Fang Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Fang Liu has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
- Tianjiang Wang
Huazhong University of Science and Technology
6 shared publications - Gwangmin Choe
Huazhong University of Science and Technology
4 shared publications Chunhwa Choe
Kim IL Sung University
2 shared publications- Chunyan Xu
Nanjing University of Science and Technology, Nanjing, China
2 shared publications - Songfeng Lu
Huazhong University of Science and Technology
2 shared publications - Aaqif Afzaal Abbasi
Department of Earth and Marine Sciences, University of Palermo, Palermo, Italy
1 shared publication
Latest Publications
- Open Access
- 13 December 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 7, 181405-181422
A Multilevel Image Thresholding Based on Hybrid Salp Swarm Algorithm and Fuzzy Entropy
The image segmentation techniques based on multi-level threshold value received lot of attention in recent years. It is because they can be used as a pre-processing step in complex image processing applications. The main problem in identifying the suitable threshold values occurs when classical image segmentation methods are employed. The swarm intelligence (SI) technique is used to improve multi-level threshold image (MTI) segmentation performance. SI technique simulates the social behaviors of swarm ecosystem, such as the behavior exhibited by different birds, animals etc. Based on SI techniques, we developed an alternative MTI segmentation method by using a modified version of the salp swarm algorithm (SSA). The modified algorithm improves the performance of various operators of the moth-flame optimization (MFO) algorithm to address the limitations of traditional SSA algorithm. This results in improved performance of SSA algorithm. In addition, the fuzzy entropy is used as objective function to determine the quality of the solutions. To evaluate the performance of the proposed methodology, we evaluated our techniques on CEC2005 benchmark and Berkeley dataset. Our evaluation results demonstrate that SSAMFO outperforms traditional SSA and MFO algorithms, in terms of PSNR, SSIM and fitness value.
- 13 December 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 7, 181405-181422
Open AccessA Multilevel Image Thresholding Based on Hybrid Salp Swarm Algorithm and Fuzzy Entropy
The image segmentation techniques based on multi-level threshold value received lot of attention in recent years. It is because they can be used as a pre-processing step in complex image processing applications. The main problem in identifying the suitable threshold values occurs when classical image segmentation methods are employed. The swarm intelligence (SI) technique is used to improve multi-level threshold image (MTI) segmentation performance. SI technique simulates the social behaviors of swarm ecosystem, such as the behavior exhibited by different birds, animals etc. Based on SI techniques, we developed an alternative MTI segmentation method by using a modified version of the salp swarm algorithm (SSA). The modified algorithm improves the performance of various operators of the moth-flame optimization (MFO) algorithm to address the limitations of traditional SSA algorithm. This results in improved performance of SSA algorithm. In addition, the fuzzy entropy is used as objective function to determine the quality of the solutions. To evaluate the performance of the proposed methodology, we evaluated our techniques on CEC2005 benchmark and Berkeley dataset. Our evaluation results demonstrate that SSAMFO outperforms traditional SSA and MFO algorithms, in terms of PSNR, SSIM and fitness value.
- Open Access
- 5 April 2019
- Published by Frontiers Media SA in Frontiers in Neuroscience
- Vol. 13, 299
Quantitative Research of 11C-CFT and 18F-FDG PET in Parkinson’s Disease: A Pilot Study With NeuroQ Software
Dopamine transporter (DAT) and glucose metabolism imaging have been applied in the diagnosis of Parkinson’s disease (PD). We explored the possibility of evaluating for PD with NeuroQ software by analyzing 11C-2β-carbomethoxy-3β-(4-fluorophenyl) tropane (11C-CFT) and 18F-FDG PET/CT. We retrospectively analyzed brain 11C-CFT and 18F-FDG PET/CT of 38 patients with parkinsonism, including 20 with PD, 10 with multiple system atrophy (MSA) and 8 with essential tremor (ET), and compared them with the PET/CT of 11 normal healthy controls (NC). PD patients were divided into mild and moderate-severe grade according to the Hoehn-Yahr (H&Y) scale. The 11C-CFT uptake in the caudate nuclei (CN) and putamen (Pu) normalized with cerebellum (CN/Cb and Pu/Cb) were obtained with a manual method and NeuroQ software, and their diagnostic performance was compared.18F-FDG uptake of specific regions was also obtained with NeuroQ, and the enhancement effect for the differential diagnosis was evaluated. There was significant agreement between the manual method and the NeuroQ method for 11C-CFT uptake by CN (r2= 0.680) and Pu (r2= 0.770). 11C-CFT uptake by CN and Pu in PD and MSA patients was significantly lower compared to NC and ET patients. The cutoffs of CN/Cb and Pu/Cb for the distinction between PD and NC were 1.71 and 2.20, respectively. No difference in uptake ratios occurred between PD and MSA. 18F-FDG uptake by the pons and cerebellum in the MSA group was markedly decreased. It was highly accurate in distinguishing between PD and MSA when combined with analysis of 11C-CFT uptake. Pu/Cb decreased significantly in mild grade PD compared to NC group (1.92 ± 0.33 vs. 2.82 ± 0.43); however no statistically significant decrease in CN/Cb was observed until moderate-severe grade PD (1.43 ± 0.11 vs. 2.23 ± 0.36). In early asymmetric PD, a statistically significant difference could be seen with Pu/Cb between the symptomatic and asymptomatic side (2.17 ± 0.30 vs. 1.95 ± 0.22). 11C-CFT and 18F-FDG PET/CT can be analyzed quantitatively with NeuroQ software, which provides an accurate method for the diagnosis and severity evaluation of PD.
