Research Interests:
ABSTRACT This paper presents a fast single-image super-resolution approach that involves learning multiple adaptive interpolation kernels. Based on the assumptions that each high-resolution image patch can be sparsely represented by... more
ABSTRACT This paper presents a fast single-image super-resolution approach that involves learning multiple adaptive interpolation kernels. Based on the assumptions that each high-resolution image patch can be sparsely represented by several simple image structures and that each structure can be assigned a suitable interpolation kernel, our approach consists of the following steps. First, we cluster the training image patches into several classes and train each class-specific interpolation kernel. Then, for each input low-resolution image patch, we select few suitable kernels of it to make up the final interpolation kernel. Since the proposed approach is mainly based on simple linear algebra computations, its efficiency can be guaranteed. And experimental comparisons with state-of-the-art super-resolution reconstruction algorithms on simulated and real-life examples can validate the performance of our proposed approach.
Research Interests:
Active contour model and mean shift are both motion detection algorithms. Each of them has its own merits and shortcomings. An active contour tends to be tracked by noise points and results in a false boundary. A mean shift vector always... more
Active contour model and mean shift are both motion detection algorithms. Each of them has its own merits and shortcomings. An active contour tends to be tracked by noise points and results in a false boundary. A mean shift vector always points to the edge area when the start point is around the object With initial curves given near the
Research Interests:
Subpixel Edge Location Using Improved LSR MO Yi, Zhongxuan Liu, Silong Peng National. ... However, continuous edges are always digitized during the process of imagery digitizing. Practicalline model turns to surface model which lead to... more
Subpixel Edge Location Using Improved LSR MO Yi, Zhongxuan Liu, Silong Peng National. ... However, continuous edges are always digitized during the process of imagery digitizing. Practicalline model turns to surface model which lead to edge discontinuity and zigzag effect. ...
Research Interests:
Research Interests:
Research Interests:
ABSTRACT In this paper, we propose a more practical and accurate method for calibrating the roadside camera used in traffic surveillance systems. Considering the characteristics of the traffic scenes, we propose a minimum calibration... more
ABSTRACT In this paper, we propose a more practical and accurate method for calibrating the roadside camera used in traffic surveillance systems. Considering the characteristics of the traffic scenes, we propose a minimum calibration condition that consists of two vanishing points and a vanishing line, which can be easily satisfied in most traffic scenes. Based on the minimum calibration condition, we provide a calibration method to estimate camera intrinsic parameters and rotation angles, which employs least squares optimization instead of closed-form computation. Compared with the existing calibration methods, our method is suitable for more traffic scenes and is able to accurately determine more camera parameters including the principal point. By making full use of video information, multiple observations of the vanishing points are available from different objects. For more accurate calibration, we present a dynamic calibration method using these observations to correct camera parameters. As for the estimation of the camera translation vector, known lengths in the road or known heights above the road are exploited. The experimental results on synthetic data and real traffic images demonstrate the accuracy, robustness, and practicability of the proposed calibration method.
Research Interests:
Research Interests:
A partitioning-based placement algorithm with priori wirelength estimation called HJ-hPl is presented in this paper. We propose a new methodology to estimate proximity of wirelengths in a netlist, which is capable of estimating not only... more
A partitioning-based placement algorithm with priori wirelength estimation called HJ-hPl is presented in this paper. We propose a new methodology to estimate proximity of wirelengths in a netlist, which is capable of estimating not only short interconnects but long interconnects accurately. We embed the wirelength estimation into the partitioning tool of our global placement, which can guide our placement towards a solution with shorter wirelengths. In addition, we employ a regular structure clustering technique to reduce the size of the original placement, which can also bring on a tighter placement result. Experimental results show that, compared to Capo10.5, mPL6, and NTU place, HJ-hPl outperforms theirs in term of wirelength and run time. The improvements in terms of average wirelength over Capo10.5, mPL6 and NPUplace are 13%, 3%, and 9%with only 19%, 91%, and 99% of their runtime,respectively. By integrating our estimated wirelength driven clustering into Capo10.5, we are able to reduce average wirelength by 3%.
