2017 Artificial Intelligence and Signal Processing Conference (AISP)最新文献

{"title":"Graph theoretical metrics and machine learning for diagnosis of Parkinson's disease using rs-fMRI","authors":"Amirali Kazeminejad, Soroosh Golbabaei, H. Soltanian-Zadeh","doi":"10.1109/AISP.2017.8324124","DOIUrl":"https://doi.org/10.1109/AISP.2017.8324124","url":null,"abstract":"In this study, we investigated the suitability of graph theoretical analysis for automatic diagnosis of Parkinson's disease. Resting state fMRI data from 18 healthy controls and 19 patients were used in the study. After data preprocessing and identifying 90 regions of interest using the AAL atlas, average time series of each region was obtained. Next, a brain network graph was constructed using the regions as nodes and the Pearson correlation between their average time series as edge weights. A percentage of edges with the highest magnitude were kept and the rest were omitted from the graph using a thresholding method ranging from 10% to 30% with 2% increments. Global graph theoretical metrics for integration (Characteristic path length and Efficiency), segregation (Clustering Coefficient and Transitivity) and small-worldness were extracted for each subject and their between group differences were subjected to statistical analysis. Local metrics, including integration, segregation, centrality (betweenness, z-score, and participation coefficient) and nodal degree, were also extracted for each subject and used as features to train a support vector machine classifier. We have shown a statistically significant increase in characteristic path length as well as a decrease in segregation metrics and efficiency in Parkinson's patients. A floating forward automatic feature selection method was used to select the 5 best features from all extracted metrics to classify patients. Our classifier was able to achieve a diagnosis accuracy of ∼95% when subjected to a leave-one-out cross-validation test. These features belonged to cuneus (right hemisphere), precuneus (left), superior (right) and middle (both) frontal gyri which were all previously reported to undergo alterations in Parkinson's disease. This investigation confirmed that global brain network alterations are associated with Parkinson's patients' symptoms and showed the potency of using graph theoretical metrics and machine learning for diagnosing the disease.","PeriodicalId":386952,"journal":{"name":"2017 Artificial Intelligence and Signal Processing Conference (AISP)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114553079","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}

{"title":"Two-stage spectrum sensing in passive radar","authors":"A. Soltani, M. R. Taban, Hamid Saeedi‐Sourck","doi":"10.1109/AISP.2017.8324093","DOIUrl":"https://doi.org/10.1109/AISP.2017.8324093","url":null,"abstract":"In this paper, we address the problem of spectrum sensing of received signal in the passive surveillant systems. The received signal is assumed to be completely unknown. We use the two-stage spectrum sensing method to facilitate the detection operation in the passive radars. Moreover, the proposed spectrum analyzer in the passive radar is compared with the one-stage spectrum sensing. Analytical and numerical results show that this method reduces the computational complexity and estimates the frequency characteristics accurately.","PeriodicalId":386952,"journal":{"name":"2017 Artificial Intelligence and Signal Processing Conference (AISP)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114707592","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}

{"title":"Spam filtering in SMS using recurrent neural networks","authors":"R. Taheri, R. Javidan","doi":"10.1109/AISP.2017.8515158","DOIUrl":"https://doi.org/10.1109/AISP.2017.8515158","url":null,"abstract":"Short Message Service (SMS) is one of the mobile communication services that allows easy and inexpensive communication. Producing unwanted messages with the aim of advertising or harassment and sending these messages on SMS have become the biggest challenge in this service. Various methods have been presented to detect unsolicited short messages; many of which are based on machine learning. Neural Networks have been applied to separate the unwanted text messages (known as spam) from normal short messages (known as ham) in SMS. To the best of our knowledge, Recurrent Neural Network (RNN) has not been used in this issue yet. In this paper, we proposed a new method which utilizes RNN to separate the ham and spam with variable length sequences; even though we used a fixed sequence length. The proposed method achieved an accuracy of 98.11, indicates a considerable improvement compared to Support Vector Machine (SVM), token-based SVM and Bayesian algorithms with accuracies of 97.81, 97.64, and 80.54, respectively.","PeriodicalId":386952,"journal":{"name":"2017 Artificial Intelligence and Signal Processing Conference (AISP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121647649","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}