- 5 April 2019
- Published by Frontiers Media SA in Frontiers in Neuroscience
- Vol. 13, 299
Open AccessQuantitative Research of 11C-CFT and 18F-FDG PET in Parkinson’s Disease: A Pilot Study With NeuroQ Software
Dopamine transporter (DAT) and glucose metabolism imaging have been applied in the diagnosis of Parkinson’s disease (PD). We explored the possibility of evaluating for PD with NeuroQ software by analyzing 11C-2β-carbomethoxy-3β-(4-fluorophenyl) tropane (11C-CFT) and 18F-FDG PET/CT. We retrospectively analyzed brain 11C-CFT and 18F-FDG PET/CT of 38 patients with parkinsonism, including 20 with PD, 10 with multiple system atrophy (MSA) and 8 with essential tremor (ET), and compared them with the PET/CT of 11 normal healthy controls (NC). PD patients were divided into mild and moderate-severe grade according to the Hoehn-Yahr (H&Y) scale. The 11C-CFT uptake in the caudate nuclei (CN) and putamen (Pu) normalized with cerebellum (CN/Cb and Pu/Cb) were obtained with a manual method and NeuroQ software, and their diagnostic performance was compared.18F-FDG uptake of specific regions was also obtained with NeuroQ, and the enhancement effect for the differential diagnosis was evaluated. There was significant agreement between the manual method and the NeuroQ method for 11C-CFT uptake by CN (r2= 0.680) and Pu (r2= 0.770). 11C-CFT uptake by CN and Pu in PD and MSA patients was significantly lower compared to NC and ET patients. The cutoffs of CN/Cb and Pu/Cb for the distinction between PD and NC were 1.71 and 2.20, respectively. No difference in uptake ratios occurred between PD and MSA. 18F-FDG uptake by the pons and cerebellum in the MSA group was markedly decreased. It was highly accurate in distinguishing between PD and MSA when combined with analysis of 11C-CFT uptake. Pu/Cb decreased significantly in mild grade PD compared to NC group (1.92 ± 0.33 vs. 2.82 ± 0.43); however no statistically significant decrease in CN/Cb was observed until moderate-severe grade PD (1.43 ± 0.11 vs. 2.23 ± 0.36). In early asymmetric PD, a statistically significant difference could be seen with Pu/Cb between the symptomatic and asymptomatic side (2.17 ± 0.30 vs. 1.95 ± 0.22). 11C-CFT and 18F-FDG PET/CT can be analyzed quantitatively with NeuroQ software, which provides an accurate method for the diagnosis and severity evaluation of PD.
- 1 September 2018
- Published by Elsevier in Neurocomputing
- Vol. 308, 245-254
A deep features based generative model for visual tracking
- 1 September 2018
- Published by Elsevier in Neurocomputing
- Vol. 308, 245-254
A deep features based generative model for visual tracking
- 15 October 2014
- Published by Zhejiang University Press in Journal of Zhejiang University SCIENCE C
- Vol. 15 (10) , 861-877
Anadvanced integrated framework for moving object tracking
This paper first introduces the concept of a geogram that captures richer features to represent the objects. The spatiogram contains some moments upon the coordinates of the pixels corresponding to each bin, while the geogram contains information about the perimeter of grouped regions in addition to features in the spatiogram. Then we consider that a convergence process of mean shift is divided into obvious dynamic and steady states, and introduce a hybrid technique of feature description, to control the convergence process. Also, we propose a spline resampling to control the balance between computational cost and accuracy of particle filtering. Finally, we propose a boosting-refining approach, which is boosting the particles positioned in the ill-posed condition instead of eliminating the ill-posed particles, to refine the particles. It enables the estimation of the object state to obtain high accuracy. Experimental results show that our approach has promising discriminative capability in comparison with the state-of-the-art approaches.
