Small Sample Face Recognition Algorithm Based on Novel Siamese Network

Zhang, Jianming,Jin, Xiaokang,Liu, Yukai,Sangaiah, Arun Kumar,Wang, Jin Korea Information Processing Society 2018 Journal of information processing systems Vol.14 No.6

In face recognition, sometimes the number of available training samples for single category is insufficient. Therefore, the performances of models trained by convolutional neural network are not ideal. The small sample face recognition algorithm based on novel Siamese network is proposed in this paper, which doesn't need rich samples for training. The algorithm designs and realizes a new Siamese network model, SiameseFacel, which uses pairs of face images as inputs and maps them to target space so that the $L_2$ norm distance in target space can represent the semantic distance in input space. The mapping is represented by the neural network in supervised learning. Moreover, a more lightweight Siamese network model, SiameseFace2, is designed to reduce the network parameters without losing accuracy. We also present a new method to generate training data and expand the number of training samples for single category in AR and labeled faces in the wild (LFW) datasets, which improves the recognition accuracy of the models. Four loss functions are adopted to carry out experiments on AR and LFW datasets. The results show that the contrastive loss function combined with new Siamese network model in this paper can effectively improve the accuracy of face recognition.

Targeting Bcr–Abl by combining allosteric with ATP-binding-site inhibitors

Zhang, Jianming,Adriá,n, Francisco J.,Jahnke, Wolfgang,Cowan-Jacob, Sandra W.,Li, Allen G.,Iacob, Roxana E.,Sim, Taebo,Powers, John,Dierks, Christine,Sun, Fangxian,Guo, Gui-Rong,Ding, Qiang,Okra Macmillan Publishers Limited. All rights reserved 2010 Nature Vol.463 No.7280

In an effort to find new pharmacological modalities to overcome resistance to ATP-binding-site inhibitors of Bcr–Abl, we recently reported the discovery of GNF-2, a selective allosteric Bcr–Abl inhibitor. Here, using solution NMR, X-ray crystallography, mutagenesis and hydrogen exchange mass spectrometry, we show that GNF-2 binds to the myristate-binding site of Abl, leading to changes in the structural dynamics of the ATP-binding site. GNF-5, an analogue of GNF-2 with improved pharmacokinetic properties, when used in combination with the ATP-competitive inhibitors imatinib or nilotinib, suppressed the emergence of resistance mutations in vitro, displayed additive inhibitory activity in biochemical and cellular assays against T315I mutant human Bcr–Abl and displayed in vivo efficacy against this recalcitrant mutant in a murine bone-marrow transplantation model. These results show that therapeutically relevant inhibition of Bcr–Abl activity can be achieved with inhibitors that bind to the myristate-binding site and that combining allosteric and ATP-competitive inhibitors can overcome resistance to either agent alone.

Small Sample Face Recognition Algorithm based on Novel Siamese Network

( Jianming Zhang ),( Xiaokang Jin ),( Yukai Liu ),( Arun Kumar Sangaiah ),( Jin Wang ) 한국정보처리학회 2018 Journal of information processing systems Vol.14 No.6

In face recognition, sometimes the number of available training samples for single category is insufficient. Therefore, the performances of models trained by convolutional neural network are not ideal. The small sample face recognition algorithm based on novel Siamese network is proposed in this paper, which doesn’t need rich samples for training. The algorithm designs and realizes a new Siamese network model, SiameseFace1, which uses pairs of face images as inputs and maps them to target space so that the L2 norm distance in target space can represent the semantic distance in input space. The mapping is represented by the neural network in supervised learning. Moreover, a more lightweight Siamese network model, SiameseFace2, is designed to reduce the network parameters without losing accuracy. We also present a new method to generate training data and expand the number of training samples for single category in AR and labeled faces in the wild (LFW) datasets, which improves the recognition accuracy of the models. Four loss functions are adopted to carry out experiments on AR and LFW datasets. The results show that the contrastive loss function combined with new Siamese network model in this paper can effectively improve the accuracy of face recognition.

