摘要:
Hyperspectral imaging (HSI) has shown great potential in the use of paddy variety identification. However, the quality of HSI images taken by a hyperspectral camera under non-ideal illumination is vulnerable to environmental influences such as shadows and noises, leading to a degraded identification result. This problem is addressed in this study by a two-stage image processing method. First, to eliminate the influence of shadows, a grayscale image based on the reflectance slope is synthesized. The synthetic reflectance slope image (SRSI) is binarized for image segmentation and shape features extraction. Secondly, an HSI image de-noising technology based on weighted spatial filtering (WSF), which integrates both spatial and spectral information of the HSI image, is proposed to reduce the influence of noises. Finally, the extracted shape, spectral and texture features are combined and input into the support vector machine for paddy variety identification. Four varieties of paddy with different origins were tested in the experiments. The experiment results showed that compared with color images, the SRSIs could help obtain more accurate shape features. The results also showed that the WSF method can significantly reduce noises and improves the paddy variety identification accuracy.
期刊:
Mathematical Problems in Engineering,2020年2020 ISSN:1024-123X
通讯作者:
Liu, R. F.
作者机构:
[Yuan, C.; Sun, C. Q.; Liu, R. F.; Tang, X. Y.] Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan 430023, Peoples R China.
通讯机构:
[Liu, R. F.] W;Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan 430023, Peoples R China.
摘要:
The purpose of image fusion is to combine the source images of the same scene into a single composite image with more useful information and better visual effects. Fusion GAN has made a breakthrough in this field by proposing to use the generative adversarial network to fuse images. In some cases, considering retain infrared radiation information and gradient information at the same time, the existing fusion methods ignore the image contrast and other elements. To this end, we propose a new end-to-end network structure based on generative adversarial networks (GANs), termed as FLGC-Fusion GAN. In the generator, using the learnable grouping convolution can improve the efficiency of the model and save computing resources. Therefore, we can have a better trade-off between the accuracy and speed of the model. Besides, we take the residual dense block as the basic network building unit and use the perception characteristics of the inactive as content loss characteristics of input, achieving the effect of deep network supervision. Experimental results on two public datasets show that the proposed method performs well in subjective visual performance and objective criteria and has obvious advantages over other current typical methods.
期刊:
Mathematical Problems in Engineering,2020年2020 ISSN:1024-123X
通讯作者:
Yuan, C.
作者机构:
[Yuan, C.; Xu, X.; Li, Y.] Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan 430023, Peoples R China.
通讯机构:
[Yuan, C.] W;Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan 430023, Peoples R China.
摘要:
Food image detection plays an essential role in visual object detection, considering its applicability in solutions that improve people's nutritional status and thus their health-care. At present, most food detection technologies are aimed at Western food and Japanese food, but few at Chinese foods. In this work, we exert effort to establish a Chinese food image dataset called CF-108 that can be used as an essential data basis for Chinese food image detection. The CF-108 dataset contains most Chinese dishes and covers large variations in presentations of the same category. In addition, we introduce a training architecture that replaces the traditional convolution in mask region convolutional neural network (Mask R-CNN) with depthwise separable convolution, namely, Mask R-DSCNN, to reduce the expensive computation cost. Experiments demonstrate that Mask R-DSCNN can significantly reduce resource consumption and improve Chinese food images' detection efficiency without hurting too much accuracy.
作者机构:
[Fan, Lili] Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan 430023, Peoples R China.;[Hou, Meichen] Univ Chinese Acad Sci, Sch Math Sci, Beijing 100049, Peoples R China.
通讯机构:
[Fan, Lili] W;Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan 430023, Peoples R China.
关键词:
non-viscous;degenerate boundary layer;rarefaction wave;outflow problem
摘要:
This article is devoted to studying the initial-boundary value problem for an ideal polytropic model of non-viscous and compressible gas. We focus our attention on the outflow problem when the flow velocity on the boundary is negative and give a rigorous proof of the asymptotic stability of both the degenerate boundary layer and its superposition with the 3-rarefaction wave under some smallness conditions. New weighted energy estimates are introduced, and the trace of the density and velocity on the boundary are handled by some subtle analysis. The decay properties of the boundary layer and the smooth rarefaction wave also play an important role.
