{"title":"Inner-knuckle-print for human authentication by using ring and middle fingers","authors":"Farzam Kharaji Nezhadian, S. Rashidi","doi":"10.1109/ICSPIS.2016.7869849","DOIUrl":"https://doi.org/10.1109/ICSPIS.2016.7869849","url":null,"abstract":"This paper proposed a new approach for inner-knuckle-print recognition. Theinner knuckle print is one of the reliable physiological characteristics among different approaches that exist in biometric. In this paper, the image of the inner surface of the middle and ring fingers are used for human verification. We considered the inner knuckle print as a texture and applied two types of feature extraction methods, namely Gabor wavelet filters and wavelet energy. Among all feature that is extracted by these approaches, fifty superior features selected by the forward feature selection algorithm. Features are classified with a new approach by using K-nearest neighbor, fuzzy K-nearest neighbor, parzen window and support vector machine classifiers. In Hong Kong Polytechnic University Contact-free 3D/2D Hand Images testing database of 1770 whole hand samples from 177 subjects, we achieved Equal Error Rate of 4.79%±1.74, 6.14%±0.09, 3.70±0.57 by using K-Nearest Neighbor, parzen window and support vector machine classifier for ring, middle and both of them together respectively.","PeriodicalId":415332,"journal":{"name":"2016 2nd International Conference of Signal Processing and Intelligent Systems (ICSPIS)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133675784","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":"Investigation of RMF-SLAM and AMF-SLAM in closed loop and open loop paths","authors":"Sayed Farzad Bahreinian, M. Palhang, M. R. Taban","doi":"10.1109/ICSPIS.2016.7869889","DOIUrl":"https://doi.org/10.1109/ICSPIS.2016.7869889","url":null,"abstract":"In this paper, performance of relative and absolute simultaneous localization and mapping (SLAM) are studied in dealing with open and closed loop paths. SLAM techniques based on absolute landmark and robot positions (AMF) has global look at the environment. This property is used to error reduction in closed loop paths. However, in open loop path, its error value increases and performance decreases strongly. In SLAM techniques based on relative positions of landmarks (RMF), the environment is observed locally by the robot. The performance of these techniques in paths with closed or open loops look alike without significant difference. One of the reasons why researchers must pay more attention to RMF SLAM in comparison with AMF SLAM methods is their provable convergence besides their significant performance especially in open loop path.","PeriodicalId":415332,"journal":{"name":"2016 2nd International Conference of Signal Processing and Intelligent Systems (ICSPIS)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123622614","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":"Modification of optimum-path forest using Markov cluster process algorithm","authors":"Hamid Bostani, M. Sheikhan","doi":"10.1109/ICSPIS.2016.7869874","DOIUrl":"https://doi.org/10.1109/ICSPIS.2016.7869874","url":null,"abstract":"Optimum-path forest (OPF) is a novel supervised graph-based classifier which reduces the classification problem into partitioning of vertices in a graph derived from the data samples. One of the main processes in OPF is identifying the optimum set of key samples named prototypes. This process is based on creating a minimum spanning tree on a complete weighted graph which is derived from the training samples; hence, it is much time-consuming for large-scale problems. In this study, for overcoming this limitation, the process of finding the prototypes in traditional OPF is modified by using Markov cluster (MCL) algorithm. The graph partitioning in MCL is based on finding key samples named attractors, which attract other related samples; so the obtained attractors can be selected as prototypes for generating optimum-path trees. Experiments on public benchmark datasets show that the speed of proposed modified OPF is improved considerably as compared to the traditional OPF.","PeriodicalId":415332,"journal":{"name":"2016 2nd International Conference of Signal Processing and Intelligent Systems (ICSPIS)","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121091342","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 lines Iranian license plate detection and recognition using subspace learning","authors":"R. Serajeh","doi":"10.1109/ICSPIS.2016.7869906","DOIUrl":"https://doi.org/10.1109/ICSPIS.2016.7869906","url":null,"abstract":"License plate detection and recognition has become high-priority research because of its many applications such as electronic payment systems, traffic monitoring and security control. This subject is divided into 2 main parts: license plate detection and character recognition. Although, there have been many studies in detection, two lines license plates such as Iranian motorcycle is less studied. Detection