDIA Omniverse生成的2000张模拟室内工业火灾的标注图像组成。本研究以U-Net++作为基线，探讨了新的SYN-FIRE数据集与四个公开可用数据集结合时对模型性能的影响。进行了两项消融研究：一项是用合成数据替换公开可用数据集中的部分真实数据，另一项是添加不同数量的合成数据。在三种分辨率下训练了200多个模型，结果表明，加入额外的合成数据分别将FireBot和BowFire数据集的DiceScore提高了[公式：见原文]至[公式：见原文]，而用合成数据替代真实数据总体上提高了性能，但也有例外情况。此外，对具有挑战性的真实火灾图像的测试证实，合成数据提高了模型的泛化能力，GRAD-CAM显著性图也支持这一点。最后，我们提供了关键要点，指出了我们研究的主要发现。SYN-FIRE数据集已公开提供，以鼓励在火灾探测和预防方面进行进一步研究。