{"title":"Employing deep learning and sparse representation for data classification","authors":"Seyed Mehdi Hazrati Fard, S. Hashemi","doi":"10.1109/AISP.2017.8324099","DOIUrl":"https://doi.org/10.1109/AISP.2017.8324099","url":null,"abstract":"Selecting a proper set of features with the best discrimination is always a challenge in classification. In this paper we propose a method, named GLLC (General Locally Linear Combination), to extract features using a deep autoencoder and reconstruct a sample based on other samples in a low dimensional space, then the class with minimum reconstruction error is selected as the winner. Extracting features along with the discrimination characteristic of the sparse model can create a robust classifier that shows simultaneous reduction of samples and features. Although the main application of GLLC is in the visual classification and face recognition, it can be used in other applications. We conduct extensive experiments to demonstrate that the proposed algorithm gain high accuracy on various datasets and outperforms the state-of-the-art methods.","PeriodicalId":386952,"journal":{"name":"2017 Artificial Intelligence and Signal Processing Conference (AISP)","volume":"224 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127192908","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}

{"title":"Visual attention-driven wireless multicasting of images using adaptive compressed sensing","authors":"Ahmad Shoja Yami, H. Hadizadeh","doi":"10.1109/AISP.2017.8324103","DOIUrl":"https://doi.org/10.1109/AISP.2017.8324103","url":null,"abstract":"Wireless multicasting of image/video signals has recently become a popular application, and various schemes have been proposed for this purpose, among them a recently-proposed scheme called SoftCast has gained a lot of attention. In SoftCast, a block-based discrete cosine transform (DCT) is applied on a given image, and the resultant coefficients are then scaled based on their expected energy within a power-distortion optimization (PDO) process. The scaled coefficients are then whitened, packetized, and transmitted over OFDM channels in an analoglike manner. Due to the linear operations used in SoftCast, each receiver is able to reconstruct the transmitted image in a graceful manner according to its channel characteristics. However, SoftCast requires a large bandwidth, and it does not consider the perceptual importance of various regions in the image. In this paper, we present a novel framework for wireless multicasting of static images. In the proposed framework, a block-wise compressed sensing (BCS) is applied on a given image to obtain measurement data. Given that due the visual attention mechanism of the brain, some parts of an image are more visually important (salient) than others, the sampling rate of various blocks is then estimated by their complexity and their visual saliency to consume the available bandwidth efficiently. The obtained data are then packetized and transmitted over OFDM channels. At the decoder side, users with different channel characteristics receive a certain number of packets, and reconstruct the transmitted image based on the available measurement data. Compared with the benchmark SoftCast scheme, the proposed framework achieves a better error resilience performance and subjective quality when some packets are lost during transmission.","PeriodicalId":386952,"journal":{"name":"2017 Artificial Intelligence and Signal Processing Conference (AISP)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131705230","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}

{"title":"Presenting an efficient selection method for dynamic classifiers","authors":"A. Tafakkor, R. Boostani","doi":"10.1109/AISP.2017.8515160","DOIUrl":"https://doi.org/10.1109/AISP.2017.8515160","url":null,"abstract":"Accurate classification of multiclass data with complex distribution is still a challenge. One of the effective ways to deal with complex data is to design a dynamic classifier in which for a given test sample, the selected ensembles of classifiers (EoCs) are changed to provide the maximum accuracy. Nevertheless, the performance of dynamic classifiers is highly dependent to its selection mechanism, which sometimes selects improper EoCs. This study aims to present an efficient selection procedure to enhance the performance of the dynamic classification by assignment of competence value to the labeling process. Prior to the classification phase, the probability distribution of samples for each class is separately estimated and an importance value (weight) is assigned to each sample according to its probability on the distribution. Next, the weight of samples is incorporated to the training phase and to find a better set of EoCs, multi-objective genetic algorithm is used to select those with higher accuracy simultaneous with diminishing the complexity. To evaluate the proposed scheme, nine datasets which coverall different aspects were driven from UCI database. Empirical results imply the superiority of the proposed method compared to the conventional dynamic selection methods.","PeriodicalId":386952,"journal":{"name":"2017 Artificial Intelligence and Signal Processing Conference (AISP)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132965277","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}

