Pub Date : 2015-03-03DOI: 10.1109/AISP.2015.7123531
Taha Hamedani, Ramin Zarei, A. Harati
In this work we proposed a novel super pixel based segmentation approach to solve energy minimization problem which can be used to deal with indoor scene labeling problem. We used Range data beside color image captured from Kinect sensor. This sensor enables us to use 3D features of structure like normal vector and 2D color features. We extracted the region of scene as super pixel based on the both color and direction change; and, consequently, we constructed our graphical model on these regions and apply Markov random field inference to assign efficient labels to them. Our evaluation on 30 scenes of challenging NYU v1 dataset shows that our proposed method reached higher values of “Correct Detection” and lower rate of “Missed instances” and “Noise instances” criteria according to Hoover evaluation method.
{"title":"Superpixel based RGB-D image segmentation using Markov random field","authors":"Taha Hamedani, Ramin Zarei, A. Harati","doi":"10.1109/AISP.2015.7123531","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123531","url":null,"abstract":"In this work we proposed a novel super pixel based segmentation approach to solve energy minimization problem which can be used to deal with indoor scene labeling problem. We used Range data beside color image captured from Kinect sensor. This sensor enables us to use 3D features of structure like normal vector and 2D color features. We extracted the region of scene as super pixel based on the both color and direction change; and, consequently, we constructed our graphical model on these regions and apply Markov random field inference to assign efficient labels to them. Our evaluation on 30 scenes of challenging NYU v1 dataset shows that our proposed method reached higher values of “Correct Detection” and lower rate of “Missed instances” and “Noise instances” criteria according to Hoover evaluation method.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134307138","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123527
Farzad Peyravi, V. Derhami, A. Latif
Social network analysis has an increasing growth as an academic field which overlaps with popular interest in social networks. Search for an expert is one of the most important issues of mining of social networks which is finding the right person with the suitable skills and knowledge. The RLS algorithm exploited Q-Learning and referrals to find experts in social network to search expert in social network. Comparison of RLS with Simple Search Algorithm, Referral Algorithm and SNPageRank shows increase in both precision and recall. RLS learns to find new experts as old experts substitute their role with new ones due to changes in social network environment.
{"title":"Reinforcement learning based search (RLS) algorithm in social networks","authors":"Farzad Peyravi, V. Derhami, A. Latif","doi":"10.1109/AISP.2015.7123527","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123527","url":null,"abstract":"Social network analysis has an increasing growth as an academic field which overlaps with popular interest in social networks. Search for an expert is one of the most important issues of mining of social networks which is finding the right person with the suitable skills and knowledge. The RLS algorithm exploited Q-Learning and referrals to find experts in social network to search expert in social network. Comparison of RLS with Simple Search Algorithm, Referral Algorithm and SNPageRank shows increase in both precision and recall. RLS learns to find new experts as old experts substitute their role with new ones due to changes in social network environment.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122117451","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123480
Seyed Mehdi Hosseini Andargoli, Javad Malekzadeh
In this paper, power allocation and target assignment is considered as a promising way to obtain low probability of interception (LPI) in the radar network. Spatial diversity in the netted radars gives us a flexibility to control power intelligently and radar assignment dynamically in such a way that not only detection performances satisfied but LPI characteristics of the network are optimized. We formulate this problem as a non-convex and nonlinear optimization problem associated with detection performance constraints. The optimum solution of general problem is complicated and cannot be solved mathematically. We relaxed problem to more tractable form for the networks with low complexity in which combination of radar's information cannot be handled. In the considered scenario, for each target only one radar is assigned and each assigned radar can only transmit one target's information. We propose a simple framework to obtain optimum power allocation and radar assignment strategy due to spatial diversity of netted radars. The framework has lower complexity compared with optimum exhaustive search algorithm and simulation results show effectiveness of proposed algorithm in satisfaction of detection performances and improvement of LPI specification of radar network.