Research Interests:
An analysis of binary mixtures of hydroxyl compound by Attenuated Total Reflection Fourier transform infrared spectroscopy (ATR FT-IR) and classical least squares (CLS) yield large model error due to the presence of unmodeled components... more
An analysis of binary mixtures of hydroxyl compound by Attenuated Total Reflection Fourier transform infrared spectroscopy (ATR FT-IR) and classical least squares (CLS) yield large model error due to the presence of unmodeled components such as H-bonded components. To accommodate these spectral variations, polynomial-based least squares (LSP) and polynomial-based total least squares (TLSP) are proposed to capture the nonlinear absorbance-concentration relationship. LSP is based on assuming that only absorbance noise exists; while TLSP takes both absorbance noise and concentration noise into consideration. In addition, based on different solving strategy, two optimization algorithms (limited-memory Broyden-Fletcher-Goldfarb-Shanno (LBFGS) algorithm and Levenberg-Marquardt (LM) algorithm) are combined with TLSP and then two different TLSP versions (termed as TLSP-LBFGS and TLSP-LM) are formed. The optimum order of each nonlinear model is determined by cross-validation. Comparison and ...
Research Interests:
Research Interests:
... Data and Results Case I: Clyde Platform The real-time hydraulics software was used both in 12¼-in. and 9⅞-in. section. ... References 1. Zamora, M.: Virtual Rheology and Hydraulics Improve Use of Oil and Synthetic-Based Muds, Oil... more
... Data and Results Case I: Clyde Platform The real-time hydraulics software was used both in 12¼-in. and 9⅞-in. section. ... References 1. Zamora, M.: Virtual Rheology and Hydraulics Improve Use of Oil and Synthetic-Based Muds, Oil & Gas Journal (3 Mar 1997) 43. ...
Research Interests:
Research Interests:
In order to eliminate the lower order polynomial interferences, a new quantitative calibration algorithm "Baseline Correction Combined Partial Least Squares (BCC-PLS)",... more
In order to eliminate the lower order polynomial interferences, a new quantitative calibration algorithm "Baseline Correction Combined Partial Least Squares (BCC-PLS)", which combines baseline correction and conventional PLS, is proposed. By embedding baseline correction constraints into PLS weights selection, the proposed calibration algorithm overcomes the uncertainty in baseline correction and can meet the requirement of on-line attenuated total reflectance Fourier transform infrared (ATR-FTIR) quantitative analysis. The effectiveness of the algorithm is evaluated by the analysis of glucose and marzipan ATR-FTIR spectra. BCC-PLS algorithm shows improved prediction performance over PLS. The root mean square error of cross-validation (RMSECV) on marzipan spectra for the prediction of the moisture is found to be 0.53%, w/w (range 7-19%). The sugar content is predicted with a RMSECV of 2.04%, w/w (range 33-68%).
Research Interests: Analytical Chemistry, Prediction, Algorithm, Quantitative analysis, Uncertainty, and 12 moreMean square error, ROOT, Calibration, Partial Least Squares, Selection, Cross Validation, Embedding, Application, Interference, Fourier Transform spectroscopy, Quantitative Analysis, and Infrared Spectrometry
Research Interests:
This paper presents a wavelet-domain Hidden Markov Tree(HMT)-based color image superresolution algorithm using multi-channel data fusion. Because there exists correlations among the three channels of a RGB color image, a channel by... more
This paper presents a wavelet-domain Hidden Markov Tree(HMT)-based color image superresolution algorithm using multi-channel data fusion. Because there exists correlations among the three channels of a RGB color image, a channel by channel superresolution method almost certain leads to color distortion. In order to solve this problem, first the low-resolution color image is converted into a gray-scale image using the
Research Interests:
A calibration transfer method for near-infrared (NIR) spectra based on spectral regression is proposed. Spectral regression method can reveal low dimensional manifold structure in high dimensional spectroscopic data and is suitable to... more
A calibration transfer method for near-infrared (NIR) spectra based on spectral regression is proposed. Spectral regression method can reveal low dimensional manifold structure in high dimensional spectroscopic data and is suitable to transfer the NIR spectra of different instruments. A comparative study of the proposed method and piecewise direct standardization (PDS) for standardization on two benchmark NIR data sets is presented. Experimental results show that spectral regression method outperforms PDS and is quite competitive with PDS with background correction. When the standardization subset has sufficient samples, spectral regression method exhibits excellent performance.