摘要:
In this paper, we study Rademacher series with d-dimensional vector-valued coefficients. We first employ a new combinatorial technique to present a sufficient condition for the Rademacher range of a sequence with a unique direction equal to
$${\mathbb R}^2$$
. This result also gives a positive answer to the question that whether the Rademacher range of
$$\{(n^{-1},n^{-1}\ln ^{-1}(n+1))\}$$
is
$${\mathbb R}^2$$
. Next, by constructing homogeneous Cantor sets, we prove that, for each
$$s\in [1,d]$$
, there exists a sequence with a unique direction such that its Rademacher range of Hausdorff dimension s is dense in
$${\mathbb R}^d$$
but not equal to
$${\mathbb R}^d$$
.
作者:
Zeng, Shan;Wang, Zhiyong;Huang, Rui*;Chen, Ling;Feng, David
期刊:
Neurocomputing,2019年335:59-71 ISSN:0925-2312
通讯作者:
Huang, Rui
作者机构:
[Zeng, Shan; Chen, Ling] Wuhan Polytech Univ, Coll Math & Comp Sci, Wuhan 430023, Hubei, Peoples R China.;[Feng, David; Wang, Zhiyong] Univ Sydney, Sch Informat Technol, Sydney, NSW 2006, Australia.;[Feng, David; Wang, Zhiyong] Univ Sydney, BMIT Res Grp, Sydney, NSW 2006, Australia.;[Huang, Rui] Chinese Univ Hong Kong, Sch Sci & Engn, Shenzhen 518172, Guangdong, Peoples R China.;[Feng, David] Shanghai Jiao Tong Univ, Med X Res Inst, Shanghai 200240, Peoples R China.
通讯机构:
[Huang, Rui] C;Chinese Univ Hong Kong, Sch Sci & Engn, Shenzhen 518172, Guangdong, Peoples R China.
关键词:
Fuzzy C-means clustering;Intuitionistic fuzzy sets;Multi-kernel model
摘要:
Fuzzy C-means (FCM) clustering has been widely applied in various data-driven applications. While traditional FCM clustering algorithms handle uncertainty with type-2 Fuzzy Sets (T2 FSs), the recently-proposed intuitionistic fuzzy sets (IFSs) have shown advantages for describing vague and uncertain data by taking both membership degree and non-membership degree into account. However, intuitionistic fuzzy C-means (IFCM) algorithms generally do not take the importance of individual attributes and the structure of the data into account, when multi-modal and imbalanced features are involved in an application. Therefore, in this paper, we propose to address this issue of IFCM with multi-kernel mapping. First, different types of features are grouped. Second, a composite kernel is constructed to map each attribute group into an individual kernel space and to linearly combine these kernels with optimimal weights. Comprehensive experiments have been conducted on a wide range of datasets, such as machine learning repository (UCI) dataset, fabric dataset, hyperspectral imaging classification and MRI brain images segmentation to demonstrate the superior performance of the proposed clustering algorithm in comparison with the state-of-the-art in the field. (C) 2019 Elsevier B.V. All rights reserved.
关键词:
hyperspectral unmixing;Gaussian mixture model;spatial group sparsity;superpixel segmentation;endmember variability;Bayesian framework
摘要:
In recent years, endmember variability has received much attention in the field of hyperspectral unmixing. To solve the problem caused by the inaccuracy of the endmember signature, the endmembers are usually modeled to assume followed by a statistical distribution. However, those distribution-based methods only use the spectral information alone and do not fully exploit the possible local spatial correlation. When the pixels lie on the inhomogeneous region, the abundances of the neighboring pixels will not share the same prior constraints. Thus, in this paper, to achieve better abundance estimation performance, a method based on the Gaussian mixture model (GMM) and spatial group sparsity constraint is proposed. To fully exploit the group structure, we take the superpixel segmentation (SS) as preprocessing to generate the spatial groups. Then, we use GMM to model the endmember distribution, incorporating the spatial group sparsity as a mixed-norm regularization into the objective function. Finally, under the Bayesian framework, the conditional density function leads to a standard maximum a posteriori (MAP) problem, which can be solved using generalized expectation-maximization (GEM). Experiments on simulated and real hyperspectral data demonstrate that the proposed algorithm has higher unmixing precision compared with other state-of-the-art methods.
