A Novel Face Recognition Algorithm based on the Deep Convolution Neural Network and Key Points Detection Jointed Local Binary Pattern Methodology

Huang, Wen-zhun,Zhang, Shan-wen The Korean Institute of Electrical Engineers 2017 Journal of Electrical Engineering & Technology Vol.12 No.1

This paper presents a novel face recognition algorithm based on the deep convolution neural network and key point detection jointed local binary pattern methodology to enhance the accuracy of face recognition. We firstly propose the modified face key feature point location detection method to enhance the traditional localization algorithm to better pre-process the original face images. We put forward the grey information and the color information with combination of a composite model of local information. Then, we optimize the multi-layer network structure deep learning algorithm using the Fisher criterion as reference to adjust the network structure more accurately. Furthermore, we modify the local binary pattern texture description operator and combine it with the neural network to overcome drawbacks that deep neural network could not learn to face image and the local characteristics. Simulation results demonstrate that the proposed algorithm obtains stronger robustness and feasibility compared with the other state-of-the-art algorithms. The proposed algorithm also provides the novel paradigm for the application of deep learning in the field of face recognition which sets the milestone for further research.

A Novel Face Recognition Algorithm based on the Deep Convolution Neural Network and Key Points Detection Jointed Local Binary Pattern Methodology

Wen-zhun Huang,Shan-wen Zhang 대한전기학회 2017 Journal of Electrical Engineering & Technology Vol.12 No.1

This paper presents a novel face recognition algorithm based on the deep convolution neural network and key point detection jointed local binary pattern methodology to enhance the accuracy of face recognition. We firstly propose the modified face key feature point location detection method to enhance the traditional localization algorithm to better pre-process the original face images. We put forward the grey information and the color information with combination of a composite model of local information. Then, we optimize the multi-layer network structure deep learning algorithm using the Fisher criterion as reference to adjust the network structure more accurately. Furthermore, we modify the local binary pattern texture description operator and combine it with the neural network to overcome drawbacks that deep neural network could not learn to face image and the local characteristics. Simulation results demonstrate that the proposed algorithm obtains stronger robustness and feasibility compared with the other state-of-the-art algorithms. The proposed algorithm also provides the novel paradigm for the application of deep learning in the field of face recognition which sets the milestone for further research.

Angelica sinensis Supercritical Fluid CO2 Extract Attenuates D-Galactose-Induced Liver and Kidney Impairment in Mice by Suppressing Oxidative Stress and Inflammation

Zhi-Zhun Mo,Zhi-Xiu Lin,ZiRen Su,Lin Zheng,Hui-Lin Li,JianHui Xie,Yan-Fang Xian,Tie-Gang Yi,Shui-Qing Huang,Jian-Ping Chen 한국식품영양과학회 2018 Journal of medicinal food Vol.21 No.9

Angelica sinensis (AS, Danggui in Chinese) is an important herbal component of various traditional formulae for the management of asthenia and its tonic effects. Although AS has been shown to ameliorate cognitive damage and nerve toxicity in D-galactose (D-gal)-elicited senescent mice brain, its effects on liver and kidney injury have not yet been explored. In this work, mice were subjected to hypodermic injection with D-gal (200 mg/kg) and orally gavaged with AS (20, 40, or 80 mg/kg) once a day for 8 successive weeks. Results revealed that AS significantly improved liver and kidney function as assessed by organ index and functional parameters. In addition, AS pretreatment effectively ameliorated the histological deterioration. AS attenuated the MDA level and markedly enhanced the activities and gene expressions of antioxidative enzymes, namely Cu, Zn-SOD, CAT, and GPx. Furthermore, AS markedly inhibited the D-gal-mediated increment of expressions of inflammatory cytokines iNOS, COX-2, IκBα, p-IκBα, and p65 and promoted the IκBα expression level in both hepatic and renal tissues. In sum, AS pretreatment could effectively guard the liver and kidney of mice from D-gal-induced injury, and the underlying mechanism was deemed to be intimately related to attenuating oxidative response and inflammatory stress.

Prognostic Value of 18F-FDG PET/CT Radiomics in Extranodal Nasal-Type NK/T Cell Lymphoma

Luo Yu,Huang Zhun,Gao Zihan,Wang Bingbing,Zhang Yanwei,Bai Yan,Wu Qingxia,Wang Meiyun 대한영상의학회 2024 Korean Journal of Radiology Vol.25 No.2

