Mr . Huanyu Li | Underwater Image Captioning | Best Researcher Award

Huanyu Li is currently pursuing a Doctoral degree (Ph.D.) at the China University of Petroleum (East China), majoring in Marine Resources and Information Engineering. His academic journey is rooted in a robust interdisciplinary curriculum that combines the fundamentals of marine sciences with advanced computational techniques. As a Ph.D. candidate, he has undertaken extensive coursework and research in areas such as underwater image processing, computer vision, deep learning, and semantic modeling. His education has been enriched through the development of research articles, participation in international academic conferences, and collaborative work with both faculty and fellow researchers. The rigorous academic environment at the China University of Petroleum has enabled him to explore emerging technologies, apply AI in marine domains, and develop innovative solutions for complex underwater imaging problems. This strong educational background has equipped him with both the theoretical knowledge and technical skillset necessary to lead research at the intersection of AI and oceanography.

Although still in the doctoral phase of his academic career, Huanyu Li has gained significant professional experience through active participation in research projects, publication of scholarly articles, and peer-review activities. He has co-authored multiple research papers in collaboration with established academics and contributed to projects involving underwater image captioning, convolutional neural networks, and AI-based image analysis. His work spans across both journal articles and conference proceedings, where he has served as a presenting author and co-investigator. Additionally, Huanyu has contributed to practical applications such as real-time seed sorting systems, lightweight model deployment, and image enhancement techniques. He has also served as a reviewer for high-impact journals including IEEE Transactions on Geoscience and Remote Sensing, Pattern Recognition, and Neural Computing and Applications, indicating his integration into the broader scientific community. These professional activities have provided him with a rich experience in experimental design, algorithm development, and scholarly communication.

Huanyu Li’s research interests lie at the intersection of artificial intelligence and marine environmental analysis. His primary focus is on underwater image intelligent processing, where he investigates techniques to enhance image quality, recognition, and interpretation in complex underwater environments. He is deeply engaged in image-semantic cross-modal learning, aiming to develop models that bridge visual content with textual understanding, such as underwater image captioning. Another core area of interest is deep learning model compression, where he explores methods to reduce model size and computational requirements without sacrificing performance—enabling real-time processing in resource-constrained scenarios. His research also touches on underwater acoustic communication and attention-based fusion techniques. These interests demonstrate a multidisciplinary approach that combines computer vision, machine learning, signal processing, and marine science. By leveraging AI-driven innovations, he aims to solve real-world challenges in ocean exploration, ecological monitoring, and underwater robotics.

While specific awards and honors are not detailed in the current profile, Huanyu Li’s accomplishments are evident through the recognition he has received from the academic community. His research has been published in leading international journals such as ISPRS Journal of Photogrammetry and Remote Sensing, Information Fusion, and IEEE Transactions on Geoscience and Remote Sensing—a significant achievement for a doctoral researcher. Furthermore, his invited participation in major conferences and peer-review roles for journals like Pattern Recognition and Neural Networks suggest high regard among his peers. These roles are typically extended to researchers with proven expertise, highlighting his growing reputation. His innovative work on underwater image captioning and real-time image processing systems positions him as a strong candidate for future academic awards, fellowships, and research grants. His academic service, publication record, and practical contributions reflect an honors-worthy trajectory within the field of intelligent vision systems and marine computing.

Huanyu Li possesses a comprehensive and evolving research skill set tailored to modern challenges in computer vision and underwater imaging. His core competencies include deep learning, particularly convolutional neural networks (CNNs), model pruning, and cross-modal learning. He is proficient in developing and optimizing AI algorithms for image captioning, classification, enhancement, and compression. His skills extend to semantic understanding of images, where he connects visual data to textual outputs, enhancing interpretability. Huanyu is also experienced in accelerated computing, using techniques like TensorRT for deploying real-time systems. Additionally, he has practical expertise in entropy-based optimization and 2D information theory for CNN design. His work includes implementing visual attention mechanisms and designing lightweight models for efficient deployment. With a hands-on approach to both algorithmic innovation and applied engineering, Huanyu’s research skills align with the needs of both academic exploration and real-world application in underwater exploration and smart marine systems.