Effectiveness of Ionic Polymer Soil Stabilizers on Warm Frozen Soil

Zhilong Zhang,Hu Zhang,Jianming Zhang,Mingtang Chai 대한토목학회 2019 KSCE JOURNAL OF CIVIL ENGINEERING Vol.23 No.7

Warm frozen soil has characteristics such as low shear strength and high compressibility. To improve its mechanical strength, the ionic polymer soil stabilizers (IPSS): named XRH stabilizer and acid cation exchange resin (CER) were adopted for reinforcing the warm frozen soil from the Beiluhe region of the Qinghai–Tibetan Plateau. To evaluate the effects of IPSS on warm frozen soil and analyze the mechanisms responsible, measurements were made of the physical, chemical and mechanical properties of warm frozen soil, both untreated and following treatment with IPSS. The addition of IPSS to the soil resulted in a significant decrease in the plasticity index, freezing temperature, and cation–exchange capacity of soil. The warm frozen soil were reinforced with XRH and CER, the cohesion increased by 87.9% and 43.1%, respectively, and the compressibility decreased by 44.5% and 41.1%, respectively. The trends of mechanical strength and unfrozen water content of reinforced warm frozen soil were the same, with extreme values being observed in response to treatment with an IPSS content of about 1% (optimal proportion). In addition, the curing mechanism of IPSS on warm frozen soil was analyzed. Overall, applying IPSS to the soil was an effective and feasible measure for reinforcing a warm frozen soil foundation.

Anthocyanins in Tree Peony (Paeonia suffruticosa) and Their Relationship with Flower Color

Yanzhao Zhang,Yan Zhang,Xujia Duan,Xin Liu,Shuiqing Yuan,Jianming Han,Yanwei Cheng 한국원예학회 2020 원예과학기술지 Vol.38 No.6

Flower color is an important trait of tree peony (Paeonia suffruticosa), an economically important species. The accurate classification of flower color and understanding of the mechanism involved in creating variations in flower color provide the basis of tree peony breeding and genetic research. This study used the International Commission of Illumination Color System, known as CIELAB or CIE L<SUP>*</SUP>a<SUP>*</SUP>b<SUP>*</SUP>, to measure the flower color of 150 cultivars of the Zhongyuan cultivar group of tree peony. Hierarchical cluster analysis and visual observations were used to divide this cultivar group into five major color lines: white, pink, red, purple, and fuchsia. The brightness (L<SUP>*</SUP>) of the flower color gradually decreased with an increase in anthocyanin content. The redness (a<SUP>*</SUP>) of the flower color was significantly negatively correlated with total anthocyanin (TA) content in the fuchsia line but was positively correlated with TA in the remaining four color lines. Six types of anthocyanins, including Cy3G, Cy3G5G, Pn3G, Pn3G5G, Pg3G, and Pg3G5G, were mainly found in the Zhongyuan cultivar group by liquid chromatography except in the white line. Stepwise regression analysis showed that Pg3G5G had a significant positive correlation with redness of flower color. Cy3G5G and Pn3G5G had a significant positive correlation with redness in pink, red, and purple color lines. However, Pn3G was negatively correlated with redness in the fuchsia line. This study lays the foundation for further development of new breeds and flower colors in tree peony and for genetic research related to flower color.

On Zeros and Fixed Points of Differences of Meromorphic Functions

Zhang, Guowei,Qi, Jianming,Zheng, Yiyuan Department of Mathematics 2018 Kyungpook mathematical journal Vol.58 No.4

In this paper, we give some results on the zeros and fixed points of the difference and the divided difference of transcendental meromorphic functions. This improves on results of Langley.

Analysis of transcriptomic profiles and physiological traits of exogenous 24-epibrassinolide alleviating salt stress in Atractylodes macrocephala Koidz.

Zhang Jianyun,Fang Huiyong,Jiang Jianming,Gu Xian,Zhao Huimei,Qiu Jinmiao,Wang Qian,Zheng Yuguang,Zheng Kaiyan 한국식물생명공학회 2024 Plant biotechnology reports Vol.18 No.1