Objective: To investigate the prognostic utility of radiomics features extracted from 18F-fluorodeoxyglucose (FDG) PET/CT combined with clinical factors and metabolic parameters in predicting progression-free survival (PFS) and overall survival (OS) in individuals diagnosed with extranodal nasal-type NK/T cell lymphoma (ENKTCL). Materials and Methods: A total of 126 adults with ENKTCL who underwent 18F-FDG PET/CT examination before treatment were retrospectively included and randomly divided into training (n = 88) and validation cohorts (n = 38) at a ratio of 7:3. Least absolute shrinkage and selection operation Cox regression analysis was used to select the best radiomics features and calculate each patient’s radiomics scores (RadPFS and RadOS). Kaplan–Meier curve and Log-rank test were used to compare survival between patient groups risk-stratified by the radiomics scores. Various models to predict PFS and OS were constructed, including clinical, metabolic, clinical + metabolic, and clinical + metabolic + radiomics models. The discriminative ability of each model was evaluated using Harrell’s C index. The performance of each model in predicting PFS and OS for 1-, 3-, and 5-years was evaluated using the time-dependent receiver operating characteristic (ROC) curve. Results: Kaplan–Meier curve analysis demonstrated that the radiomics scores effectively identified high- and low-risk patients (all P < 0.05). Multivariable Cox analysis showed that the Ann Arbor stage, maximum standardized uptake value (SUVmax), and RadPFS were independent risk factors associated with PFS. Further, β2-microglobulin, Eastern Cooperative Oncology Group performance status score, SUVmax, and RadOS were independent risk factors for OS. The clinical + metabolic + radiomics model exhibited the greatest discriminative ability for both PFS (Harrell’s C-index: 0.805 in the validation cohort) and OS (Harrell’s C-index: 0.833 in the validation cohort). The time-dependent ROC analysis indicated that the clinical + metabolic + radiomics model had the best predictive performance. Conclusion: The PET/CT-based clinical + metabolic + radiomics model can enhance prognostication among patients with ENKTCL and may be a non-invasive and efficient risk stratification tool for clinical practice.

Nano/Microstructures of Some DMS Materials Fabricated by High Magnetic Field-hydrothermal Synthesis

YingLi,Mingyuan Zhu,Hongmin Jin,Zhun Wang,Yibing Li,Tao Yang,Huang Jin 한국자기학회 2008 한국자기학회 학술연구발표회 논문개요집 Vol.- No.-

Study of generalized electric spring modeling based on Hamilton’s principle and its stability

Xiaohu Wang,Chaohui Zhao,Xinyuan Chen,Zhun Huang 전력전자학회 2024 JOURNAL OF POWER ELECTRONICS Vol.24 No.5

The proposed generalized electric spring (G-ES) topology effectively reduces the redundancy caused by the parallel structure of multiple electric springs in a microgrid system. A modeling method for the G-ES is urgently needed to accurately determine the G-ES parameters and to evaluate its transient operation characteristics. The correspondence between Hamilton’s principle in mechanics and electricity is introduced, and the feasibility of Hamilton’s principle under the smart load topology is verified. The modeling method of the G-ES under Hamilton’s principle is discussed, followed by the application of repetitive control strategies on the G-ES. The filter parameters are designed using the normalization method and a simplified parameter design using repetitive controllers was applied. This article introduces the impedance analysis method into the generalized electric spring to analyze its stability under weak network conditions. This work also derives the stability judgment criteria for generalized electric springs. Finally, the feasibility of modeling the G-ES based on the Hamilton’s principle was verified using MATLAB and a DSP hardware system. In addition, the consistency between the G-ES and a single intelligent load was verified under the Hamilton’s modeling principle. The G-ES stability and dynamic performance under this modeling method meet industry requirements.

Molecular Cloning, Tissue Distribution and Segmental Ontogenetic Regulation of b^0,+ Amino Acid Transporter in Lantang Pigs

Zhi, Ai-Min,Feng, Ding-Yuan,Zhou, Xiang-Yan,Zou, Shi-Geng,Huang, Zhi-Yi,Zuo, Jian-Jun,Ye, Hui,Zhang, Chang-Ming,Dong, Ze-Min,Liu, Zhun Asian Australasian Association of Animal Productio 2008 Animal Bioscience Vol.21 No.8

Cationic amino acid transporter $b^{0,+}AT$ (HGMW-approved gene symbol SLC7A9, solute carrier family 7, member 9) plays a crucial role in amino acid nutrition. In the present study, we describe the cloning and sequencing of porcine $b^{0,+}AT$. Based on the sequence of porcine $b^{0,+}AT$ deposited in the NCBI (National Center for Biotechnological Information), we identified a putative porcine homologue. Using rapid amplification of cDNA ends (RACE), the full-length cDNA encoding porcine $b^{0,+}AT$ was isolated. The porcine $b^{0,+}AT$ cDNA was 1,680 bp long, encoding a 487 amino acid trans-membrane protein. The predicted amino acid sequence was found to have 88.9% and 87.1% identity with human and mouse $b^{0,+}AT$, respectively. Real-time RT-PCR indicated porcine $b^{0,+}AT$ transcripts expressed in heart, kidney, muscle and small intestine. The small intestine had the highest $b^{0,+}AT$ mRNA abundance while the muscle had the lowest (p<0.05). Along the longitudinal axis, the ileum had the highest $b^{0,+}AT$ mRNA abundance while the colon had the lowest (p<0.05). The $b^{0,+}AT$ mRNA level was highest on day 7 and 90 in the duodenum (p<0.05). It increased from day 1 to day 26 in the jejunum (p>0.05) and had the highest abundance on day 60 (p<0.05). There was, however, no difference between day 1, 7, 26, 30, 90 and 150 (p>0.05). The strongest $b^{0,+}AT$ expression appeared on day 7 in the ileum before weaning, and then decreased till day 30 but rose gradually again from day 60 to 150 (p<0.05).