{"title":"Target assignment and power allocation for LPI radar networks","authors":"Seyed Mehdi Hosseini Andargoli, Javad Malekzadeh","doi":"10.1109/AISP.2015.7123480","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123480","url":null,"abstract":"In this paper, power allocation and target assignment is considered as a promising way to obtain low probability of interception (LPI) in the radar network. Spatial diversity in the netted radars gives us a flexibility to control power intelligently and radar assignment dynamically in such a way that not only detection performances satisfied but LPI characteristics of the network are optimized. We formulate this problem as a non-convex and nonlinear optimization problem associated with detection performance constraints. The optimum solution of general problem is complicated and cannot be solved mathematically. We relaxed problem to more tractable form for the networks with low complexity in which combination of radar's information cannot be handled. In the considered scenario, for each target only one radar is assigned and each assigned radar can only transmit one target's information. We propose a simple framework to obtain optimum power allocation and radar assignment strategy due to spatial diversity of netted radars. The framework has lower complexity compared with optimum exhaustive search algorithm and simulation results show effectiveness of proposed algorithm in satisfaction of detection performances and improvement of LPI specification of radar network.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122531982","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123521
H. Hadizadeh
This paper presents a rotation-invariant texture descriptor, which is robust to noise. In the proposed method, a given gray-scale texture image is first filtered by a set of Gabor wavelets filters. The filters are designed such that their half-peak magnitude support in the frequency spectrum touch each other with no overlap to reduce redundant information. After that a number of local binary patterns called “Local Gabor Wavelets Binary Patterns” (LGWBPs) are computed based on the obtained Gabor wavelets filters responses via global measures. The histogram of the computed LGWBPs is then used as a texture feature vector. Extensive experiments were conducted on the well-known Outex, and CUReT databases in the presence of different levels of Gaussion noise. Experimental results indicate that the proposed method can be utilized as a suitable noise-robust and rotation-invariant texture descriptor for texture classification.
{"title":"Noise-resistant and rotation-invariant texture description and representation using local Gabor wavelets binary patterns","authors":"H. Hadizadeh","doi":"10.1109/AISP.2015.7123521","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123521","url":null,"abstract":"This paper presents a rotation-invariant texture descriptor, which is robust to noise. In the proposed method, a given gray-scale texture image is first filtered by a set of Gabor wavelets filters. The filters are designed such that their half-peak magnitude support in the frequency spectrum touch each other with no overlap to reduce redundant information. After that a number of local binary patterns called “Local Gabor Wavelets Binary Patterns” (LGWBPs) are computed based on the obtained Gabor wavelets filters responses via global measures. The histogram of the computed LGWBPs is then used as a texture feature vector. Extensive experiments were conducted on the well-known Outex, and CUReT databases in the presence of different levels of Gaussion noise. Experimental results indicate that the proposed method can be utilized as a suitable noise-robust and rotation-invariant texture descriptor for texture classification.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116441499","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123497
Roya Aliabkabri Sani, A. Ghasemi
In the recent decades, many intrusion detection systems (IDSs) have been proposed to enhance the security of networks. A class of IDSs is based on clustering of network traffic into normal and abnormal according to some features of the connections. The selected distance function to measure the similarity and dissimilarity of sessions' features affect the performance of clustering based IDSs. The most popular distance metric, which is used in designing these IDSs is the Euclidean distance function. In this paper, we argue that more appropriate distance functions can be deployed for IDSs. We propose a method of learning an appropriate distance function according to a set of supervision information. This metric is derived by solving a semi-definite optimization problem, which attempts to decrease the distance between the similar, and increases the distances between the dissimilar feature vectors. The evaluation of this scheme over Kyoto2006+ dataset shows that the new distance metric, can improve the performance of a support vector machine (SVM) clustering based IDS in terms of normal detection and false positive rates.