{"title":"A new method for malware detection using opcode visualization","authors":"F. Manavi, A. Hamzeh","doi":"10.1109/AISP.2017.8324117","DOIUrl":"https://doi.org/10.1109/AISP.2017.8324117","url":null,"abstract":"Malware is a program that is developed with malicious purpose, such as sabotage the computer system, information theft or other malicious actions. Various methods have been defined for detecting and classifying malware. This paper proposes a new malware detection method based on the opcodes within an executable file by using image processing techniques. In opcode level, the proposed method shows promising results with less complexity in comparison with previous studies. There are several steps in the proposed method, which includes generating a graph of operational codes (opcodes) from an executable file and converting this graph to an image and then using “GIST” method in order to extract features from each image. In the final step machine learning methods such as Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Ensemble are used for classification.","PeriodicalId":386952,"journal":{"name":"2017 Artificial Intelligence and Signal Processing Conference (AISP)","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132383124","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}

{"title":"Single sample face identification utilizing sparse discriminative multi manifold embedding","authors":"Z. Azimifar, A. Nazemi, Fatemeh Shahali","doi":"10.1109/AISP.2017.8324123","DOIUrl":"https://doi.org/10.1109/AISP.2017.8324123","url":null,"abstract":"This paper describes three methods to improve single sample dataset face identification. The recent approaches to address this issue use intensity and do not guarantee for the high accuracy under uncontrolled conditions. This research presents an approach based on Sparse Discriminative Multi Manifold Embedding (SDMME), which uses feature extraction rather than intensity and normalization for pre-processing to reduce the effects of uncontrolled condition such as illumination. In the worst case of illumination this study improves identification accuracy about 17% compare to current methods.","PeriodicalId":386952,"journal":{"name":"2017 Artificial Intelligence and Signal Processing Conference (AISP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129241550","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}

{"title":"An optimal motion planning and obstacle avoidance algorithm based on the finite time velocity obstacle approach","authors":"Sepehr Samavati, M. Zarei, M. T. Masouleh","doi":"10.1109/AISP.2017.8324091","DOIUrl":"https://doi.org/10.1109/AISP.2017.8324091","url":null,"abstract":"This paper addresses a collision-free motion planning algorithm of mobile robots based on the synergy of discrete motion planning and optimal Finite-time Velocity Obstacle (FVO). The proposed approach is employed for motion planning of a mobile robot to the end of passing through an unknown environment. In this regard, predicated on the non-convex nature of the FVO constraints, they are approximated by a parabolic function and the approximated function is used to calculate the optimal next step velocity of the mobile robot in order to ensure the collision free motion. The reported results reveal that by considering the maximum velocity of the mobile robot, obtained computation time is less than 0.00015 seconds in each stage for single mobile robot scenarios which can be considered fast enough for robot motion planning tasks for the mobile robot under study.","PeriodicalId":386952,"journal":{"name":"2017 Artificial Intelligence and Signal Processing Conference (AISP)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129252848","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}

{"title":"A novel PCA perspective mapping for robust lane detection in urban streets","authors":"A. Bosaghzadeh, Seidfarbod Seidali Routeh","doi":"10.1109/AISP.2017.8324126","DOIUrl":"https://doi.org/10.1109/AISP.2017.8324126","url":null,"abstract":"Lane detection has been an active research subject in recent years which has a wide range of applications in intelligent transportation systems. Various approaches have been proposed to solve this problem but most of them suffer from lack of robustness towards noise, illumination and occlusion. In this paper, we propose a novel lane detection method which consists of three major parts:(1) lane detection, (2) perspective mapping and (3) lane selection. In order to solve the perspective issue in the acquisition of input image, we introduce a novel method based on Principal Component Analysis (PCA). More over, we employ a rotated rectangle model which leads to more accurate lane detection. The proposed method is evaluated through five different scenarios namely, occlusion, illumination change, crowded scene, noncrowded scene and noisy image. Experimental results proved that our method is accurate and robust to aforementioned scenarios. A comparison between a respectable lane detection method and ours demonstrates that the proposed method has better accuracy and robustness.","PeriodicalId":386952,"journal":{"name":"2017 Artificial Intelligence and Signal Processing Conference (AISP)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115936213","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}