Ran Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Ran Liu has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Arun Anand
Sardar Patel University
1 shared publication - Bahram Javidi
University of Connecticut
1 shared publication Dipak K. Dey
Department of Statistics, University of Connecticut, 215 Glenbrook Rd. U-4120, Storrs, Connecticut 06269, USA
1 shared publication- Guanhua Zhou
School of Instrumentation Science and Optoelectronic Engineering, Beihang University, Beijing 100191, China
1 shared publication - Lin Wang1 shared publication
- Masayoshi Takahashi
New Industry Creation Hatchery Center, Tohoku University, 6-6-10 Aoba, Aramaki, Aoba-ku, Sendai 980-8579, Japan
1 shared publication
Latest Publications
- 2 February 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 11 (2) , 572-584
An Ensemble of Classifiers Based on Positive and Unlabeled Data in One-Class Remote Sensing Classification
One-class remote sensing classification refers to the situations when users are only interested in one specific land type without considering other types. The positive and unlabeled learning (PUL) algorithm, which trains a binary classifier from positive and unlabeled data, has been shown to be promising in one-class classification. The implementation of PUL by a single classifier has been investigated. However, implementing PUL using multiple classifiers and creating classifier ensembles based on PUL have not been studied. In this research, we investigate the implementations of PUL using several classifiers, including generalized linear model, generalized additive model, multivariate adaptive regression splines, maximum entropy, backpropagation neural network, and support vector machine, as well as three ensemble methods based on majority vote, weighted average, and weighted vote combination rules. These methods are applied in classifying the urban areas from four remote sensing imagery of different spatial resolutions, including aerial photograph, Landsat 8, WorldView-3, and Gaofen-1. Experimental results show that classifiers can successfully extract the urban areas with high accuracies, and the ensemble methods based on weighted average and weighted vote generally outperform the individual classifiers on different datasets. We conclude that PUL is a promising method in one-class remote sensing classification, and the classifier ensemble based on PUL can significantly improve the accuracy.
- 2 February 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 11 (2) , 572-584
An Ensemble of Classifiers Based on Positive and Unlabeled Data in One-Class Remote Sensing Classification
One-class remote sensing classification refers to the situations when users are only interested in one specific land type without considering other types. The positive and unlabeled learning (PUL) algorithm, which trains a binary classifier from positive and unlabeled data, has been shown to be promising in one-class classification. The implementation of PUL by a single classifier has been investigated. However, implementing PUL using multiple classifiers and creating classifier ensembles based on PUL have not been studied. In this research, we investigate the implementations of PUL using several classifiers, including generalized linear model, generalized additive model, multivariate adaptive regression splines, maximum entropy, backpropagation neural network, and support vector machine, as well as three ensemble methods based on majority vote, weighted average, and weighted vote combination rules. These methods are applied in classifying the urban areas from four remote sensing imagery of different spatial resolutions, including aerial photograph, Landsat 8, WorldView-3, and Gaofen-1. Experimental results show that classifiers can successfully extract the urban areas with high accuracies, and the ensemble methods based on weighted average and weighted vote generally outperform the individual classifiers on different datasets. We conclude that PUL is a promising method in one-class remote sensing classification, and the classifier ensemble based on PUL can significantly improve the accuracy.
- 3 March 2014
- Published by Optica Publishing Group in Journal of the Optical Society of America A
- Vol. 31 (4) , 677-684
Entropy-based clustering of embryonic stem cells using digital holographic microscopy
Embryonic stem (ES) cells are an important factor in the development of cell-based therapeutic strategies. In this work, the use of digital holographic interferometric microscopy and statistical identification for automatic discrimination of ES cells and fibroblast (FB) cells is discussed in detail. The proposed algorithm first reduces the complex data structure to lower dimensions. Then, based on asymptotic normality, model-based clustering and linear discriminant analysis are applied to the transformed data to obtain the classification between ES and FB cells. The proposed algorithm is robust because it does not depend on parametric assumptions and can be extended to the classification of other cell image data. Experimental results are presented to demonstrate the performance of the system.
- 3 March 2014
- Published by Optica Publishing Group in Journal of the Optical Society of America A
- Vol. 31 (4) , 677-684
Entropy-based clustering of embryonic stem cells using digital holographic microscopy
Embryonic stem (ES) cells are an important factor in the development of cell-based therapeutic strategies. In this work, the use of digital holographic interferometric microscopy and statistical identification for automatic discrimination of ES cells and fibroblast (FB) cells is discussed in detail. The proposed algorithm first reduces the complex data structure to lower dimensions. Then, based on asymptotic normality, model-based clustering and linear discriminant analysis are applied to the transformed data to obtain the classification between ES and FB cells. The proposed algorithm is robust because it does not depend on parametric assumptions and can be extended to the classification of other cell image data. Experimental results are presented to demonstrate the performance of the system.
- 15 November 2005
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 5 (21536996) , 3553-3556
A new cross-track radiometric correction method (VRadCor) for airborne hyperspectral image of operational modular imaging spectrometer(OMIS)
- 15 November 2005
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 5 (21536996) , 3553-3556
A new cross-track radiometric correction method (VRadCor) for airborne hyperspectral image of operational modular imaging spectrometer(OMIS)