期刊:
ANNALES DE L INSTITUT HENRI POINCARE-ANALYSE NON LINEAIRE,2019年36(1):1-25 ISSN:0294-1449
通讯作者:
Ruan, Lizhi
作者机构:
[Fan, Lili] Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan 430023, Hubei, Peoples R China.;[Ruan, Lizhi] Cent China Normal Univ, Sch Math & Stat, Hubei Key Lab Math Phys, Wuhan 430079, Hubei, Peoples R China.;[Xiang, Wei] City Univ Hong Kong, Dept Math, Kowloon, Tat Chee Ave, Hong Kong, Peoples R China.
通讯机构:
[Ruan, Lizhi] C;Cent China Normal Univ, Sch Math & Stat, Hubei Key Lab Math Phys, Wuhan 430079, Hubei, Peoples R China.
期刊:
International Journal of Control, Automation and Systems,2019年17(5):1141-1148 ISSN:1598-6446
通讯作者:
Zhao, Jiemei
作者机构:
[Zhao, Jiemei] Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan 430023, Hubei, Peoples R China.;[Hu, Zhonghui] China Shipbldg Ind Corp, Res Inst 709, Wuhan 430074, Hubei, Peoples R China.
通讯机构:
[Zhao, Jiemei] W;Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan 430023, Hubei, Peoples R China.
关键词:
Discrete and distributed delays;generalized reciprocally convex combination;optimization method;reachable set estimation
摘要:
In this study, we investigate the reachable set estimation for linear systems with discrete delay and distributed delay as well as disturbances. Based on the reciprocally convex combination lemma, free-weighting matrix approach and convex analysis technique, improved delay-dependent linear matrix inequalities (LMIs) criteria are derived for finding an ellipsoid to bound the reachable sets of such systems. Moreover, this result is extended to the one for linear systems with discrete delay, distributed delay and disturbances as well as polytopic-type uncertainties. In addition, the proposed results include some existing ones as special cases since the initial conditions of the systems are not required to be zero. Finally, two numerical examples are provided to show that our results have less conservative.
摘要:
One of the main concerns in rank aggregation tasks for metasearch service is how to retrieve and aggregate the large-scale candidate search results efficiently. Much work has been done to implement metasearch service engines with different rank aggregation algorithms. However, the performance of these metasearch engines can hardly be improved. In this paper, we transform the top-k ranking task into a multi-objective programming problem when user preferences are considered along with user queries. We build an improved discrete multi-objective programming model to make the aggregate rankings satisfy user queries and user preferences both, and then propose a user preferences-based rank aggregation algorithm accordingly. Based on discrete particle swarm optimization algorithm, we improve the encoding scheme, the initialization methods, the position and velocity definition, the integrating updating process, the turbulence operator, and the external archive updating and leader selection strategy to make sure the candidate results that fit the user’s preferences can be located quickly and accurately in a large-scale discrete solution space. We have our proposed algorithm tested on three different benchmark datasets: a public dataset, the real-world datasets and the synthetic simulation datasets. The experimental results demonstrate the efficacy and convergence efficiency of the proposed algorithm over the baseline rank aggregation methods especially when dealing with large amount of candidate results. And when the set of candidate results is of normal size, the proposed algorithm is proved to perform not worse than the baseline methods.
