Pub Date : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207468
Shuqin Wang, Hongzhi Zhao
Time shift technique and cubic spline interpolation are combined to reconstruct the irregularly sampled, aliased seismic data. The spatial aliasing is reduced by linear time shift, and the irregular sampling is handled by cubic spline interpolation. The method is applicable to both uniform sampling with missing traces and non-uniform sampling. It can handle linear, nonlinear and interfered events. The underling assumption is that the dip range of all events, within the whole data set or spatiotemporal window, is not too large. This method is feasible in practical applications since field data usually satisfy this assumption. As a one-pass and easily parallelized method, this technique has attractive computational cost and memory demand. For 3D seismic data, only 2D interpolation along spatial direction is required for each time slice. This shows great potential on huge volume data, especially for 3D marine data. Experiments on both synthetic and field data demonstrate the capability of the proposed method.
{"title":"Cubic-spline reconstruction of irregular seismic data using linear time shift","authors":"Shuqin Wang, Hongzhi Zhao","doi":"10.1109/ICWAPR.2009.5207468","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207468","url":null,"abstract":"Time shift technique and cubic spline interpolation are combined to reconstruct the irregularly sampled, aliased seismic data. The spatial aliasing is reduced by linear time shift, and the irregular sampling is handled by cubic spline interpolation. The method is applicable to both uniform sampling with missing traces and non-uniform sampling. It can handle linear, nonlinear and interfered events. The underling assumption is that the dip range of all events, within the whole data set or spatiotemporal window, is not too large. This method is feasible in practical applications since field data usually satisfy this assumption. As a one-pass and easily parallelized method, this technique has attractive computational cost and memory demand. For 3D seismic data, only 2D interpolation along spatial direction is required for each time slice. This shows great potential on huge volume data, especially for 3D marine data. Experiments on both synthetic and field data demonstrate the capability of the proposed method.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132045143","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207464
Bin Fang, Jing Chen, Yuanyan Tang
Incremental learning has attracted increasing attention in the past decade. Since many real tasks are high-dimensional problems, dimensionality reduction is the important step. Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) are two of the most widely used dimensionality reduction algorithms. However, PCA is an unsupervised algorithm. It is known that PCA is not suitable for classification tasks. Generally, LDA outperforms PCA when classification problem is involved. However, the major shortcoming of LDA is that the performance of LDA is degraded when encountering singularity problem. Recently, the modified LDA, Maximum margin criterion (MMC) was proposed to overcome the shortcomings of PCA and LDA. Nevertheless, MMC is not suitable for incremental data. The paper proposes an incremental extension version of MMC, called Incremental Maximum margin criterion (IMMC) to update projection matrix when new observation is coming, without repetitive learning. Since the approximation intermediate eigenvalue decomposition is introduced, it is low in computational complexity.
在过去的十年里，渐进式学习吸引了越来越多的关注。由于许多实际任务是高维问题，降维是重要的一步。主成分分析(PCA)和线性判别分析(LDA)是目前应用最广泛的两种降维算法。然而，PCA是一种无监督算法。众所周知，PCA不适用于分类任务。一般来说，当涉及分类问题时，LDA优于PCA。然而，LDA的主要缺点是在遇到奇异性问题时性能会下降。近年来，为了克服PCA和LDA的不足，提出了改进的LDA——最大边际准则(Maximum margin criterion, MMC)。然而，MMC不适合增量数据。本文提出了MMC的增量扩展版本，即增量最大边界准则(incremental Maximum margin criterion, IMMC)，用于在新观测值到来时更新投影矩阵，而无需重复学习。由于引入了近似中间特征值分解，计算复杂度较低。
{"title":"A linear subspace learning algorithm for incremental data","authors":"Bin Fang, Jing Chen, Yuanyan Tang","doi":"10.1109/ICWAPR.2009.5207464","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207464","url":null,"abstract":"Incremental learning has attracted increasing attention in the past decade. Since many real tasks are high-dimensional problems, dimensionality reduction is the important step. Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) are two of the most widely used dimensionality reduction algorithms. However, PCA is an unsupervised algorithm. It is known that PCA is not suitable for classification tasks. Generally, LDA outperforms PCA when classification problem is involved. However, the major shortcoming of LDA is that the performance of LDA is degraded when encountering singularity problem. Recently, the modified LDA, Maximum margin criterion (MMC) was proposed to overcome the shortcomings of PCA and LDA. Nevertheless, MMC is not suitable for incremental data. The paper proposes an incremental extension version of MMC, called Incremental Maximum margin criterion (IMMC) to update projection matrix when new observation is coming, without repetitive learning. Since the approximation intermediate eigenvalue decomposition is introduced, it is low in computational complexity.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132282759","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207489
Si Li, Hui He, Weiran Xu, Jun Guo
In recent years, sentiment analysis has become a hot topic in the study of natural language processing. Methods of machine learning are widely used to the sentiment analysis. This paper presents an approach for Chinese sentiment analysis at phrase-level. A LMR template is designed to tag word features, like position, orientation, part of speech (POS), and so on. Then, Maximum Entropy (ME) model is employed to extract sentiment words. Parts of the first Chinese Opinion Analysis Evaluation (COAE2008) corpus are used in evaluation. Experimental results show that ME model with LMR template can achieve a good performance.
