Search

Face recognition using an image-processing technique is a valuable and effective approach to verifying the identity of a human being. Authorizing users based on human face identification remains one of the biggest challenges. However, with the improvement of deep learning and the creation of deep convolutional neural networks, the performance of face recognition has significantly improved. However, the outcomes from different models and networks are quite different in terms of accuracy. This paper proposes a face recognition model based on a Convolutional neural network (CNN). The recognition rate of the AT&T and AR data set on the proposed model is 99.17% and 99.12%. more...

Deep metric learning (DML) has achieved state-of-the-art results in several deep learning applications. However, this type of deep learning models has not been tested on the classification of electrical brain waves (EEG) for brain computer interface (BCI) applications. For the first time, we propose a triplet network to classify motor imagery (MI) EEG signals. Stockwell Transform has been used for converting the EEG signals in the time domain into the frequency domain, which resulted in improved DML classification accuracy in comparison to DML with Short Term Fourier Transform (0.647 vs. 0.431). DML model was trained with a topogram of concatenated 64 EEG channel spectrograms. The training batch was comprised of triplet pairs of the anchor, positive, and negative labeled epochs. The triplet network was able to train an embedding feature space that minimized the Euclidean distance between the embeddings of spectrograms of the same class and increased the distance between the embeddings of different labeled images. The proposed method has been tested on an EEG dataset of 109 untrained subjects. We showed that the DML classifier is able to converge with an extremely small number of training samples (~ 120 EEG trials) for only one subject per model, mitigating the well-known issue of the large inter-individual variability of human MI-BCI EEG which degrades the classification performance. The proposed preprocessing pipeline and the Triplet Network provide a promising method to classify MI-BCI EEG signals with much less training samples than the previous methods. more...

The effective educational systems aimed at improving Cognitive Skills (CS) both in and out of institutions. Such system relies on timely prediction of CS during students' activities in intuitions. Meanwhile, literature is saturated with the number of approaches which have used study schedules, biological and environmental factors to predict CS. However, the loopholes in prior studies have become the main source of inspiration for the current attempt. In this study, we propose a Bayesian Neural Network which predicts CS by iterative manipulations of CS under the profound influence of Student's Basic Attributes. Initially, the study classifies the Basic Attributes into three factors (1.age group, 2. gender, and 3. parent's cohabitation status) which have multiple layers. Furthermore, the technique splits the range of CS into 20 periodic outcome variables (with a period of 0.5). Eventually, the network iteratively estimates each outcome of CS by feed-forward process through Basic Attributes layers. We have reviewed the performance of the proposed network by using a students' score dataset. The results have illustrated that the current technique obtained significant prediction accuracy in terms of accuracy measures. more...

This paper presents novel reduced size diagonal slot embedded Triangular Gasket fractal patch antenna for multiband wireless applications. The diagonal slot geometry is embedded in triangular patch antenna for initial size reduction of the basic cell and further miniaturization is achieved by fractalization of the cell upto third order. Three frequency resonances of 2.4, 6.5 and 9.23 GHz are optimized for WiFi, WiMAX and WLAN (Wireless Local Area Network) applications in the S, C and X frequency bands. Size reduction upto 55.32 and 80.74% is achieved in terms of whole antenna area and copper cladding remaining respectively in comparison to triangular patch antenna. Antenna 3D (Three-Dimensional) modeling, simulation and optimization for the desired BW (Bandwidth) and gain requirements are done in HFSS (High Frequency Structure Simulator) and CST (Computer Simulation Technology). Effects of embedding diagonal slot in the basic triangular cell is analyzed in terms of the diagonal slot length and width with corresponding frequency response variations. Microstrip feed line dimensions and ground plane separation from the radiating top layer is optimized for achieving acceptable BWs less than -10 dB for the desired multiple resonances. E-field, H-field, current density and surface current plots are presented to verify the radiations for multiband wireless applications. Proposed slot embedded fractal radiator is fabricated and measured frequency responses and gain patterns are demonstrated in comparison with simulated results for verification of the concept. more...

This paper presents novel mobile phone antenna for radiations simultaneously in ten frequency bands for applications of LTE (Long-Term Evolution), GPS (Global Positioning System), GSM (Global System for Mobile Communications), PCS (Public and Commercial Services), DCS (Distributed Control System), UMTS (Universal Mobile Telecommunications System) and WiMAX (Worldwide Interoperability for Microwave Access). The antenna demonstrates novel characteristics for resolving the challenge of simultaneous radiation capabilities for the GSM and GPS applications with frequencies of 850/900 and 1575 MHz respectively, along with resonances at frequency band of 700 MHz for LTE applications, 1800 and 1900 MHz for PCS and DCS applications, 2100 MHz for UMTS applications, 2400 and 2500 MHz for LTE applications and 3300 MHz for WiMAX applications with S11 = - 6 dB matching criteria (VSWR 3:1). The antenna architecture comprises top and bottom copper layers embedded with monopole radiating element, branch line, slots and various stub lines. The desired operating bands are achieved in a compact area with overall dimensions of 0.8x60x120 mm for the height, width and length respectively of the antenna providing suitable platform for mobile handset applications. Omni-directional radiation pattern characteristics are achieved throughout the range of frequencies by designing the proposed antenna in monopole configuration. Proposed antenna is fabricated and results for the surface currents, s-parameters and 3D (Three-Dimensional) gain plots are illustrated for the proof of concept. more...