摘要:
In this paper we study a multi-layer tumor model which is expressed as a free boundary problem of a system of partial differential equations. The problem consists of two elliptic equations describing the distribution of nutrient concentration and the pressure between tumor cells, respectively, in an unbounded strip-like region in which the tumor occupies. This region has two disjoint boundaries: While the lower part is fixed, the upper part, which stands for the tumor surface, can move as the tumor grows. Under certain conditions, the problem can be proved to admit a unique equilibrium which corresponds to the flat upper boundary. We first convert the model into a parabolic differential equation in certain function space. Next we compute the spectrum of the linearized problem at the equilibrium. By applying the geometric theory of parabolic differential equations in Banach spaces, we prove that if the cell-to-cell adhesiveness coefficient gamma is larger than a threshold value gamma*, then the unique flat equilibrium is asymptotically stable, whereas in the case 0 < gamma < gamma* the flat equilibrium is unstable. (C) 2018 Elsevier Ltd. All rights reserved.
期刊:
Human-centric Computing and Information Sciences,2019年9(1):1-16 ISSN:2192-1962
通讯作者:
Zhang, Cong
作者机构:
[Chao, Han-Chieh; Liu, Yu; Zhang, Cong; Wang, Song] Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan, Hubei, Peoples R China.;[Hang, Bo] Hubei Univ Arts & Sci, Sch Comp Engn, Xiangyang, Peoples R China.
通讯机构:
[Zhang, Cong] W;Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan, Hubei, Peoples R China.
关键词:
Audio;Attention computational model;Information entropy;Exponential moving average
摘要:
The main down-top attention model usually uses some characteristics of audio signal to extract the auditory saliency map at present. But existing audio attention computational model based image saliency mostly doesn't consider the continuity and attenuation mechanism of the human brain on paying attention to some occurred events in our real environment. To address these issues, we propose a model based on information entropy of two channels and exponential moving average (EMA) to simulate the processing of sound signals from the basilar membrane to the cochlear nucleus, and get the local information entropy of image and audio channel, finally we use the relevant EMA methods to calculate the auditory attention map. Some experimental results on artificial audio signal and real-world audio signal belonging to a public corpus show that the proposed model not only can detect the attention event, but also reflect the mechanism of human brain, and the experimental results of talk show in real environment are better. The main down-top attention model usually uses some characteristics of audio signal to extract the auditory saliency map at present. But existing audio attention computational model based image saliency mostly doesn't consider the continuity and attenuation mechanism of the human brain on paying attention to some occurred events in our real environment. To address these issues, we propose a model based on information entropy of two channels and exponential moving average (EMA) to simulate the processing of sound signals from the basilar membrane to the cochlear nucleus, and get the local information entropy of image and audio channel, finally we use the relevant EMA methods to calculate the auditory attention map. Some experimental results on artificial audio signal and real-world audio signal belonging to a public corpus show that the proposed model not only can detect the attention event, but also reflect the mechanism of human brain, and the experimental results of talk show in real environment are better.
关键词:
Near-infrared spectroscopy;Sesame oil;Lignan;Linear discriminant analysis;Partial least squares
摘要:
Near-infrared spectroscopy (NIRS) combined with chemometric methods were used to discriminate the sesame oils from different Chinese provinces and determinate the lignans (sesamin and sesamolin) in these sesame oils. The geographic discriminant model was constructed by principal component analysis (PCA) combining with linear discriminant analysis (LDA), and the quantitative analysis models of lignans were built using partial least squares (PLS) regression. Multiplicative scatter correction (MSC) and competitive adaptive reweighted sampling (CARS) were adopted to optimize the regression models. It was found that the discriminant model could recognize the sesame oils from different Chinese provinces correctly, and the contents of sesamin and sesamolin calculated from the optimized models and high-performance liquid chromatography (HPLC) analysis are rather close. Reasonable results proved the feasibility of near-infrared spectroscopy (NIRS) combined with chemometric methods for geographic origin of sesame oils and quantitate analysis of lignans in sesame oils.
