张金霞，东南大学自动化学院副教授（破格），博士生导师，东南大学至善青年学者A类。研究方向为图文多模态、计算机视觉、缺陷检测等。作为项目负责人主持国家自然科学基金项目、江苏省自然科学基金项目和若干企业预研项目。在国际学术期刊和国际会议上发表学术论文40余篇，包括TII、PR、Applied Energy、TMM等学科顶级期刊。第一发明人授权发明专利8项，公开发明专利6项。指导研究生电子设计竞赛获全国总决赛二等奖、全国总决赛最佳论文奖和华东分赛一等奖。指导本科毕业设计获校优和院优。参与教学的工程设计导论（研讨）课获江苏省一流课程。参与编写的教材《神经网络设计》入选江苏省重点教材和“十三五”国家重点图书出版规划（高等教育出版社）。以并列第一编者的身份编著一本《深度学习》教材，目前已出版5000+册。现为中国自动化学会模式识别与机器智能专委会通讯委员，江苏省人工智能学会模式识别专委会委员，江苏省自动化学会会员。担任Pattern Recognition, Neurocomputing等刊物审稿人，曾担任ACPR 2019 PC member，IscIDE2019 PC member等。

学生招收情况：每年招收博士研究生1-2名，南京校区硕士研究生2-3名，苏州校区硕士研究生1-2名，无锡校区硕士研究生1-2名。

欢迎勤奋、踏实、好学、对图文多模态、计算机视觉和机器学习感兴趣的同学。

联系方式：jinxiazhang@seu.edu.cn

工作经历

2019-04至今，东南大学，自动化学院，副教授（破格）

2016-03至2019-03，东南大学，自动化学院，讲师

2012-10至2014-09，美国哈佛医学院，视觉注意实验室，访问学者，合作导师：Jeremy Wolfe教授

讲授课程

本科生：《数据结构》、《工程设计导论》、《人工智能通识导论》

科研项目

Ø  面向海上光伏电站智慧运维的持续缺陷检测方法研究，东南大学南通海洋高等研究院面上项目，主持

Ø  电商领域可控多模态图文理解和生成，阿里巴巴（中国）网络技术有限公司预研项目，主持

Ø  基于显著区域与语义类别精准匹配的复杂场景弱监督语义分割研究，广东省自然科学基金面上项目，主持

Ø  结合感知分组和视觉注意特性的显著目标检测研究，国家青年基金项目，主持

Ø  基于视觉认知特性的复杂图像显著目标检测研究，江苏省青年基金项目，主持

Ø  AI果实识别系统，企业项目，主持

Ø  光伏组件热斑、隐裂等缺陷诊断分析技术，国家重点研发计划项目，项目骨干

Ø  分布式光伏系统智慧运维技术，企事业单位委托项目，项目骨干

Ø  智能机器人面对复杂场景时的显著目标检测研究，重点实验室课题，主持

Ø  面向复杂自然场景图像的显著区域检测研究，重点实验室课题，主持

部分发表论文 

Ø  Shenghao Dong, Jinxia Zhang*, Yu Shen, Dehong Gao. Referring Solar Cell Defect Segmentation in Electroluminescence Images. IEEE Transactions ON Industrial Informatics (Q1 top一区), 2025, Accepted.

Ø  Jiale Huang, Dehong Gao, Jinxia Zhang*, et al. FashionFAE: Fine-grained Attributes Enhanced Fashion Vision-Language Pre-training[C]//ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE,2025.(CCF-B)

Ø   Yang Hu, Jinxia Zhang*, Kaihua Zhang, et al. Shifting Spotlight for Co-supervision: A Simple yet Efficient Single-branch Network to See Through Camouflage[C]//ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE,2025.(CCF-B)

Ø  Zechao Zhan, Dehong Gao, Jinxia Zhang*, et al. MADiff: Text-Guided Fashion Image Editing with Mask Prediction and Attention-Enhanced Diffusion[C]//ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE,2025.(CCF-B) 

Ø  Yeyuan Wang, Dehong Gao, Lei Yi, Linbo Jin, Jinxia Zhang, Libin Yang, Xiaoyan Cai, Enhancing Fine-Grained Vision-Language Pretraining with Negative Augmented Samples, AAAI, 2025 (CCF-A).

Ø  Yeyuan Wang, Dehong Gao, Bin Li, Rujiao Long, Lei Yi, Xiaoyan Cai, Libin Yang, Jinxia Zhang, Shanqing Yu, Qi Xuan. CoF: Coarse to Fine-Grained Image Understanding for Multi-modal Large Language Models, ICASSP, 2025. (CCF-B)

Ø   Qingyu Lu, Liang Ding, Kanjian Zhang, Jinxia Zhang, Dacheng Tao. MQM-APE: Toward High-Quality Error Annotation Predictors with Automatic Post-Editing in LLM Translation Evaluators, The 31st International Conference on Computational Linguistics, 2025 (CCF-B).