Slotted circular printed layered patch antenna is designed, simulated and fabricated for 5G (Fifth Generation) wireless communication applications. The antenna consists of slots in the main radiating circular patch element for miniaturizing the size of the radiating element and providing dual band radiation characteristics. The feed line is separated on bottom substrate layer with EBG (Electromagnetic Band-Gap) embedded for enhancing the gain characteristics of the antenna. Superstrate layer is also used for improving the gain of the antenna where the distance from the radiating antenna element is optimized for maximizing the impedance bandwidth and radiation characteristics. The feed realization and impedance matching of the radiating slotted circular patch antenna is done by inducing slot at the middle ground plane of the slot embedded circular patch antenna system. The proposed configuration provides power radiation gain values of more than 5 dB for the Ka band of communications, whereas the impedance bandwidth of the antenna is verified for the dual resonances at 27.5 and 28.5 GHz. Dual band radiation characteristics are attained by embedding and optimizing the slot length and width in the circular patch radiator element that is placed on the upper face of the substrate RT Rogers Duroid 5880 layer. The length of the microstrip feed line embedded in the lower layer of the substrate is optimized for providing required bandwidth characteristics for the dual frequency point radiations. The antenna configuration is designed, modeled and simulated in CST (Central Standard Time) Microwave studio. The antenna is fabricated and measured vs simulated frequency response, gain patterns and current density plots are presented for the verification of antenna operation in the desired frequency bands. more...

Semantic segmentation for accurate visual perception is a critical task in computer vision. In principle, the automatic classification of dynamic visual scenes using predefined object classes remains unresolved. The challenging problems of learning deep convolution neural networks, specifically ResNet-based DeepLabV3+ (the most recent version), are threefold. The problems arise due to (1) biased centric exploitations of filter masks, (2) lower representational power of residual networks due to identity shortcuts, and (3) a loss of spatial relationship by using per-pixel primitives. To solve these problems, we present a proficient approach based on DeepLabV3+, along with an added evaluation metric, namely, Unified DeepLabV3+ and S3core, respectively. The presented unified version reduced the effect of biased exploitations via additional dilated convolution layers with customized dilation rates. We further tackled the problem of representational power by introducing non-linear group normalization shortcuts to solve the focused problem of semi-dark images. Meanwhile, to keep track of the spatial relationships in terms of the global and local contexts, geometrically bunched pixel cues were used. We accumulated all the proposed variants of DeepLabV3+ to propose Unified DeepLabV3+ for accurate visual decisions. Finally, the proposed S3core evaluation metric was based on the weighted combination of three different accuracy measures, i.e., the pixel accuracy, IoU (intersection over union), and Mean BFScore, as robust identification criteria. Extensive experimental analysis performed over a CamVid dataset confirmed the applicability of the proposed solution for autonomous vehicles and robotics for outdoor settings. The experimental analysis showed that the proposed Unified DeepLabV3+ outperformed DeepLabV3+ by a margin of 3% in terms of the class-wise pixel accuracy, along with a higher S3core, depicting the effectiveness of the proposed approach. more...

Super-pixels represent perceptually similar visual feature vectors of the image. Super-pixels are the meaningful group of pixels of the image, bunched together based on the color and proximity of singular pixel. Computation of super-pixels is highly affected in terms of accuracy if the image has high pixel intensities, i.e., a semi-dark image is observed. For computation of super-pixels, a widely used method is SLIC (Simple Linear Iterative Clustering), due to its simplistic approach. The SLIC is considerably faster than other state-of-the-art methods. However, it lacks in functionality to retain the content-aware information of the image due to constrained underlying clustering technique. Moreover, the efficiency of SLIC on semi-dark images is lower than bright images. We extend the functionality of SLIC to several computational distance measures to identify potential substitutes resulting in regular and accurate image segments. We propose a novel SLIC extension, namely, SLIC++ based on hybrid distance measure to retain content-aware information (lacking in SLIC). This makes SLIC++ more efficient than SLIC. The proposed SLIC++ does not only hold efficiency for normal images but also for semi-dark images. The hybrid content-aware distance measure effectively integrates the Euclidean super-pixel calculation features with Geodesic distance calculations to retain the angular movements of the components present in the visual image exclusively targeting semi-dark images. The proposed method is quantitively and qualitatively analyzed using the Berkeley dataset. We not only visually illustrate the benchmarking results, but also report on the associated accuracies against the ground-truth image segments in terms of boundary precision. SLIC++ attains high accuracy and creates content-aware super-pixels even if the images are semi-dark in nature. Our findings show that SLIC++ achieves precision of 39.7%, outperforming the precision of SLIC by a substantial margin of up to 8.1%. more...