of this kind of license plates is more complicated because of background clutter. In order to solve this problem, this paper uses vertical edge detection and template matching. The following step is character recognition. There are many challenges such as skew, pose variations, blurriness, being dirty, illumination changes and some other distortions in real scenes. Our approach uses subspace learning to have proper performance under these conditions. To show the efficiency of our approach, the database including 60101 digits images is collected and the recognition algorithm is tested on it.","PeriodicalId":415332,"journal":{"name":"2016 2nd International Conference of Signal Processing and Intelligent Systems (ICSPIS)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133231063","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 statistical detector for additive image watermarking based on dual-tree complex wavelet transform","authors":"Milad Barazandeh, M. Amirmazlaghani","doi":"10.1109/ICSPIS.2016.7869886","DOIUrl":"https://doi.org/10.1109/ICSPIS.2016.7869886","url":null,"abstract":"This paper introduces a new detector for digital watermarking based on dual-tree complex wavelet transform (DT-CWT). The DT-CWT benefits from the high directionality and shift invariance, which ensure the high efficiency of proposed method. The watermark is additively embedded in the magnitude components of DT-CWT coefficients. Also, the watermark detection is formulated as a binary hypothesis test. So, the performance of the detector is dependent on the accuracy of the statistical modeling. In this paper, magnitude components of the coefficients are modeled by Gamma, Rayleigh and Weibull distributions. Also, the optimal detector is designed using the likelihood ratio test. The simulation results show that proposed method provides high imperceptibility and robustness against various types of attacks.","PeriodicalId":415332,"journal":{"name":"2016 2nd International Conference of Signal Processing and Intelligent Systems (ICSPIS)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120843722","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":"Facial expression recognition with discriminatory graphical models","authors":"Behzad Hasani, M. Arzani, M. Fathy, K. Raahemifar","doi":"10.1109/ICSPIS.2016.7869885","DOIUrl":"https://doi.org/10.1109/ICSPIS.2016.7869885","url":null,"abstract":"Discriminating probabilistic graphical models are reliable tools for a sequence labeling task. Conditional Random Fields (CRFs) are discriminative models which will enable us to label a sequence of input data. Other variations of CRFs have been proposed. Hidden Conditional Random Fields (HCRFs) incorporate hidden states to the CRF model and assign a label for the whole input sequence as the model's output. Latent-Dynamic Conditional Random Fields (LDCRFs) also incorporate hidden variable states to the CRFs, in addition, these models are able to label each output variables separately. These models can capture subtle changes among different classes which will help us to achieve better recognition results. In this work we experiment various models and settings in order to achieve better results in facial expression recognition from sequence of videos. We use CRF and LDCRF models and train them with Limited-memory BFGS and Conjugate Gradient parameter learning methods. For each model we use various feature vectors in order to achieve better recognition results. We use Active Appearance Model (AAM) landmark points, Histogram of Oriented Gradients (HOG) and Uniform Local Binary Pattern (U-LBP) as our feature vectors in our models. We show which combination of learning methods and feature vectors are suitable for CRF and LDCRF discriminative models.","PeriodicalId":415332,"journal":{"name":"2016 2nd International Conference of Signal Processing and Intelligent Systems (ICSPIS)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116850391","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":"Human activity recognition in ambient assisted living environments using a convex optimization problem","authors":"V. Ghasemi, A. Pouyan","doi":"10.1109/ICSPIS.2016.7869899","DOIUrl":"https://doi.org/10.1109/ICSPIS.2016.7869899","url":null,"abstract":"Human activity recognition (HAR) is of crucial importance in ambient assisted living (AAL) environments. In such environments, residents' data are collected via unobtrusive sensors to be further interpreted as human activities. Essential services are provided for the users according to the HAR results. In this paper, a novel scheme is represented for HAR, which is based on minimizing a convex objective function. For this purpose, given a sensor data stream, a belief vector (BV) is calculated for each sensor event. A BV is calculated by weighting the posterior probabilities of activities. It denotes the possibility of each activity, given a single sensor event. Then, the BVs are smoothed by minimizing a convex objective