Department of Electrical Engineering

Laboratory of Image Processing, Computer Vision and AI Applications

The "Image Processing, Computer Vision and AI Applications Lab" is engaged in pioneering research within the domains of image processing, computer vision, artificial intelligence (AI) applications, and industrial Internet of Things (IoT).

Established in 1996, this lab has a long-standing history of academic excellence, having trained over one hundred master's and doctoral students. Over the years, the lab has successfully executed numerous government and industry-funded projects, contributing to significant advancements in its fields of expertise. In addition to publishing several well-regarded journal articles, the lab has also been recognized with multiple invention patents.

As one of the key core laboratories of the Department of Electrical Engineering, This lab is committed to advancing the theoretical and practical aspects of image processing and computer vision, with the aim of contributing to significant developments in medical diagnostics, smart manufacturing, and AI-driven innovations. The research conducted here is inherently interdisciplinary, incorporating deep learning, machine learning, and advanced imaging techniques to address complex challenges across both healthcare and industrial contexts.

More specifically, recent research topics include:

• Advanced medical image segmentation and analysis
• Real-time object detection and tracking in dynamic environments
• Development of AI-driven diagnostic tools
• Integration of IoT technologies with computer vision for smart manufacturing
• Novel machine learning algorithms for image classification
• Automated quality inspection systems in industrial settings

The following list represents some of the lab's most significant achievements in recent years.

1. Collaborative Diagnosis in Mixed-Reality using Deep-Learning Networks and RE-WAPICP Algorithm

2. Study of a Deep Convolution Network with Enhanced Region Proposal Network in the Detection of Cancerous Lung Tumors

3. Advancing Barrett's Esophagus Segmentation: A Deep Learning Ensemble Approach with Data Augmentation and Model Collaboration

3. Tongue Image Segmentation and Constitution Identification with Deep Learning

4. Inspection and Classification of Semiconductor Wafer Surface Defects Using CNN Deep Learning Networks

5. Automatic Recognition of Oral drug with Deep Learning

6. Automatic Surgical Instrument Recognition

7. Improving Night Time Driving Safety Using Vision-Based Classification Techniques

8. Enhancing OLED Thin Film Encapsulation: Comparative Analysis of Machine Learning Approaches for Thickness Prediction

9. Image Processing Techniques for Identifying PCB Layout Rules in High-Speed SFI Interfaces