{"title":"Automatic Chinese sentiment word extraction based on Aximum Entropy","authors":"Si Li, Hui He, Weiran Xu, Jun Guo","doi":"10.1109/ICWAPR.2009.5207489","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207489","url":null,"abstract":"In recent years, sentiment analysis has become a hot topic in the study of natural language processing. Methods of machine learning are widely used to the sentiment analysis. This paper presents an approach for Chinese sentiment analysis at phrase-level. A LMR template is designed to tag word features, like position, orientation, part of speech (POS), and so on. Then, Maximum Entropy (ME) model is employed to extract sentiment words. Parts of the first Chinese Opinion Analysis Evaluation (COAE2008) corpus are used in evaluation. Experimental results show that ME model with LMR template can achieve a good performance.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134478670","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207416
Lin Chen, Bin Fang, Yi Wang, Guang-Zhou Lu, Ji-Ye Qian, Chunyan Li
The particles in urinary microscopic images are hard to classify because of noisy background and strong variability of objects in shape and texture. In order to overcome these difficulties, firstly, a new method of texture feature extraction using the distance mapping based on a set of local grayvalue invariants is introduced and the feature is robust to the shift and rotation. Secondly, we reduce the high dimensional feature into a lower dimensional space using PCA. Thirdly, a multiclass SVM is applied to classify 5 categories of particles after trained them reasonably. Finally the experiment results achieve an average of accuracy of 90.02% and a F1 value of 90.44%.
{"title":"Automated classfication of particles in urinary sediment","authors":"Lin Chen, Bin Fang, Yi Wang, Guang-Zhou Lu, Ji-Ye Qian, Chunyan Li","doi":"10.1109/ICWAPR.2009.5207416","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207416","url":null,"abstract":"The particles in urinary microscopic images are hard to classify because of noisy background and strong variability of objects in shape and texture. In order to overcome these difficulties, firstly, a new method of texture feature extraction using the distance mapping based on a set of local grayvalue invariants is introduced and the feature is robust to the shift and rotation. Secondly, we reduce the high dimensional feature into a lower dimensional space using PCA. Thirdly, a multiclass SVM is applied to classify 5 categories of particles after trained them reasonably. Finally the experiment results achieve an average of accuracy of 90.02% and a F1 value of 90.44%.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122238230","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207445
Marthinus C. du Plessis, J. Olivier
This paper presents a classification method which discriminates between two radar transmitters based on the received pulses. A simple radar transmitter model is presented to which a non-stationary signal classifier is applied. The classifier is a support vector machine which is applied to the radar pulse's time-frequency representation. The time-frequency representation is refined using particle swarm optimization to increase the classification accuracy. The classification accuracy is tested in an additive white Gaussian noise channel. An acceptable classification accuracy is reported for component tolerances as small as 2% on the transmitter's modulator.