期刊:
TRANSACTIONS OF THE AMERICAN MATHEMATICAL SOCIETY,2019年371(11):7559-7583 ISSN:0002-9947
通讯作者:
Wang, Zhi-Qiang
作者机构:
[Wang, Qingfang; Peng, Shuangjie] Cent China Normal Univ, Sch Math & Stat, Wuhan 430079, Hubei, Peoples R China.;[Wang, Qingfang] Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan 430023, Hubei, Peoples R China.;[Wang, Zhi-Qiang] Tianjin Univ, Ctr Appl Math, Tianjin 300072, Peoples R China.;[Wang, Zhi-Qiang] Utah State Univ, Dept Math & Stat, Logan, UT 84322 USA.
通讯机构:
[Wang, Zhi-Qiang] T;[Wang, Zhi-Qiang] U;Tianjin Univ, Ctr Appl Math, Tianjin 300072, Peoples R China.;Utah State Univ, Dept Math & Stat, Logan, UT 84322 USA.
摘要:
We consider the following nonlinear Schrödinger system with mixed couplings in \mathbb {R}^3: \begin{equation*} -\Delta u_i + \lambda _i u_i=\mu _i u_i^3+\sum \limits _{j=1,j\neq i}^N\beta _{ij}u_j^2u_i,\ \ \ i=1,\cdots ,N, \end{equation*} where \lambda _i, \mu _i>0, \beta _{ij}=\beta _{ji} (i,j=1,\cdots ,N, i\neq j). The system appears in modeling of Bose-Einstein condensates theory. While most existing works in the literature are concerned with purely attractive or purely repulsive couplings (i.e., all \beta _{ij} have the same signs), we examine the effect of mixed nonlinear couplings on the solution structure and obtain vector solutions with some of the components synchronized between them while being segregated with the rest of the components simultaneously.
期刊:
JOURNAL D ANALYSE MATHEMATIQUE,2019年137(1):231-249 ISSN:0021-7670
通讯作者:
Peng, Shuangjie
作者机构:
[Peng, Shuangjie] Cent China Normal Univ, Sch Math & Stat, Wuhan 430079, Hubei, Peoples R China.;[Peng, Yanfang] Giuzhou Normal Univ, Sch Math Sci, Guiyang 550001, Guizhou, Peoples R China.;[Wang, Qingfang] Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan 430079, Hubei, Peoples R China.
通讯机构:
[Peng, Shuangjie] C;Cent China Normal Univ, Sch Math & Stat, Wuhan 430079, Hubei, Peoples R China.
摘要:
We study the following elliptic system with critical exponent:
$$\left\{ {\begin{array}{*{20}{c}}
{ - \Delta u = {\lambda _1}u + {u_1}|u{|^{2*-2}}u + \beta |u{{|^{\frac{{2*}}{2} - 2}{{u|v|}}^{\frac{{2*}}{2}}}},\;\;x \in \Omega } \\
{ - \Delta v = {\lambda _2}v + {u_2}|v{|^{2*-2}}v + \beta |v{{|^{\frac{{2*}}{2} - 2}{{v|u|}}^{\frac{{2*}}{2}}}},\;\;x \in \Omega } \\ \;\;\;\;\;\;\;
{u = v = 0,\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;x \in \partial \Omega ,}
\end{array}} \right.\;\;$$
where Ω is a smooth bounded domain in
$$\mathbb{R}^N,\;N=5,2*:=\frac{2N}{N-2}$$
is the critical Sobolev exponent, μ1,μ2 > 0,
$$\beta \in ( - \sqrt {{\mu _1},{\mu _2}} ,0)$$
, 0 < λ1, λ2 < λ1(Ω), λ1(Ω) is the first eigenvalue of —Δ in
$$H^1_0(\Omega)$$
. In [10], Chen, Lin and Zou established a sign-changing solution of the above system in the case N ≥ 6 for β < 0 and λ1, λ2 ∈ (0, λ1(Ω)). We show that in dimension N = 5, for λ1 and λ2 slightly smaller than λ1(Ω), the above system has a sign-changing solution in the following sense: one component changes sign and has exactly two nodal domains, while the other one is positive.