Exogenous 24-epibrassinolide (24-epiBL) application is a preferred method for improving salt tolerance in plants. However, information about the effects of exogenous 24-epiBL application related to salt response in Atractylodes macrocephala Koidz. (AMK) has rarely been reported. Here, we aimed to investigate the effects of exogenous 24-epiBL (0.01 mg L−1) on enhancing salt tolerance in AMK seedlings under salt stress (5 g L−1 NaCl) based on seed germination experiments. Our results indicated that NaCl treatment suppresses AMK seed germination, while 24-epiBL treatment under nonstress and stress conditions had positive effects on plant seedling growth. Treatment with NaCl significantly inhibited radicle elongation and root activity to endure salt stress, and 24-epiBL treatment promoted growth and alleviated the damage to osmotic stress, ion stress sub- stances, and oxidative stress by regulating free proline (Pro) and soluble protein (SP) contents and enhancing soluble sugar (SS) conversion. Moreover, the expression of normal development process genes was affected, and differentially expressed genes (DEGs) in multiple pathways associated with amino acid metabolism, signal transduction, carbohydrate metabolism, and lipid metabolism were regulated under NaCl treatment. Further analysis of the AMK transcriptome profile revealed that exogenous 24-epiBL enhanced salt tolerance through hormone regulation and carbohydrate metabolism promotion, salt- responsive gene expression inhibition, and induction of pathways related to terpenoid and polyketide metabolism as well as pentose and glucuronate interconversions for plant growth promotion. Collectively, our findings successfully demonstrate that exogenous 24-epiBL plays a positive role in regulating the AMK response to salt stress, which provides potential genetic resources to improve salt tolerance in plants.

UDE-based Nonlinear Path-following Control of Autonomous Underwater Vehicles With Multiple Uncertainties and Input Saturation

Jianming Miao,Xingyu Sun,Kankan Deng,Xi Gong,Chao Peng,Tao Liu,Haosu Zhang 제어·로봇·시스템학회 2024 International Journal of Control, Automation, and Vol.22 No.3

This paper proposes a novel UDE-based nonlinear path-following (PF) control scheme for the underactuated autonomous underwater vehicle (AUV) with multiple uncertainties and input saturation. Firstly, to overcomethe drawback of dependence on the accurate mathematical model, the unknown time-varying sideslip angular speedin the error dynamic model is treated as kinematic uncertainty, and the linear superposition of parameter uncertainties, environmental perturbations, and other unmodeled dynamics is regarded as lumped dynamic uncertainty. Bothkinematic and dynamic uncertainties are estimated and compensated by the designed uncertainty and disturbanceestimators (UDEs). Secondly, a novel augmented backstepping controller based on the estimates generated by UDEsis proposed to achieve PF control of the underactuated AUV. The nonlinear tracking differentiator (NTD) is adoptedto prevent the “explosion of complexity” inherent in the traditional back-stepping method, and the auxiliary dynamical system is utilized to solve the issue of input saturation. Lastly, the stability analysis of the entire enclosed-loopframework is presented. Simulations and comparative analyses are presented to demonstrate the effectiveness androbustness of the presented strategy

Indium doping effect on the magnetic properties of Y-type hexaferrite Ba0.5Sr1.5Zn2(Fe1-xInx)12O22

Min Zhang,Lihua Yin,Qiangchun Liu,Zhenfa Zi,Jianming Dai,Yuping Sun 한국물리학회 2018 Current Applied Physics Vol.18 No.9

The effect of indium doping on structural and magnetic properties of Y-type hexaferrite Ba0.5Sr1.5Zn2(Fe1- xInx)12O22 (x=0, 0.02, 0.04, 0.06, 0.08 and 0.1) prepared by the solid state reaction method was investigated. The Rietveld refinement method was used to analyze the X-ray diffraction patterns. The magnetic transition temperatures associated with the proper-screw spin phase to the collinear ferrimagnetic spin phase transition can be efficiently modulated by varying indium content. The magnetic transition temperature increases to a maximum with indium content x=0.04 and then decreases with x, suggesting the possibility that electrically controlled magnetization reversal can be can be effectively tailored by varying indium content. The saturation magnetization at room temperature was decreased as increasing indium content, which can be explained as the metal ions occupation. It is worthy to note that the coercivity of In-doped samples was decreased drastically compared that of undoped sample, which is probably resulted from the reduction in anisotropy field with substitution of In3+ for Fe3+. The In-doped hexaferrite Ba0.5Sr1.5Zn2(Fe1-xInx)12O22 may be potential candidates for application in magnetoelectric devices.

Li- and Mn-Rich Cathode Materials: Challenges to Commercialization

Zheng, Jianming,Myeong, Seungjun,Cho, Woongrae,Yan, Pengfei,Xiao, Jie,Wang, Chongmin,Cho, Jaephil,Zhang, Ji-Guang Wiley Blackwell (John Wiley Sons) 2017 ADVANCED ENERGY MATERIALS Vol.7 No.6