{"title":"Radar transmitter classification using a non-stationary signal classifier","authors":"Marthinus C. du Plessis, J. Olivier","doi":"10.1109/ICWAPR.2009.5207445","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207445","url":null,"abstract":"This paper presents a classification method which discriminates between two radar transmitters based on the received pulses. A simple radar transmitter model is presented to which a non-stationary signal classifier is applied. The classifier is a support vector machine which is applied to the radar pulse's time-frequency representation. The time-frequency representation is refined using particle swarm optimization to increase the classification accuracy. The classification accuracy is tested in an additive white Gaussian noise channel. An acceptable classification accuracy is reported for component tolerances as small as 2% on the transmitter's modulator.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125907032","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207476
Haina Jiang, Xiaobin Xu, Chenglin Wen
This paper provides a method based on Dezert-Smarandache Theory (DSmT) for simultaneous faults diagnosis when evidence is dependent. Firstly, according to the characteristics of simultaneous faults, a frame of discernment is given for both single fault and simultaneous faults diagnosis, the DSmT combination rule applicable to simultaneous faults diagnosis is introduced. Secondly, the dependence of original evidence is classified according to three main factors in information acquisition and extraction, a method for evidence decorrelation is provided. On the other hand, the weights for measuring evidence credibility are given to modify independent evidence based on Generalized Ambiguity Measure. Next, DSmT combination rule is used to aggregate the modified evidence. Finally, an example of rotor faults diagnosis is given to illustrate effectiveness of the proposed method.
{"title":"The combination method for dependent evidence and its application for simultaneous faults diagnosis","authors":"Haina Jiang, Xiaobin Xu, Chenglin Wen","doi":"10.1109/ICWAPR.2009.5207476","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207476","url":null,"abstract":"This paper provides a method based on Dezert-Smarandache Theory (DSmT) for simultaneous faults diagnosis when evidence is dependent. Firstly, according to the characteristics of simultaneous faults, a frame of discernment is given for both single fault and simultaneous faults diagnosis, the DSmT combination rule applicable to simultaneous faults diagnosis is introduced. Secondly, the dependence of original evidence is classified according to three main factors in information acquisition and extraction, a method for evidence decorrelation is provided. On the other hand, the weights for measuring evidence credibility are given to modify independent evidence based on Generalized Ambiguity Measure. Next, DSmT combination rule is used to aggregate the modified evidence. Finally, an example of rotor faults diagnosis is given to illustrate effectiveness of the proposed method.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127325343","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207406
Yan Liu, Shengli Zhao, Chen Yi
By analyzing the causes and influencing factors of corrosion of reinforcing steel, the RBF neural network model for predicting reinforcement corrosion is founded. And actual data is analyzed through an example and results are compared with the BP network model. The testing results show that RBF network model for predicting reinforcement corrosion can become a new effective assessment model with better prediction results and higher recognition precision.
{"title":"The forecast for corrosion of reinforcing steel based on RBF neural network","authors":"Yan Liu, Shengli Zhao, Chen Yi","doi":"10.1109/ICWAPR.2009.5207406","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207406","url":null,"abstract":"By analyzing the causes and influencing factors of corrosion of reinforcing steel, the RBF neural network model for predicting reinforcement corrosion is founded. And actual data is analyzed through an example and results are compared with the BP network model. The testing results show that RBF network model for predicting reinforcement corrosion can become a new effective assessment model with better prediction results and higher recognition precision.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117040099","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207481
Junying Gan, Sibin He
Singularity problem of LDA algorithm is overcome by Two-dimensional LDA(2DLDA), and Support Vector Machine(SVM) has the character of Structural Risk Minimization. In this paper, two methods are combined and used for face recognition. Firstly, the original images are decomposed into high-frequency and low-frequency components with the help of Wavelet Transform(WT). The high-frequency components are ignored, while the low-frequency components can be obtained. Then, the liner discriminant features are extracted by 2DLDA, and SVM is selected to perform face recognition. Experimental results based on ORL(Olivetti Research Laboratory) and Yale face database show the validity of 2DLDA+SVM for face recognition.
