Nilanjan Ray, PhD
Professor, Faculty of Science - Computing Science
406 Athabasca Hall
9119 - 116 St NWEdmonton ABT6G 2E8
Area of Study / Keywords
Computer vision Image analysis Medical image processing Deep learning
I joined Computing Science, University of Alberta as an Assistant Professor in 2006. I was promoted to Associate Professor in 2013.
- B.Engg., Mechanical Egnineering, Jadavpur University, India, 1995
- M.Tech., Computer Science, Indian Statistical Institute, India, 1997
- Ph.D., Electrical Engineering, University of Virginia, USA, 2003
- IEEE Transactions on Image Processing (2013-2017)
- IET Image Processing (2016-2022)
- 2017 AI/GI/CRV Conference, Edmonton
- Scott T. Acton and Nilanjan Ray, Biomedical image analysis: Tracking. Morgan & Claypool Publishers, 2006.
- Scott T. Acton and Nilanjan Ray, Biomedical image analysis: Segmentation. Morgan & Claypool Publishers, 2009.
Computer vision and deep learning; image and video analysis, medical image processing
Medical image and video analysis and general computer vision problems including segmentation, registration, object detection and classification.
- Pedestrian and vehicle counting
- Microscopy image and video segmentation and registration
- Breast cancer histopathology image analysis
- Oilsand slurry image and video analysis
- Non-invasive blood pressure monitoring
- Brain tumor localization
An introduction to basic digital image processing theory, and the tools that make advanced image manipulation possible for ordinary users. Image processing is important in many applications: editing and processing photographs, special effects for movies, drawing animated characters starting with photographs, analyzing and enhancing remote imagery, and detecting suspects from surveillance cameras. Image processing concepts are introduced using tools like Photoshop and GIMP. Exposure to simple image processing programming with Java and MATLAB. This course is preparation for more advanced courses in the Digital Media area. Prerequisites: Any introductory-level Computing Science course, plus knowledge of introductory-level MATH and STAT; or consent of the instructor or SCI 100. Open to students in the Faculty of Arts, Engineering and Sciences; others require consent of the instructor.
Introduction to visual recognition to recognize objects and classify scenes or images automatically by a computer. Supervised and unsupervised machine learning principles and deep learning techniques will be utilized for visual recognition. Successful commercial systems based on visual recognition range from entertainment to serious scientific research: face detection and recognition on personal devices, social media. Prerequisites: CMPUT 115 or 175; one of MATH 100, 113, 114, 117, 134, 144, 154; MATH 125; STAT 141, 151 or 235.
A major essay on an agreed topic.
Quality assessment of image and video (or 3D data) is essential in many applications, which deliver educational content, medical images, games, movies, video-on-demand and so on. In order to generate high quality image and video, especially given the sheer volume of consumer demand and under constrained resources, e.g., time and bandwidth, it is necessary to understand the image and video processing pipeline from the initial creation limitations to the final display at the receiver. This course focuses on reviewing various image/video processing techniques, as well as the quality assessment metrics proposed in the literature. Sections offered at an increased rate of fee assessment; refer to the Tuition and Fees page in the University Regulations sections of the Calendar.