模式识别与智能系统、图像与视频的内容理解与分析、工业视觉。

1989年6月，毕业于5822yh银河国际计算机系软件专业；

1989年7月-1991年7月，就职于中国科学院遥感应用研究所（北京）；

1991年8月-1996年6月，就职于中国煤田地质总局航测遥感局；

1996年7月-2002年7月，就职于日本太平洋软件银行株式会社（东京）；

2002年8月-今，就职于5822yh银河国际；

现任5822yh银河国际软件学院研究员、博士生导师。 IEEE 会员， CSIG 会员，CCF 会员，美国项目管理学会认证的项目管理专家（PMP）。

主要研究方向为计算机视觉、模式识别与智能系统，发表高水平文章90余篇、获得授权发明专利20余项。

承担多项国际合作项目、国家重点研发计划课题和企业合作项目， 研究图像、视频内容理解与分析方法及其在工业界的落地应用。面向智能复印机、钢铁质量表面在线检测、智能冰箱等应用进行了大量工程实践，完成了多项国际合作研究项目和面向工业界的实际落地应用项目。项目团队在面向产品线的表面缺陷检测方面已具备项目产业化能力，其“热态高速线材表面缺陷在线检测系统研发”项目已在宝山钢铁股份有限公司的生产线上应用，系统稳定、高效，实现了产品质检自动化，达到国际领先水平。

基于视频图像内容分析的铁路线路危情监测智能化提升关键技术研究

城轨列车振动信号典型异常数据治理方法研究

户外全天候监控视频的异常行为检测关键技术及其应用

多传感技术融合的高铁周界入侵报警技术研究

基于图像的钢铁产品缺陷检测及分类的人工智能平台研发

铁路安检多维信息智慧感知技术与平台

银行票据及钞票算法智能技术研究

安检行业违禁品检测与识别方法研究

热态高速线材成品在线表面检测软件系统研发

基于深度学习的酸洗图像表面缺陷识别与三维视觉伺服技术研究项目

基于深度信息的视频中人体动作识别方法研究

铁路客站智能视频监控分析系统

[1] Zefeng Li, Yuehu Liu, Yonghong Song*, Image Caption Method from Coarse to Fine Based on Dual Encoder-Decoder Framework, IJCNN2024,Yokohama, Japan, 30 June - 5 July 2024.

[2] Hanxiao Li, Yonghong Song*, Tong Geng, Semi-Supervised Crowd Counting Based on Hard Pseudo-Labels, IJCNN2024,Yokohama, Japan, 30 June - 5 July 2024.

[3] Ke Jia, Yonghong Song*, Video Anomaly Detection Via Self-supervised Learning With Frame Interval and Rotation Prediction, IEEE Conference on Multimedia Expo 2024, July 15-19 ,2024, in Niagara Falls.

[4] DaiLiang Zhou, YongHong Song*，PNSP: Overcoming Catastrophic Forgetting Using Primary Null Space Projection in Continual Learning，Pattern Recognition Letters 179（2024）137-143，Available online 15 February 2024.

[5] Jiawei Zheng, Yonghong Song*，Effective Representation Learning is More Effective in Reinforcement Learning than You Think，2024 IEEE International Conference on Robotics and Automation (ICRA).

[6] Zijian Chen , Yonghong Song* , Yue Ma , Guofu Li , Rui Wang , Hao Hu, Interaction in Transformer for Change Detection in VHR Remote Sensing Images, Transactions on Geoscience and Remote Sensing, On page(s): 1-12，Print ISSN: 0196-2892，Online ISSN: 1558-0644 .

[7] Yuanzhao Shao, Yonghong Song*, CENet: improve counting performance of X-ray surface mounted chip counter via scale favor and cell extraction, DOI:10.1007/s10845-023-02223-z，Journal of Intelligent Manufacturing.

[8] Mengyuan Zhao，Yonghong Song*, ABNORMAL-AWARE LOSS AND FULL DISTILLATION FOR UNSUPERVISED ANOMALY DETECTION BASED ON KNOWLEDGE DISTILLATION, the 2023 IEEE International Conference on Image Processing, Oct 9, 2023 - Oct 12, 2023, Oct 9, 2023 - Oct 12, 2023.

[9] Xiaoyao Wu, Yonghong Song*, SGLP-Net: Sparse Graph Label Propagation Network for Weakly-Supervised Temporal Action Localization, The 2023 International Conference on Neural Information Processing , Changsha, China, November 20-23. Neural Information Processing，eBook ISBN：978-981-99-8073-4，Print ISBN：978-981-99-8072-7，P.149-161.

