A5024015565- Image Enhancement Techniques
- Visual Attention and Saliency Detection
- Color Science and Applications
- Medical Image Segmentation Techniques
- Medical Imaging Techniques and Applications
- Image and Object Detection Techniques
- Image and Video Quality Assessment
- Radiomics and Machine Learning in Medical Imaging
- Advanced Image and Video Retrieval Techniques
- Advanced Image Fusion Techniques
- Advanced Image Processing Techniques
- Digital Image Processing Techniques
- Image and Signal Denoising Methods
Dalhousie University
2013-2018
Lappeenranta-Lahti University of Technology
2017
Northeastern University
2010-2012
Computational visual attention systems detect regions of interest in images. These have a broad range applications areas such as computer vision, computational aesthetics, and non-photorealistic rendering. However, almost all the to date are designed for low dynamic (LDR) images may not be suitable analyzing saliency high (HDR) We propose novel algorithm analysis HDR that is based on virtual photographs. Taking photographs inverse process generating from multiple LDR exposures, photograph...
We present a saliency-based parameter tuning algorithm that can optimize the parameters of tone mapping operators automatically by minimizing saliency distortion caused process mapping. The employs an improved detection model for HDR images, and is quantified as Kullback-Leibler divergence between distributions mapped images those corresponding images. show minimization be accomplished employing evolution strategy with individuals representing settings fitness values based on distortion....
Tone mapping is the process of transforming high dynamic range images for display on low devices. While many tone operators have been proposed, there no single operator that generates optimal results under all conditions. Blending from multiple with varying weights allows leveraging strengths each considered. Prior work has used interactive evolution as a tool blended mapping. In this paper, we build recent progress in development objective quality measures mapped us to automate evolving...
Liver segmentation is a prerequisite for liver cancer CAD, and the result of it affects accuracy rate feature extraction recognition directly. Euclidean distance transformation one main methods in medical image segmentation. However, simply using will lead to some problems, such as over-segmentation excessively high cost calculation. Therefore, this paper puts forward new method segmentation, which combines high-speed complete with statistical analysis. The effectiveness has been verified...
With the development of high-dynamic-range images and tone mapping operators comes a need for image quality evaluation mapped images. However, because significant difference in dynamic range between images, conventional assessment algorithms that predict distortion based on magnitude intensity or normalized contrast are not suitable this task. In article, we present feature-based metric predicts perceived by measuring important features affect judgment. Our utilizes multi-exposed virtual...