This paper addresses ELSP (Economic Lot Scheduling Problem) using EBPA (Extended Basic Period Approach) model with PoT (Power of Two) policy. The objective is to solve the ELSP using HCA (Hybrid Cuckoo Search Algorithm). The proposed approach improves the solution (i.e. minimizes the total cost which is the sum of setup and inventory holding costs) obtained through GA (Genetic Algorithm).The solution obtained from HCA is compared with GA on 17 Bomberger?s problems. The comparison indicates the superiority of the proposed HCA over GA with respect to the solution quality. more...

Several real-time visual monitoring applications such as surveillance, mental state monitoring, driver drowsiness and patient care, require equipping high-quality cameras with wireless sensors to form visual sensors and this creates an enormous amount of data that has to be managed and transmitted at the sensor node. Moreover, as the sensor nodes are battery-operated, power utilization is one of the key concerns that must be considered. One solution to this issue is to reduce the amount of data that has to be transmitted using specific compression techniques. The conventional compression standards are based on complex encoders (which require high processing power) and simple decoders and thus are not pertinent for battery-operated applications, i.e., VSN (primitive hardware). In contrast, compressive sensing (CS) a distributive source coding mechanism, has transformed the standard coding mechanism and is based on the idea of a simple encoder (i.e., transmitting fewer data-low processing requirements) and a complex decoder and is considered a better option for VSN applications. In this paper, a CS-based joint decoding (JD) framework using frame prediction (using keyframes) and residual reconstruction for single-view video is proposed. The idea is to exploit the redundancies present in the key and non-key frames to produce side information to refine the non-key frames’ quality. The proposed method consists of two main steps: frame prediction and residual reconstruction. The final reconstruction is performed by adding a residual frame with the predicted frame. The proposed scheme was validated on various arrangements. The association among correlated frames and compression performance is also analyzed. Various arrangements of the frames have been studied to select the one that produces better results. The comprehensive experimental analysis proves that the proposed JD method performs notably better than the independent block compressive sensing scheme at different subrates for various video sequences with low, moderate and high motion contents. Also, the proposed scheme outperforms the conventional CS video reconstruction schemes at lower subrates. Further, the proposed scheme was quantized and compared with conventional video codecs (DISCOVER, H-263, H264) at various bitrates to evaluate its efficiency (rate-distortion, encoding, decoding). more...

This paper proposes an intrusion detection technique based on DBN (Deep Belief Network) to classify four intrusion classes and one normal class using KDD-99 dataset. The proposed technique is based on two phases: in first phase it removes the class imbalance problem and in the next, it applies DBN followed by FFNN (Feed-Forward Neural Network) to build a prediction model. The obtained results are compared with those given in [9]. The prediction accuracy of our model shows promising results on both intrusion and normal patterns more...

Devices in a visual sensor network (VSN) are mostly powered by batteries, and in such a network, energy consumption and bandwidth utilization are the most critical issues that need to be taken into consideration. The most suitable solution to such issues is to compress the captured visual data before transmission takes place. Compressive sensing (CS) has emerged as an efficient sampling mechanism for VSN. CS reduces the total amount of data to be processed such that it recreates the signal by using only fewer sampling values than that of the Nyquist rate. However, there are few open issues related to the reconstruction quality and practical implementation of CS. The current studies of CS are more concentrated on hypothetical characteristics with simulated results, rather than on the understanding the potential issues in the practical implementation of CS and its computational validation. In this paper, a low power, low cost, visual sensor platform is developed using an Arduino Due microcontroller board, XBee transmitter, and uCAM-II camera. Block compressive sensing (BCS) is implemented on the developed platform to validate the characteristics of compressive sensing in a real-world scenario. The reconstruction is performed by using the joint multi-phase decoding (JMD) framework. To the best of our knowledge, no such practical implementation using off the shelf components has yet been conducted for CS. more...

Radial basis function neural networks are used in a variety of applications such as pattern recognition, nonlinear identification, control and time series prediction. In this paper, the learning algorithm of radial basis function neural networks is analyzed in a feedback structure. The robustness of the learning algorithm is discussed in the presence of uncertainties that might be due to noisy perturbations at the input or to modeling mismatch. An intelligent adaptation rule is developed for the learning rate of RBFNN which gives faster convergence via an estimate of error energy while giving guarantee to the l2 stability governed by the upper bounding via small gain theorem. Simulation results are presented to support our theoretical development. more...