李康

Kang Li

Postdoctoral Researcher State Key Laboratory of Fire Science, University of Science and Technology of China Hefei, Anhui, China

Email: kanglee07@ustc.edu.cn

Phone: (+86) 551 63600770

Address: Jinzhai Road 96, State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei 230026, Anhui, P. R. China

ORCID: 0000-0001-5163-6904

About Me

I am a Postdoctoral Researcher at the State Key Laboratory of Fire Science, University of Science and Technology of China (USTC). My research lies at the intersection of artificial intelligence, computer vision, and fire science, with a focus on semantic segmentation, image generation, lightweight neural network design, intelligent fire and smoke detection, moving fire source detection, and multimodal AI. My work focuses on developing accurate, efficient, and robust visual perception methods for complex environments, particularly for early smoke identification and intelligent fire monitoring. My recent research includes lightweight smoke segmentation, spatial-frequency feature learning, fine-grained smoke perception, multimodal fire detection, and AI-assisted image restoration. I have published several first-author papers in IEEE Transactions on Image Processing, Pattern Recognition, IEEE Transactions on Consumer Electronics, Displays and other international journals. My collaborative research has also appeared in npj Heritage Science, a Nature Portfolio journal.

Research Interests

• Intelligent fire and smoke detection​
• Image generation and generative AI​
• Computer vision and image understanding
• Multimodal AI and multi-source information fusion​
• AI-enabled early fire warning and intelligent fire safety
• Lightweight neural networks and efficient deep learning​
• Semantic segmentation and fine-grained visual perception

Biography

• 2025-present: Postdoctoral Researcher, University of Science and Technology of China
• 2022-2025: Ph.D. in Computational Mathematics, Shanghai Normal University
• 2019-2022: Master's Degree in Information and Communication Engineering, Shanghai Normal University

Research Overview

My research focuses on integrating artificial intelligence with fire science to address critical challenges in fire prevention, detection, monitoring, and emergency response.

Selected Publications

1. Kang Li, Yuhuan Peng, Chong Wang, Wei Dai, Qixing Zhang, Feiniu Yuan. PPASNet: A Lightweight Progressive Posterior-guided Asymmetric Spatial Network for Early Smoke Segmentation. IEEE Transactions on Image Processing (CCF-A, IF=15.3), accepted, 2026

2. Kang Li, Feiniu Yuan, Chunmei Wang, Chunli Meng. Multi-Stage Group Interaction and Cross-Domain Fusion Network for Real-Time Smoke Segmentation. IEEE Transactions on Image Processing (CCF-A, IF=15.3), 35: 124-135, 2026. DOI: 10.1109/TIP.2025.3646455

3. Kang Li, Feiniu Yuan, Chunmei Wang. Frequency-Space Interaction With Hierarchical Aggregation Network for Lightweight Smoke Image Segmentation. IEEE Transactions on Consumer Electronics (IF=9.9), 71(2): 2632-2643, 2025.DOI: 10.1109/TCE.2025.3572007

4. Kang Li, Feiniu Yuan, Chunmei Wang. An Effective Multi-Scale Interactive Fusion Network with Hybrid Transformer and CNN for Smoke Image Segmentation. Pattern Recognition (CCF-B, IF=9.1), 159: 111177, 2025. DOI: 10.1016/j.patcog.2024.111177

5. Kang Li, Yuhuan Peng, Hengsheng Zhang, Wei Dai, Chong Wang, Feiniu Yuan, Qixing Zhang. ULSFNet: Ultra-Lightweight Spatial-Frequency Fusion Network with Gaussian Enhancement for Early Smoke Segmentation. Displays (IF=3.9), 96: 103730, 2027. DOI:10.1016/j.displa.2026.103730

6. Feiniu Yuan, Kang Li (Co-first Author), Chunmei Wang, Zhijun Fang. A Lightweight Network for Smoke Semantic Segmentation. Pattern Recognition (CCF-B, IF=7.5), 137: 109289, 2023.DOI: 10.1016/j.patcog.2022.109289

7. Yuhuan Peng, Fukang Song, Kang Li, Feiniu Yuan. A Shared Degradation-Aware and Context-Guided Framework for Restoring Ancient Chinese Inscriptions and Murals. npj Heritage Science (Nature Portfolio), 2026. DOI: 10.1038/s40494-026-02950-5

8. Qie Gao, Haihui Wang, Zhenhai Qin, Linhao Fan, Kang Li, Chong Wang, Qixing Zhang. Using Multimodal Large Language Models for False Alarm Reduction in Image-Based Fire Detection. Fire Technology, 62(6), 2026. DOI: 10.1007/s10694-026-02000-3

9. Dena Zhu, Zhe Yang, Mingjin Wang, Lei Zhang, Kang Li (Corresponding Author), Feiniu Yuan. A Hybrid Network of Transformer and CNN by Attention-Driven Edge Enhancement for Smoke Segmentation. KSII Transactions on Internet and Information Systems, 19(11): 3984-4003, 2025. DOI: 10.3837/tiis.2025.11.013

10. Gang Wen, Fangrong Zhou, Yutang Ma, Hao Pan, Hao Geng, Jun Cao, Kang Li (Corresponding Author), Feiniu Yuan. A Dense Multi-Scale Context and Asymmetric Pooling Embedding Network for Smoke Segmentation. IET Computer Vision, 18(2): 236-246, 2024. DOI: 10.1049/cvi2.12246

11. Feiniu Yuan, Kang Li, Chunmei Wang, Jinting Shi, Yaowen Zhu. Fully Extracting Feature Correlation Between and Within Stages for Semantic Segmentation. Digital Signal Processing (IF=2.9), 127: 103578, 2022. DOI: 10.1016/j.dsp.2022.103578

12. Feiniu Yuan, Yaowen Zhu, Kang Li, Zhijun Fang, Jinting Shi. An Anisotropic Non-Local Attention Network for Image Segmentation. Machine Vision and Applications, 33(2): 23, 2022. DOI: 10.1007/s00138-021-01265-8

Selected Patents

1. Qixing Zhang, Yuxue Zhang, Chong Wang, Kang Li. Method for Pan-Tilt Camera Deployment and Automatic Calibration for Fire Localization on High-Rise Building Facades.CN120876622B, 2025-11-28.

2. Chunmei Wang, Kang Li, Feiniu Yuan. Semantic Segmentation Method and System Based on Deep Convolutional Neural Networks. CN113902925B, 2025-01-03.
3. Chunmei Wang, Kang Li, Feiniu Yuan. Lightweight Smoke Segmentation Method and System Based on Deep Learning. CN113936241B, 2024-11-26.

Selected Projects

Professional Service

I serve as a reviewer for journals including:

• IEEE Transactions on Industrial Informatics
• Pattern Recognition
• Fire Safety Science
• Machine Vision and Applications
• The Journal of Supercomputing
• Signal, Image and Video Processing
• Data Mining and Knowledge Discovery
• International Journal of Multimedia Information Retrieval
• International Journal of Machine Learning and Cybernetics
• Journal of King Saud University Computer and Information Sciences
• ...