Huanyu Li is a highly promising and technically proficient early-career researcher whose contributions in underwater image processing and intelligent systems are both innovative and well-published. His strong academic output, interdisciplinary research, and peer recognition make him a strong candidate for the Best Researcher Award, particularly in the emerging researcher or early-career category.

To further elevate his candidacy for broader or senior-level recognition, increasing leadership visibility, international collaboration, and showcasing real-world impact would be beneficial in the future.

• Title: Underwater Image Captioning via Attention Mechanism Based Fusion of Visual and Textual Information
  Authors: Li Li, Huanyu Li, Peng Ren
  Year: 2025
  Citation: DOI: 10.1016/j.inffus.2025.103269

• Title: Underwater Image Captioning: Challenges, Models, and Datasets
  Authors: Huanyu Li, Hao Wang, Ying Zhang, Li Li, Peng Ren
  Year: 2025
  Citation: DOI: 10.1016/j.isprsjprs.2024.12.002

• Title: An Underwater Acoustic Semantic Communication Approach to Underwater Image Transmission
  Authors: Ying Zhang, Huanyu Li, Bingyu Li, Li Li, Weibo Zhang, Hao Wang, Peng Ren
  Year: 2025
  Citation: DOI: 10.1007/s44295-025-00054-7

• Title: A Real-Time and High-Performance MobileNet Accelerator Based on Adaptive Dataflow Scheduling for Image Classification
  Authors: Xiaoting Sang, Tianyi Ruan, Chunlei Li, Huanyu Li, Rui Yang, Zhen Liu
  Year: 2024
  Citation: DOI: 10.1007/s11554-023-01378-5

• Title: INSPIRATION: A Reinforcement Learning-Based Human Visual Perception-Driven Image Enhancement Paradigm for Underwater Scenes
  Authors: Hao Wang, Shuhan Sun, Linlin Chang, Huanyu Li, Weibo Zhang, Alejandro C. Frery, Peng Ren
  Year: 2024
  Citation: DOI: 10.1016/j.engappai.2024.108411

• Title: A Real-Time and High-Performance MobileNet Accelerator Based on Adaptive Dataflow Scheduling for Image Classification (Preprint)
  Authors: Xiaoting Sang, Tianyi Ruan, Chunlei Li, Huanyu Li, Rui Yang, Zhen Liu
  Year: 2023
  Citation: DOI: 10.21203/rs.3.rs-3132056

• Title: An Accelerating Convolutional Neural Networks via a 2D Entropy Based-Adaptive Filter Search Method for Image Recognition
  Authors: Chunlei Li, Huanyu Li, Guangshuai Gao, Zhen Liu, Peng Liu
  Year: 2023
  Citation: DOI: 10.1016/j.asoc.2023.110326

• Title: Real-Time Seed Sorting System via 2D Information Entropy-Based CNN Pruning and TensorRT Acceleration
  Authors: Chunlei Li, Huanyu Li, Liang Liao, Zhen Liu, Yizhou Dong
  Year: 2023
  Citation: DOI: 10.1049/ipr2.12747

• Title: A Block-Based and Highly Parallel CNN Accelerator for Seed Sorting
  Authors: Xiaoting Sang, Zhenghui Hu, Huanyu Li, Chunlei Li, Zhen Liu
  Year: 2022
  Citation: DOI: 10.1155/2022/5608573

• Title: Rapid and High-Purity Seed Grading Based on Pruned Deep Convolutional Neural Network
  Authors: Huanyu Li, Cuicao Zhang, Chunlei Li, Zhen Liu, Yizhou Dong, Shuhang Tang
  Year: 2022
  Citation: DOI: 10.1007/978-3-031-02444-3_8

• Title: AMDet: An Efficient Infrared Small Object Detection Model Based on Visual Attention and Multi-Dilation Feature
  Authors: Cuicao Zhang, Yizhou Dong, Huanyu Li, Chunlei Li, Zhen Liu
  Year: 2021
  Citation: DOI: 10.1145/3497623.3497644

• Title: SeedSortNet: A Rapid and Highly Efficient Lightweight CNN Based on Visual Attention for Seed Sorting
  Authors: Chunlei Li, Huanyu Li, Zhen Liu, Bing Li, Yong Huang
  Year: 2021
  Citation: DOI: 10.7717/peerj-cs.639