Ø  Jinxia Zhang, Xinyi Chen, Haikun Wei, Kanjian Zhang, A lightweight network for photovoltaic cell defect detection in electroluminescence images based on neural architecture search and knowledge distillation, Applied Energy  (Q1 top一区), 2024, 355, 122184.

Ø  Jinxia Zhang, Yu Shen,Jiacheng Jiang,Shixiong Fang,Liping Chen,Tingting Yan,Zuoyong Li,Kanjian Zhang, Weili Guo, Haikun Wei. Automatic detection of defective solar cells in electroluminescence images via global similarity and concatenated saliency guided network, IEEE Transactions ON Industrial Informatics (Q1 top一区), 2023, 19(6), 7335-7345.

Ø  Jingzheng Deng, Jinxia Zhang*, Zewen Hu, Liantao Wang, Jiacheng Jiang, Xinchao Zhu, Xinyi Chen, Yin Yuan, Chao Wang. RGB-D salient object ranking based on depth stack and truth stack for complex indoor scenes[J]. Pattern Recognition, 2023, 137: 109251.

Ø  Yu Shen, He, Z., Xu, Z., Wang, Y., Li, C., Jinxia Zhang*, Wei, H*. Modeling of photovoltaic modules under common shading conditions. Energy( 1区), 256, 124618，2022. 已检索WOS:000834159800009

Ø  Yu Shen, Tao Fan, Guangzhi Lai, Zhixiong Na, Hu Liu, Ziyao Wang, Yiye Wang, Yiping Jiao, Xinyi Chen, Zhouwei Lou, Jinxia Zhang*, Kanjian Zhang, Haikun Wei*, Modified U-Net based photovoltaic array extraction from complex scene in aerial infrared thermal imagery. Solar Energy(2区), 240:90-103, 2022 已检索WOS:000809727300004 

Ø  Yu Shen, Zhen Xu, Yiye Wang, Chenxi Li, Peizhen Peng, Jinxia Zhang*, Kanjian Zhang, Haikun Wei*, Temperature Estimation of Partial Shaded Photovoltaic  Modules Using a Multi-physics Model, Journal of Photovoltaics(2区), 2022

Ø   Qinghua Ren, Shijian Lu, Jinxia Zhang, Renjie Hu. Salient object detection by fusing local and global contexts. IEEE Transactions on Multimedia(1区), PP(99):1-1, 2020.

Ø  Jinxia Zhang, Krista A. Ehinger, Haikun Wei, Kanjian Zhang, Jingyu Yang. A novel graph-based optimization framework for salient object detection. Pattern Recognition(1区), 64: 39-50, 2017.  WOS: 000392682400004. 

Ø  Jinxia Zhang, Shixiong Fang, Haifeng Zhao, Guang-Hai Liu, Haikun Wei, Lihuan Chen, Kanjian Zhang. Multiple Gestalt principles-based graph for salient region detection. Journal of Electronic Imaging, 27(5), 051227 1-9, 2018. WOS: 000443279700026.

Ø  Jinxia Zhang, Xue Gong, Daryl Fougnie, Jeremy Wolfe. How humans react to changing rewards during visual foraging. Attention Perception & Psychophysics. 79(8): 2299-2309, 2017. WOS: 000414337800005.

Ø  Jinxia Zhang, Shixiong Fang, Krista A. Ehinger, Weili Guo, Wankou Yang, Haikun Wei. Probabilistic Hypergraph Optimization for Salient Object Detection. P368-378, ISciDE2017. WOS:000441466300032. 

Ø  Yu Shen, Xinyi Chen, Jinxia Zhang*, Liping Xie, Kanjian Zhang and Haikun Wei*, A Robust Automatic Method for Removing Projective Distortion of Photovoltaic Modules from Close Shot Images. PRCV, 2020

Ø   Wenjie Liu, Jinxia Zhang*, Haikun Wei*, Kanjian Zhang, Shixiong Fang, Liping Xie, Jian Ge. Short-term PV power prediction considering the influence of aerosol. EEMS, 2020

Ø  Jinxia Zhang, Xue Gong, Daryl Fougnie, Jeremy Wolfe. Using the past to anticipate the future in human foraging behavior. Vision Research, 111:66-74, 2015.WOS:000356027200007. 

Ø  Jinxia Zhang, Jundi Ding, Jingyu Yang. Exploiting global rarity, local contrast and central bias for salient region learning. Neurocomputing(2区), 144:569-580, 2014. WOS:000341677800054.

Ø  Z. Bylinskii¹, E. DeGennaro¹, R. Rajalingham¹, H. Ruda¹, Jinxia Zhang¹, J.K. Tsotsos(The first 5 authors, listed alphabetically according to last name, contributed equally). Towards the quantitative evaluation of visual attention models, Vision Research, 116:258-268, 2015. WOS: 000364270400013.

Ø  Jinxia Zhang, Krista A. Ehinger, Jundi Ding, Jingyu Yang. A prior-based graph for salient object detection. ICIP , Paris, France, 27-30 October 2014, P1175-1178. WOS:000370063601069.