Image Generation Method Based on Image Edge Information

Yanqiu Liu,Xiuhui Wang 보안공학연구지원센터 2016 International Journal of Multimedia and Ubiquitous Vol.11 No.4

An algorithm was proposed to generate automatically embedded image. With the use of outline matching and shape deformation, the proposed method can insert an interior object with partial deformation into an outer object image, having better results by considering the cavity of the outer object. The embedded object is easily observed because its color is opposite to the outer object while the outer object’s cavity can be part of the interior object. By designing method which is applied to express object’s edge strength, the matching result of the algorithm can be optimized.

Gait Recognition Algorithm Based on Feature Fusion of GEI Dynamic Region and Gabor Wavelets

( Jun Huang ),( Xiuhui Wang ),( Jun Wang ) 한국정보처리학회 2018 Journal of information processing systems Vol.14 No.4

The paper proposes a novel gait recognition algorithm based on feature fusion of gait energy image (GEI) dynamic region and Gabor, which consists of four steps. First, the gait contour images are extracted through the object detection, binarization and morphological process. Secondly, features of GEI at different angles and Gabor features with multiple orientations are extracted from the dynamic part of GEI, respectively. Then averaging method is adopted to fuse features of GEI dynamic region with features of Gabor wavelets on feature layer and the feature space dimension is reduced by an improved Kernel Principal Component Analysis (KPCA). Finally, the vectors of feature fusion are input into the support vector machine (SVM) based on multi classification to realize the classification and recognition of gait. The primary contributions of the paper are: a novel gait recognition algorithm based on based on feature fusion of GEI and Gabor is proposed; an improved KPCA method is used to reduce the feature matrix dimension; a SVM is employed to identify the gait sequences. The experimental results suggest that the proposed algorithm yields over 90% of correct classification rate, which testify that the method can identify better different human gait and get better recognized effect than other existing algorithms.

Laminar flow and chaotic advection mixing performance in a static mixer with perforated helical segments

Huibo Meng,Xiuhui Jiang,Yanfang Yu,Zongyong Wang,Jianhua Wu 한국화학공학회 2017 Korean Journal of Chemical Engineering Vol.34 No.5

The laminar flow and chaotic mixing characteristics of a high-viscosity fluid in static mixers with staggered perforated helical segments were numerically investigated in the range of Re=0.1-150. The numerical results of pressure drop of Kenics static mixer have a good agreement with the reported data from the literature. The effects of aspect ratio Ar and Reynolds number on the mixing performance of Modified Kenics Static Mixers (MKSM) were evaluated by Darcy friction coefficient, shear rate, stretching rate, and Lyapunov exponent, respectively. The product of f×Re for MKSM linearly increased with the increase of Re, but it was constant under Re<10. The values of shear rate in the first perforated hole of mixing elements gradually became much larger by 1.10%-11.78% than those in the second perforated hole with the increasing Re. With the increase of dimensionless axial mixing length, the stretching rate increased linearly and the sensitivity for initial condition gradually weakened. A larger Ar is beneficial for micro-mixing in creeping flow. The average Lyapunov exponent linearly increases with the increase of Re. The profiles of Lyapunov exponent at different dimensionless perforated diameter (d/W) and perforated spacing (s/W) indicate that the chaotic mixing in MKSM is much more sensitive to d/W than s/W. A dimensionless parameter η taking into account the mixing degree and pressure drop was employed to evaluate the mixing efficiency. The optimization of perforated helical segments with the highest mixing efficiency at Re=100 was d/W=0.55 and s/W=1.2.

Gait Recognition Algorithm Based on Feature Fusion of GEI Dynamic Region and Gabor Wavelets

Huang, Jun,Wang, Xiuhui,Wang, Jun Korea Information Processing Society 2018 Journal of information processing systems Vol.14 No.4

The paper proposes a novel gait recognition algorithm based on feature fusion of gait energy image (GEI) dynamic region and Gabor, which consists of four steps. First, the gait contour images are extracted through the object detection, binarization and morphological process. Secondly, features of GEI at different angles and Gabor features with multiple orientations are extracted from the dynamic part of GEI, respectively. Then averaging method is adopted to fuse features of GEI dynamic region with features of Gabor wavelets on feature layer and the feature space dimension is reduced by an improved Kernel Principal Component Analysis (KPCA). Finally, the vectors of feature fusion are input into the support vector machine (SVM) based on multi classification to realize the classification and recognition of gait. The primary contributions of the paper are: a novel gait recognition algorithm based on based on feature fusion of GEI and Gabor is proposed; an improved KPCA method is used to reduce the feature matrix dimension; a SVM is employed to identify the gait sequences. The experimental results suggest that the proposed algorithm yields over 90% of correct classification rate, which testify that the method can identify better different human gait and get better recognized effect than other existing algorithms.

Kernel Fisher Discriminant Analysis for Natural Gait Cycle Based Gait Recognition

Huang, Jun,Wang, Xiuhui,Wang, Jun Korea Information Processing Society 2019 Journal of information processing systems Vol.15 No.4

This paper studies a novel approach to natural gait cycles based gait recognition via kernel Fisher discriminant analysis (KFDA), which can effectively calculate the features from gait sequences and accelerate the recognition process. The proposed approach firstly extracts the gait silhouettes through moving object detection and segmentation from each gait videos. Secondly, gait energy images (GEIs) are calculated for each gait videos, and used as gait features. Thirdly, KFDA method is used to refine the extracted gait features, and low-dimensional feature vectors for each gait videos can be got. The last is the nearest neighbor classifier is applied to classify. The proposed method is evaluated on the CASIA and USF gait databases, and the results show that our proposed algorithm can get better recognition effect than other existing algorithms.