{"title":"Learning a new distance metric to improve an SVM-clustering based intrusion detection system","authors":"Roya Aliabkabri Sani, A. Ghasemi","doi":"10.1109/AISP.2015.7123497","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123497","url":null,"abstract":"In the recent decades, many intrusion detection systems (IDSs) have been proposed to enhance the security of networks. A class of IDSs is based on clustering of network traffic into normal and abnormal according to some features of the connections. The selected distance function to measure the similarity and dissimilarity of sessions' features affect the performance of clustering based IDSs. The most popular distance metric, which is used in designing these IDSs is the Euclidean distance function. In this paper, we argue that more appropriate distance functions can be deployed for IDSs. We propose a method of learning an appropriate distance function according to a set of supervision information. This metric is derived by solving a semi-definite optimization problem, which attempts to decrease the distance between the similar, and increases the distances between the dissimilar feature vectors. The evaluation of this scheme over Kyoto2006+ dataset shows that the new distance metric, can improve the performance of a support vector machine (SVM) clustering based IDS in terms of normal detection and false positive rates.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114569791","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123530
H. Mahmoudi, M. Homayounpour
This paper represents a statistically framework for a Persian spoken dialogue system. The framework is based on the Partially Observable Markov Decision Process (POMDP). A Bayesian network is used to represent the states of the POMDP model. It is shown that Bayesian approaches can improve the spoken dialogue system performance by handling uncertainties. Also Natural Actor Critic (NAC) algorithm is used for learning in spoken dialogue system and finally a framework for collecting training data is proposed. We compare the system with a handcrafted spoken dialogue system to show the efficiency of the proposed framework.
{"title":"A Persian spoken dialogue system using POMDPs","authors":"H. Mahmoudi, M. Homayounpour","doi":"10.1109/AISP.2015.7123530","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123530","url":null,"abstract":"This paper represents a statistically framework for a Persian spoken dialogue system. The framework is based on the Partially Observable Markov Decision Process (POMDP). A Bayesian network is used to represent the states of the POMDP model. It is shown that Bayesian approaches can improve the spoken dialogue system performance by handling uncertainties. Also Natural Actor Critic (NAC) algorithm is used for learning in spoken dialogue system and finally a framework for collecting training data is proposed. We compare the system with a handcrafted spoken dialogue system to show the efficiency of the proposed framework.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125340585","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123486
Ali Akbar Nasiri, M. Fathy
It is very important to protect fingerprint templates in the fingerprint recognition systems. Fuzzy vault is a promising and applicable scheme for this purpose. It can protect biometric templates. Also it can perform secure key management. Alignment of the query fingerprint sample in the encrypted domain and the template fingerprint sample is a challenging task. In this paper, we propose an alignment-free fingerprint cryptosystem based on multiple fuzzy vaults. In the proposed method, in registration phase, multiple vaults are constructed for one fingerprint and in verification phase, if at least two of the vaults are decoded successfully by the query fingerprint, the secret will be recovered. The Experiments of the proposed fingerprint cryptosystem are conducted on FVC2002-DBla and FVC2002-DB2a data sets to evaluate the performance of the proposed fingerprint cryptosystem.
{"title":"Alignment-free fingerprint cryptosystem based on multiple fuzzy vaults","authors":"Ali Akbar Nasiri, M. Fathy","doi":"10.1109/AISP.2015.7123486","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123486","url":null,"abstract":"It is very important to protect fingerprint templates in the fingerprint recognition systems. Fuzzy vault is a promising and applicable scheme for this purpose. It can protect biometric templates. Also it can perform secure key management. Alignment of the query fingerprint sample in the encrypted domain and the template fingerprint sample is a challenging task. In this paper, we propose an alignment-free fingerprint cryptosystem based on multiple fuzzy vaults. In the proposed method, in registration phase, multiple vaults are constructed for one fingerprint and in verification phase, if at least two of the vaults are decoded successfully by the query fingerprint, the secret will be recovered. The Experiments of the proposed fingerprint cryptosystem are conducted on FVC2002-DBla and FVC2002-DB2a data sets to evaluate the performance of the proposed fingerprint cryptosystem.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127522603","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123500
M. J. Moghaddam, Matin Hosseini, R. Safabakhsh
Traffic is an issue that many big cities are confronted with because of ever-increasing population growth. In this paper we propose a two phase traffic light control system based on fuzzy Q-learning for an isolated 4-way intersection. The states and actions of the Q-learning variables is set by a fuzzy algorithm which can be learned through environmental interactions and taking advantage of fuzzy logic. The proposed algorithm was simulated for a period of one hour for each of 14 different traffic conditions. Comparison with other methods was carried out on the 14 traffic conditions. The results showed that the proposed algorithms decrease the total waiting time and the mean of queue length.