Research Interests:
A low power reconfigurable DCT architecture is proposed, which can be run at three transform precision levels for different demands. Using the character of energy distribution of the DCT matrix after 2D DCT operation, we selected the best... more
A low power reconfigurable DCT architecture is proposed, which can be run at three transform precision levels for different demands. Using the character of energy distribution of the DCT matrix after 2D DCT operation, we selected the best DCT bases which achieve considerable power reduction in DCT operation with minimum image quality degradation. The reconfigurable architecture can achieve power saving
Research Interests:
A partitioning-based placement algorithm with priori wirelength estimation called HJ-hPl is presented in this paper. We propose a new methodology to estimate proximity of wirelengths in a netlist, which is capable of estimating not only... more
A partitioning-based placement algorithm with priori wirelength estimation called HJ-hPl is presented in this paper. We propose a new methodology to estimate proximity of wirelengths in a netlist, which is capable of estimating not only short interconnects but long interconnects accurately. We embed the wirelength estimation into the partitioning tool of our global placement, which can guide our placement towards a solution with shorter wirelengths. In addition, we employ a regular structure clustering technique to reduce the size of the original placement, which can also bring on a tighter placement result. Experimental results show that, compared to Capo10.5, mPL6, and NTU place, HJ-hPl outperforms theirs in term of wirelength and run time. The improvements in terms of average wirelength over Capo10.5, mPL6 and NPUplace are 13%, 3%, and 9%with only 19%, 91%, and 99% of their runtime,respectively. By integrating our estimated wirelength driven clustering into Capo10.5, we are able to reduce average wirelength by 3%.
Research Interests:
Abstract This paper presents a novel framework based on three-dimensional (3D) grid graph for automatic mosaic construction of multilayer microscopic images. Firstly the multilayer structure is divided into several slices, and... more
Abstract This paper presents a novel framework based on three-dimensional (3D) grid graph for automatic mosaic construction of multilayer microscopic images. Firstly the multilayer structure is divided into several slices, and two-dimensional (2D) scanning manner is ...
Research Interests:
This paper presents a method to deals with two types of alignment error for construction of mosaics, which are built from large-scale microscopic images. The type I error is defined as the difference between the mapping model used for the... more
This paper presents a method to deals with two types of alignment error for construction of mosaics, which are built from large-scale microscopic images. The type I error is defined as the difference between the mapping model used for the alignment and the actual between-image geometric distortion, and the type II error is defined as the erroneous alignment due to mismatch. Firstly, a global alignment model based on two-dimensional (2D) grid graph is proposed to eliminate error accumulation induced by type I error. Secondly, characteristic of global alignment error caused by type II error is analyzed. Finally, a minimum cycle method is proposed to eliminate the type II error. Iteratively solving the global alignment model leads to a global consistently mosaics for large-scale microscopic images
Research Interests:
We present a new receiver scheme, termed aswavelet receiver, for doubly-selective channels to combat the annoying Doppler effect. The key point is to convert the Doppler effect to Doppler diversity, taking advantage of the diversity... more
We present a new receiver scheme, termed aswavelet receiver, for doubly-selective channels to combat the annoying Doppler effect. The key point is to convert the Doppler effect to Doppler diversity, taking advantage of the diversity technique to improve system performance. To this end, a new framework based on multiresolution analysis (MRA) is established. In this framework, we find that RAKE
Research Interests:
The size of training set as well as the usage thereof is an important issue of learning-based super-resolution. In this paper, we presented an adaptive learning method for face hallucination using locality preserving projections (LPP). By... more
The size of training set as well as the usage thereof is an important issue of learning-based super-resolution. In this paper, we presented an adaptive learning method for face hallucination using locality preserving projections (LPP). By virtue of the ability to reveal the non-linear structure hidden in the high-dimensional image space, LPP is an efficient manifold learning method to analyze
Research Interests:
Recently, there has ken a growing interest in investigating signal-adaptive multirate filter banks. In this paper, we propose a direct method of constructing signal-adaptive orthogonal wavelet filter banks with better energy compaction... more
Recently, there has ken a growing interest in investigating signal-adaptive multirate filter banks. In this paper, we propose a direct method of constructing signal-adaptive orthogonal wavelet filter banks with better energy compaction performance than that of Daubechies wavelet ...
Research Interests:
... National ASIC Design Engineering Center Institute of Automation, Chinese Academy of Sciences Beijing 100080-2728, China { tao.chen}{ruosan.guo}{silong.peng} @mail.ia ... A novel algorithm based on multiscale first fun-damental form... more
... National ASIC Design Engineering Center Institute of Automation, Chinese Academy of Sciences Beijing 100080-2728, China { tao.chen}{ruosan.guo}{silong.peng} @mail.ia ... A novel algorithm based on multiscale first fun-damental form was presented by Scheunders et al in (41 ...
Research Interests:
Abstract In this paper the technique of directional empirical mode decomposition (DEMD) and its application to texture segmentation are presented. Empirical mode decomposition (EMD) decomposes signals by sifting and then analyzes the... more
Abstract In this paper the technique of directional empirical mode decomposition (DEMD) and its application to texture segmentation are presented. Empirical mode decomposition (EMD) decomposes signals by sifting and then analyzes the instantaneous frequency of ...