Kernel Fisher Discriminant Analysis for Natural Gait Cycle Based Gait Recognition

Jun Huang,Xiuhui Wang,Jun Wang 한국정보처리학회 2019 Journal of information processing systems Vol.15 No.4

This paper studies a novel approach to natural gait cycles based gait recognition via kernel Fisher discriminantanalysis (KFDA), which can effectively calculate the features from gait sequences and accelerate the recognitionprocess. The proposed approach firstly extracts the gait silhouettes through moving object detection andsegmentation from each gait videos. Secondly, gait energy images (GEIs) are calculated for each gait videos, andused as gait features. Thirdly, KFDA method is used to refine the extracted gait features, and low-dimensionalfeature vectors for each gait videos can be got. The last is the nearest neighbor classifier is applied to classify. The proposed method is evaluated on the CASIA and USF gait databases, and the results show that ourproposed algorithm can get better recognition effect than other existing algorithms.

Tobacco Sales Bill Recognition Based on Multi-Branch Residual Network

Yuxiang Shan,Cheng Wang,Qin Ren,Xiuhui Wang 한국정보처리학회 2022 Journal of information processing systems Vol.18 No.3

Tobacco sales enterprises often need to summarize and verify the daily sales bills, which may consume substantialmanpower, and manual verification is prone to occasional errors. The use of artificial intelligencetechnology to realize the automatic identification and verification of such bills offers important practicalsignificance. This study presents a novel multi-branch residual network for tobacco sales bills to improve theefficiency and accuracy of tobacco sales. First, geometric correction and edge alignment were performed onthe input sales bill image. Second, the multi-branch residual network recognition model is established andtrained using the preprocessed data. The comparative experimental results demonstrated that the correctrecognition rate of the proposed method reached 98.84% on the China Tobacco Bill Image dataset, which issuperior to that of most existing recognition methods.

Research on grid‑connected harmonic current suppression of three‑phase four‑wire energy storage inverters

Hongyang Qing,Chunjiang Zhang,Xiuhui Chai,Hao He,Xiaohuan Wang 전력전자학회 2023 JOURNAL OF POWER ELECTRONICS Vol.23 No.6

When a three-phase four-wire grid-connected energy storage inverter is connected to unbalanced or single-phase loads, a large grid-connected harmonic current is generated due to the existence of a zero-sequence channel. A controller design approach for grid-connected harmonic current suppression is proposed based on proportion–integral–repetitive (PI–repetitive) control for a three-level neutral point clamped (3L-NPC) three-phase four-wire inverter. By designing the variable parameters n (gain coeffi cient of the PI controller) and Qs (gain of the repetitive controller), the eff ect of the PI–repetitive controller gain on current harmonic suppression is analyzed using a three-dimensional amplitude gain curve. A simplifi ed impedance model in the d 0-frame for a three-phase four-wire inverter is proposed. Based on the impedance model in the d0-frame, the system stability is analyzed under diff erent PI–repetitive control gains by the generalized Nyquist criterion. Finally, the optimal controller design is obtained by a gain characteristic and system stability analysis. The controller obtained by this harmonic suppression analysis method can simultaneously ensure the best grid-connected current quality of the three-phase four-wireinverter and the dynamic stability of the system. Simulation and experimental results verify the effectiveness and correctness of the proposed controller design approach for grid-connected harmonic current suppression.

Tobacco Retail License Recognition Based on Dual Attention Mechanism

Yuxiang Shan,Qin Ren,Cheng Wang,Xiuhui Wang 한국정보처리학회 2022 Journal of information processing systems Vol.18 No.4

Images of tobacco retail licenses have complex unstructured characteristics, which is an urgent technicalproblem in the robot process automation of tobacco marketing. In this paper, a novel recognition approachusing a double attention mechanism is presented to realize the automatic recognition and information extractionfrom such images. First, we utilized a DenseNet network to extract the license information from the inputtobacco retail license data. Second, bi-directional long short-term memory was used for coding and decodingusing a continuous decoder integrating dual attention to realize the recognition and information extraction oftobacco retail license images without segmentation. Finally, several performance experiments were conductedusing a largescale dataset of tobacco retail licenses. The experimental results show that the proposed approachachieves a correction accuracy of 98.36% on the ZY-LQ dataset, outperforming most existing methods.

Antibody-conjugated gold nanoparticles as nanotransducers for second near-infrared photo-stimulation of neurons in rats

Liu Jiansheng,Li Jiajia,Zhang Shu,Ding Mengbin,Yu Ningyue,Li Jingchao,Wang Xiuhui,Li Zhaohui 나노기술연구협의회 2022 Nano Convergence Vol.9 No.13

Infrared neural stimulation with the assistance of photothermal transducers holds great promise as a mini-invasive neural modulation modality. Optical nanoparticles with the absorption in the near-infrared (NIR) window have emerged as excellent photothermal transducers due to their good biocompatibility, surface modifiability, and tunable optical absorption. However, poor activation efficiency and limited stimulation depth are main predicaments encountered in the neural stimulation mediated by these nanoparticles. In this study, we prepared a targeted polydopamine (PDA)-coated gold (Au) nanoparticles with specific binding to thermo-sensitive ion channel as nanotransducers for second near-infrared (NIR-II) photo-stimulation of neurons in rats. The targeted Au nanoparticles were constructed via conjugation of anti-TRPV1 antibody with PEGylated PDA-coated Au nanoparticles and thus exhibited potent photothermal performance property in the second NIR (NIR-II) window and converted NIR-II light to heat to rapidly activate Ca 2+ influx of neurons in vitro. Furthermore, wireless photothermal stimulation of neurons in living rat successfully evoke excitation in neurons in the targeted brain region as deep as 5 mm beneath cortex. This study thus demonstrates a remote-controlled strategy for neuromodulation using photothermal nanotransducers.