二维LDA(2DLDA)克服了LDA算法的奇异性问题，支持向量机(SVM)具有结构风险最小化的特点。本文将两种方法结合起来进行人脸识别。首先，利用小波变换将原始图像分解为高频和低频分量;忽略高频分量，得到低频分量。然后，利用2DLDA提取线性判别特征，选择支持向量机进行人脸识别。基于ORL(Olivetti Research Laboratory)和耶鲁大学人脸数据库的实验结果表明，2DLDA+SVM用于人脸识别是有效的。
{"title":"Face recognition based on 2DLDA and support vector machine","authors":"Junying Gan, Sibin He","doi":"10.1109/ICWAPR.2009.5207481","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207481","url":null,"abstract":"Singularity problem of LDA algorithm is overcome by Two-dimensional LDA(2DLDA), and Support Vector Machine(SVM) has the character of Structural Risk Minimization. In this paper, two methods are combined and used for face recognition. Firstly, the original images are decomposed into high-frequency and low-frequency components with the help of Wavelet Transform(WT). The high-frequency components are ignored, while the low-frequency components can be obtained. Then, the liner discriminant features are extracted by 2DLDA, and SVM is selected to perform face recognition. Experimental results based on ORL(Olivetti Research Laboratory) and Yale face database show the validity of 2DLDA+SVM for face recognition.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121527450","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207441
Liying Lang, Wei-wei Gu
Because AdaBoost Cascade face detection algorithm has a very outstanding performance, AdaBoost face detection is the mainstream algorithm currently. But it can produce misjudgment at a similar facial feature regional, particularly in the detection of more complicated image background circumstances misjudgment is even more serious. In view of reasons above, in this paper, a new algorithm was proposed and named A-SCS algorithm, which is increased skin color segmentation after detected face region use the AdaBoost algorithm. This algorithm makes full use of the image useful information, and greatly reduced the possibility of misjudgment. Compare to AdaBoost algorithm and skin color segmentation algorithm, the algorithm mentioned in this paper reduced the false detecting rate in complex background image, At the same time, it is of definite robustness. Simulated experimental results by Matlab indicate that this algorithm is faster and accuracy. Therefore it can be applied to real-time face detection system.
{"title":"A robustness and real-time face detection algorithm in complex background","authors":"Liying Lang, Wei-wei Gu","doi":"10.1109/ICWAPR.2009.5207441","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207441","url":null,"abstract":"Because AdaBoost Cascade face detection algorithm has a very outstanding performance, AdaBoost face detection is the mainstream algorithm currently. But it can produce misjudgment at a similar facial feature regional, particularly in the detection of more complicated image background circumstances misjudgment is even more serious. In view of reasons above, in this paper, a new algorithm was proposed and named A-SCS algorithm, which is increased skin color segmentation after detected face region use the AdaBoost algorithm. This algorithm makes full use of the image useful information, and greatly reduced the possibility of misjudgment. Compare to AdaBoost algorithm and skin color segmentation algorithm, the algorithm mentioned in this paper reduced the false detecting rate in complex background image, At the same time, it is of definite robustness. Simulated experimental results by Matlab indicate that this algorithm is faster and accuracy. Therefore it can be applied to real-time face detection system.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"322 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122708679","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207437
Li-wen Han, Na-Duo Yuan, Bin Xie, Ai-jun Hu
The paper considers the refinement of a family of cubic parametric splines, introduced for the shape-preserving approximation of equation of M-band (≥ 2 is an integer) with different choice of finer knot sequences, which contain the existent binary refinement. The Advantage of M-band(≥ 2) is discussed.
{"title":"Refining cubic parametric splines with tension properties","authors":"Li-wen Han, Na-Duo Yuan, Bin Xie, Ai-jun Hu","doi":"10.1109/ICWAPR.2009.5207437","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207437","url":null,"abstract":"The paper considers the refinement of a family of cubic parametric splines, introduced for the shape-preserving approximation of equation of M-band (≥ 2 is an integer) with different choice of finer knot sequences, which contain the existent binary refinement. The Advantage of M-band(≥ 2) is discussed.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124009488","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：