- 15 October 2014
- Published by Zhejiang University Press in Journal of Zhejiang University SCIENCE C
- Vol. 15 (10) , 861-877
Anadvanced integrated framework for moving object tracking
This paper first introduces the concept of a geogram that captures richer features to represent the objects. The spatiogram contains some moments upon the coordinates of the pixels corresponding to each bin, while the geogram contains information about the perimeter of grouped regions in addition to features in the spatiogram. Then we consider that a convergence process of mean shift is divided into obvious dynamic and steady states, and introduce a hybrid technique of feature description, to control the convergence process. Also, we propose a spline resampling to control the balance between computational cost and accuracy of particle filtering. Finally, we propose a boosting-refining approach, which is boosting the particles positioned in the ill-posed condition instead of eliminating the ill-posed particles, to refine the particles. It enables the estimation of the object state to obtain high accuracy. Experimental results show that our approach has promising discriminative capability in comparison with the state-of-the-art approaches.
- 26 June 2014
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 74 (21) , 9771-9794
Moving object tracking based on geogram
- 26 June 2014
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 74 (21) , 9771-9794
Moving object tracking based on geogram
- 23 May 2014
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 74 (17) , 7195-7220
Visual tracking based on particle filter with spline resampling
- 23 May 2014
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 74 (17) , 7195-7220
Visual tracking based on particle filter with spline resampling
- 22 May 2014
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 74 (18) , 7595-7619
An advanced association of particle filtering and kernel based object tracking
- 22 May 2014
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 74 (18) , 7595-7619
An advanced association of particle filtering and kernel based object tracking
- 10 October 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 25 (4) , 728-737
An Ordered-Patch-Based Image Classification Approach on the Image Grassmannian Manifold
This paper presents an ordered-patch-based image classification framework integrating the image Grassmannian manifold to address handwritten digit recognition, face recognition, and scene recognition problems. Typical image classification methods explore image appearances without considering the spatial causality among distinctive domains in an image. To address the issue, we introduce an ordered-patch-based image representation and use the autoregressive moving average (ARMA) model to characterize the representation. First, each image is encoded as a sequence of ordered patches, integrating both the local appearance information and spatial relationships of the image. Second, the sequence of these ordered patches is described by an ARMA model, which can be further identified as a point on the image Grassmannian manifold. Then, image classification can be conducted on such a manifold under this manifold representation. Furthermore, an appropriate Grassmannian kernel for support vector machine classification is developed based on a distance metric of the image Grassmannian manifold. Finally, the experiments are conducted on several image data sets to demonstrate that the proposed algorithm outperforms other existing image classification methods.
- 10 October 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 25 (4) , 728-737
An Ordered-Patch-Based Image Classification Approach on the Image Grassmannian Manifold
This paper presents an ordered-patch-based image classification framework integrating the image Grassmannian manifold to address handwritten digit recognition, face recognition, and scene recognition problems. Typical image classification methods explore image appearances without considering the spatial causality among distinctive domains in an image. To address the issue, we introduce an ordered-patch-based image representation and use the autoregressive moving average (ARMA) model to characterize the representation. First, each image is encoded as a sequence of ordered patches, integrating both the local appearance information and spatial relationships of the image. Second, the sequence of these ordered patches is described by an ARMA model, which can be further identified as a point on the image Grassmannian manifold. Then, image classification can be conducted on such a manifold under this manifold representation. Furthermore, an appropriate Grassmannian kernel for support vector machine classification is developed based on a distance metric of the image Grassmannian manifold. Finally, the experiments are conducted on several image data sets to demonstrate that the proposed algorithm outperforms other existing image classification methods.
- 8 June 2013
- Published by Springer Nature in Quantum Information Processing
- Vol. 12 (10) , 3265-3274
An efficient quantum search engine on unsorted database
- 8 June 2013
- Published by Springer Nature in Quantum Information Processing
- Vol. 12 (10) , 3265-3274
An efficient quantum search engine on unsorted database