function. The objective function is formulated based on common assumptions in AAL environments. The final activity inference is based on the smoothed belief vectors. The proposed method is evaluated using a well-known and publicly available dataset. It is compared to three widely adopted HAR benchmarks, which are based on probabilistic learning graphical models. Simulations show that the proposed method outperforms the benchmarks, having an accuracy of 79.97% and an average F-measure of 78.2%.","PeriodicalId":415332,"journal":{"name":"2016 2nd International Conference of Signal Processing and Intelligent Systems (ICSPIS)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116253277","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":"FPGA-based convolutional neural network accelerator design using high level synthesize","authors":"Sina Ghaffari, Saeed Sharifian","doi":"10.1109/ICSPIS.2016.7869873","DOIUrl":"https://doi.org/10.1109/ICSPIS.2016.7869873","url":null,"abstract":"Convolutional Neural Networks are well known for their outstanding results in recent years in computer vision applications. Two hardware architectures for implementing these networks are proposed. The first one is more application specific which is suitable for smaller convolutional networks and the second one is more extendable and can be easily used for larger networks. A new method for computing hyperbolic tangent activation function is proposed to reduce computation time. LeNet convolutional network architecture is implemented to recognize handwritten digits using MNIST dataset as a sample. With the proposed architectures, it is indicated that latency is reduced while maintaining accuracy. Experiments are done on Xilinx Zynq FPGA using High Level Synthesis.","PeriodicalId":415332,"journal":{"name":"2016 2nd International Conference of Signal Processing and Intelligent Systems (ICSPIS)","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122785104","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}
Mohsen Soltanpour, K. Faez, Saeed Sharifian, V. Pourahmadi
{"title":"Enhance evoked potentials detection using RBF neural networks: Application to brain-computer interface","authors":"Mohsen Soltanpour, K. Faez, Saeed Sharifian, V. Pourahmadi","doi":"10.1109/ICSPIS.2016.7869850","DOIUrl":"https://doi.org/10.1109/ICSPIS.2016.7869850","url":null,"abstract":"Brain-Computer Interface (BCI) is a specific type of human-computer interface that stablish the direct communication between human and computers by analyzing brain activities. Oddball paradigms are used in BCI to generate Event-Related Potentials (ERPs), like the P300 response, on targets selected by the user. A P300 speller is based on this principle, where the detection of P300, allows the user to spell characters. The P300 speller is divided into in two classification problems. The first classification is for detect P300 in the electroencephalogram (EEG). The second one for determining the desired symbol by user. A new method for the detection of P300 waves is presented. This model is based on a Radial Basis Function Neural Network (RBFNN). The architecture of the network is adapted to the detection of P300 response in the time domain. Principal Component Analysis (PCA) is used for feature extraction. These models are tested and compared on the Data set III of BCI competition (2004). The proposed approach provides also a new way for analyzing brain activities due to the receptive field of the RBFNN models.","PeriodicalId":415332,"journal":{"name":"2016 2nd International Conference of Signal Processing and Intelligent Systems (ICSPIS)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128687292","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":"Maximum likelihood estimation of generalized linear models with generalized Gaussian residuals","authors":"Hamidreza Hakimdavoodi, Maryam Amirmazlghani","doi":"10.1109/ICSPIS.2016.7869893","DOIUrl":"https://doi.org/10.1109/ICSPIS.2016.7869893","url":null,"abstract":"Assumption of normally distributed residuals is one of the big challenges in the generalized linear models (GLM). Recently, generalized Gaussian distribution (GGD) is used widely to analyze and model heavy-tail signals. Consequently, investigations for robust estimation of regression coefficients have led us to introduce GG-GLM, which models the GLM residuals using GGD. This model can deal with broad range of residuals distributions that have lower or heavier tailed behavior than the Gaussian distribution. The model parameters are estimated by a maximum likelihood (ML) approach. Experimental results on the synthetic data confirms the superior performance of GG-GLM model in comparison with the GLM model during the heavy tailed behavior of data.","PeriodicalId":415332,"journal":{"name":"2016 2nd International Conference of Signal Processing and Intelligent Systems (ICSPIS)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115574811","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}