[10]  Yongbiao Deng, Yonghong Song*, Few-Shot Steel Plate Surface Defect Detection with Multi-Relation Aggregation and Adaptive Support Learning, ISIJ International, Vol. 63 (2023), No. 10, pp. 1727–1737，https://doi.org/10.2355/isijinternational.ISIJINT-2023-118.

[11]  Pengcheng Guo, Yonghong Song*, Yuanlin Zhang ; DCMAI: A Dynamical Cross-Modal Alignment Interaction Framework for Document Key Information Extraction，IEEE Transactions on Circuits and Systems for Video Technology，DOI：10.1109/TCSVT.2023.3287296，Date of Publication: 19 June 2023.

[12]  Zhang Y, Song Y*, Hu H, et al. Mixed-former: multi-fusion remote sensing change detection[J]. International Journal of Remote Sensing, 2023, 44(11): 3507-3528. DOI：10.1080/01431161.2023.2224100.

[13]  Mengbo Sun, Yonghong Song*, Hongda Wang, Efficient Temporal Action Localization with Temporal Attention and Gaussian Weight,, 2023 International Joint Conference on Neural Networks，June 18 - 23, 2023，Gold Coast Convention and Exhibition Centre ，Queensland, Australia.

[14]  Yuanyuan Zhang，Yonghong Song*, Anqi Li, Handwritten Chinese Character Generation via Embedding, Decomposition and Discrimination, IJCNN 2023，June 18 - 23, 2023，Gold Coast Convention and Exhibition Centre ，Queensland, Australia.

[15]  Jiahao Liu, Pengcheng Guo, Yonghong Song* ，NC-WAMKD: NEIGHBORHOOD CORRECTION WEIGHT-ADAPTIVE MULTI-TEACHER KNOWLEDGE DISTILLATION FOR GRAPH-BASED SEMI-SUPERVISED NODE CLASSIFICATION，ICASSP 2023.

[16]  Huang Shien, Yonghong Song*, Rong Junyan, Liu Tianshuai, Huang Dong, Lu Hongbin, Single-view cone beam CT reconstruction with swin transformer based deep learning, SPIE Medical Imaging 2023, 19 - 23 February 2023, San Diego, California, United States.

[17]  Yaochen Li , Haochuan Hou, Zikun Dong, Yujie Zang, Ying Zhang,and Yonghong Song*, Spatiotemporal Analysis of Static and Dynamic Traffic Elements From Road Scenes, IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, VOL. 24, NO. 3, MARCH 2023, P.3297-3311.

[18]  S. Zhu, Y. Song*, Yu Zhang and Y. Zhang, "ECFNet: A Siamese network with fewer FPs and fewer FNs for change detection of remote-sensing images", IEEE Geosci. Remote Sens. Lett., vol. 20, Jan. 2023. DOI：10.1109/LGRS.2023.3238553，WOS:000966553600001.

[19]  杜鹏、宋永红*、张鑫瑶，基于自注意力模态融合网络的跨模态行人再识别方法研究，《自动化学报》，2022年6月，第48卷 第6期，pp:1457-1468，DOI：10.16383/j.aas.c190376.

[20]  Ji Wu, Yonghong Song*, Yuanlin Zhang, Siyuan Zhu, Wang Zezhong, Hengyu Zhu，SA-CenterNet: Scale Adaptive CenterNet for X-RAY Luggage Image Detection，ICPR 2022.

[21]  XuMingjie,Songyonghong,XieKangknag, Adaptive Sparse Self-attention for Object Detection,IJCNN2022.

[22] 李晓玉，宋永红* ，余涛. 结合感受野增强和全卷积网络的场景文字检测方法，《自动化学报》，2022年3月，第48 卷第03 期，pp:797-807，DOI：10.16383/j.aas.c190376.

[23]  王鑫，宋永红* ，张元林. 基于显著性特征提取的图像描述算法，《自动化学报》，2022年3月，第48卷第03 期，pp:  735-746，DOI：10.16383/j.aas.c190279.

[24]  Yaochen Li,Yuhui Hong,Yonghong Song*,Chao Zhu, Ying Zhang,Ruihao Wang，SiamPolar: Semi-supervised Realtime Video Object Segmentation with Polar Representation，Journal:  Neurocomputing，Volume 467, 7 January 2022, Pages 491-503.