{"title":"Traffic light control based on fuzzy Q-leaming","authors":"M. J. Moghaddam, Matin Hosseini, R. Safabakhsh","doi":"10.1109/AISP.2015.7123500","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123500","url":null,"abstract":"Traffic is an issue that many big cities are confronted with because of ever-increasing population growth. In this paper we propose a two phase traffic light control system based on fuzzy Q-learning for an isolated 4-way intersection. The states and actions of the Q-learning variables is set by a fuzzy algorithm which can be learned through environmental interactions and taking advantage of fuzzy logic. The proposed algorithm was simulated for a period of one hour for each of 14 different traffic conditions. Comparison with other methods was carried out on the 14 traffic conditions. The results showed that the proposed algorithms decrease the total waiting time and the mean of queue length.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131488310","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123499
Hadi Shahraki, S. Zahiri
In this paper a particle swarm classifier is proposed to classify fuzzy data sets. This classifier is able to find the decision hyperplanes between different classes with fuzzy samples. The performance of the proposed classifier has been tested on various fuzzy data sets. The experimental results show that our proposed classifier is able to classify fuzzy data sets as other common uncertain data classifiers. Also the results obtained from classifying some crisp data sets show the powerfulness of our proposed classifier for crisp data sets is same as the traditional particle swarm classifier.
{"title":"Particle swarm classifier for fuzzy data sets","authors":"Hadi Shahraki, S. Zahiri","doi":"10.1109/AISP.2015.7123499","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123499","url":null,"abstract":"In this paper a particle swarm classifier is proposed to classify fuzzy data sets. This classifier is able to find the decision hyperplanes between different classes with fuzzy samples. The performance of the proposed classifier has been tested on various fuzzy data sets. The experimental results show that our proposed classifier is able to classify fuzzy data sets as other common uncertain data classifiers. Also the results obtained from classifying some crisp data sets show the powerfulness of our proposed classifier for crisp data sets is same as the traditional particle swarm classifier.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131889807","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123483
Seyedeh Hamideh Shalforoushan, Mehrdad Jalali
Link prediction is as an effective technique in social network analysis to find out the relations between users and has received great concentration by many researchers in recent studies. In this paper a method is proposed for friend recommendation in social networks using Bayesian networks. The Bayesian network is a reliable model to understand the relations between variables and has been used in many areas for prediction. This method with considering effective features on creating friendships, suggests friends to users accurately. First, the goal is to find attributes and similarities that have the most effect on creating a friendship. After that friends with most common similarities will be suggested to each other. The results of the proposed method are compared with those obtained from different algorithms like Friend Of Friend and it is found that the method used in this paper significantly improves the accuracy of friend suggestion due to inclusion of several features.
链接预测作为社交网络分析中发现用户之间关系的一种有效技术，近年来受到了许多研究者的关注。本文提出了一种基于贝叶斯网络的社交网络好友推荐方法。贝叶斯网络是一个可靠的模型来理解变量之间的关系，并已用于许多领域的预测。这种方法考虑了建立友谊的有效功能，准确地向用户推荐朋友。首先，我们的目标是找到对建立友谊最有影响的特质和相似之处。之后，朋友最常见的相似之处将被推荐给对方。将本文方法的结果与Friend of Friend等不同算法的结果进行比较，发现本文方法由于包含了多个特征，显著提高了朋友推荐的准确率。
{"title":"Link prediction in social networks using Bayesian networks","authors":"Seyedeh Hamideh Shalforoushan, Mehrdad Jalali","doi":"10.1109/AISP.2015.7123483","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123483","url":null,"abstract":"Link prediction is as an effective technique in social network analysis to find out the relations between users and has received great concentration by many researchers in recent studies. In this paper a method is proposed for friend recommendation in social networks using Bayesian networks. The Bayesian network is a reliable model to understand the relations between variables and has been used in many areas for prediction. This method with considering effective features on creating friendships, suggests friends to users accurately. First, the goal is to find attributes and similarities that have the most effect on creating a friendship. After that friends with most common similarities will be suggested to each other. The results of the proposed method are compared with those obtained from different algorithms like Friend Of Friend and it is found that the method used in this paper significantly improves the accuracy of friend suggestion due to inclusion of several features.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116623067","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：