刁华彬

主要学习经历:

2013.09-2017.06 青岛理工大学,应用物理学,学士

2017.09-2023.01 中国科学院大学&中国科学院微电子研究所,电路与系统,硕博连读,博士

2023.05-至今 安徽工程大学,计算机与信息学院,讲师,硕士生导师

研究方向:机器学习、计算机视觉、模型轻量化、区块链

邮箱diaohuabin@ahpu.edu.cn

 

主持/参与科研项目:

1. 产学研合作项目,面向工业产线的多源异构数据融合及智能分析系统,2024/1-2025/3,在研、主持。

2. 芜湖市科技计划,基于深度学习的机器人视觉分拣关键技术研究,2023/1-2024/12,在研、主持。

3. 国家重点研发计划,优质果蔬智能化品质分级技术装备研发,结题,参与

 

科研论文:

1. Diao H, Li G, Xu S, et al. Self-distillation enhanced adaptive pruning of convolutional neural networks[J]. Pattern Recognition, 2025, 157: 110942.

2. Diao H, Li G, Xu S, et al. Attention Round for post-training quantization[J]. Neurocomputing, 2024, 565: 127012.

3. Diao H, Li G, Xu S, et al. PA-NAS: Partial operation activation for memory-efficient architecture search[J]. Applied Intelligence, 2022, 52(8): 9373-9387.

4. Diao H, Hao Y, Xu S, et al. Implementation of lightweight convolutional neural networks via layer-wise differentiable compression[J]. Sensors, 2021, 21(10): 3464.

5. Wang W, Kong C, Diao H. Path Planning of UGV via Deep Reinforcement Learning with Observation Enhancement[C]//2024 IEEE 2nd International Conference on Control, Electronics and Computer Technology (ICCECT). IEEE, 2024: 1552-1558.

6. Wang W, Kong C, Diao H. Proximal Policy Optimization Based USV Navigation Under Auxiliary Soft Policy Constraints[C]//2024 IEEE 2nd International Conference on Control, Electronics and Computer Technology (ICCECT). IEEE, 2024: 1559-1565.

7. Wang W, Huang S, Diao H. Robust Strategy Generation for Automatic Navigation of Unmanned Surface Vehicles through Improved DDPG Algorithm[J]. Journal of Robotics, 2024, 2024(1): 7328944.

 

发明专利:

1. 史朝坤,刁华彬等;一种基于剪枝轻量化模型的包裹识别方法与系统(ZL 2022 1 0447343.7)