Intelligent Transport System: Classification of Traffic Signs Using Deep Neural Networks in Real Time

Traffic control has been one of the most common and irritating problems since the time automobiles have hit the roads. Problems like traffic congestion have led to a significant time burden around the world and one significant solution to these problems can be the proper implementation of the Intelligent Transport System (ITS). It involves the integration of various tools like smart sensors, artificial intelligence, position technologies and mobile data services to manage traffic flow, reduce congestion and enhance driver's ability to avoid accidents during adverse weather. Road and traffic signs’ recognition is an emerging field of research in ITS. Classification problem of traffic signs needs to be solved as it is a major step in our journey towards building semi-autonomous/autonomous driving systems. The purpose of this work focuses on implementing an approach to solve the problem of traffic sign classification by developing a Convolutional Neural Network (CNN) classifier using the GTSRB (German Traffic Sign Recognition Benchmark) dataset. Rather than using hand-crafted features, our model addresses the concern of exploding huge parameters and data method augmentations. Our model achieved an accuracy of around 97.6% which is comparable to various state-of-the-art architectures.

Simulation Based Performance Comparison of Different Control Methods of ZSI Feeding Industrial Drives

Industrial drives are source of serious power quality problems. In this, two typical industrial drives have been dealt with, namely, FOC induction motor drives and DTC induction motor drive. The Z-source inverter is an emerging topology of power electronic converters which is capable of buck boost characteristics. The performances of different control methods based Z-source inverters feeding these industrial drives have been investigated, in this work. The test systems have been modeled and simulated in MATLAB/SIMULINK. The results obtained after carrying out these simulations have been used to draw the conclusions.

Performance Analysis of IDMA Scheme Using Quasi-Cyclic Low Density Parity Check Codes

The next generation mobile communication systems i.e. fourth generation (4G) was developed to accommodate the quality of service and required data rate. This project focuses on multiple access technique proposed in 4G communication systems. It is attempted to demonstrate the IDMA (Interleave Division Multiple Access) technology. The basic principle of IDMA is that interleaver is different for each user whereas CDMA employs different signatures. IDMA inherits many advantages of CDMA such as robust against fading, easy cell planning; dynamic channel sharing and IDMA increase the spectral efficiency and reduce the receiver complexity. In this, performance of IDMA is analyzed using QC-LDPC coding scheme further it is compared with LDPC coding and at last BER is calculated and plotted in MATLAB.

Heart Rate Variability Analysis for Early Stage Prediction of Sudden Cardiac Death

In present scenario, cardiovascular problems are growing challenge for researchers and physiologists. As heart disease have no geographic, gender or socioeconomic specific reasons; detecting cardiac irregularities at early stage followed by quick and correct treatment is very important. Electrocardiogram is the finest tool for continuous monitoring of heart activity. Heart rate variability (HRV) is used to measure naturally occurring oscillations between consecutive cardiac cycles. Analysis of this variability is carried out using time domain, frequency domain and non-linear parameters. This paper presents HRV analysis of the online dataset for normal sinus rhythm (taken as healthy subject) and sudden cardiac death (SCD subject) using all three methods computing values for parameters like standard deviation of node to node intervals (SDNN), square root of mean of the sequences of difference between adjacent RR intervals (RMSSD), mean of R to R intervals (mean RR) in time domain, very low-frequency (VLF), low-frequency (LF), high frequency (HF) and ratio of low to high frequency (LF/HF ratio) in frequency domain and Poincare plot for non linear analysis. To differentiate HRV of healthy subject from subject died with SCD, k –nearest neighbor (k-NN) classifier has been used because of its high accuracy. Results show highly reduced values for all stated parameters for SCD subjects as compared to healthy ones. As the dataset used for SCD patients is recording of their ECG signal one hour prior to their death, it is therefore, verified with an accuracy of 95% that proposed algorithm can identify mortality risk of a patient one hour before its death. The identification of a patient’s mortality risk at such an early stage may prevent him/her meeting sudden death if in-time and right treatment is given by the doctor.

Numerical Studies on Thrust Vectoring Using Shock-Induced Self Impinging Secondary Jets

Numerical studies have been carried out using a validated two-dimensional standard k-omega turbulence model for the design optimization of a thrust vector control system using shock induced self-impinging supersonic secondary double jet. Parametric analytical studies have been carried out at different secondary injection locations to identifying the highest unsymmetrical distribution of the main gas flow due to shock waves, which produces a desirable side force more lucratively for vectoring. The results from the parametric studies of the case on hand reveal that the shock induced self-impinging supersonic secondary double jet is more efficient in certain locations at the divergent region of a CD nozzle than a case with supersonic single jet with same mass flow rate. We observed that the best axial location of the self-impinging supersonic secondary double jet nozzle with a given jet interaction angle, built-in to a CD nozzle having area ratio 1.797, is 0.991 times the primary nozzle throat diameter from the throat location. We also observed that the flexible steering is possible after invoking ON/OFF facility to the secondary nozzles for meeting the onboard mission requirements. Through our case studies we concluded that the supersonic self-impinging secondary double jet at predesigned jet interaction angle and location can provide more flexible steering options facilitating with 8.81% higher thrust vectoring efficiency than the conventional supersonic single secondary jet without compromising the payload capability of any supersonic aerospace vehicle.

Optimization of End Milling Process Parameters for Minimization of Surface Roughness of AISI D2 Steel

The present work analyses different parameters of end milling to minimize the surface roughness for AISI D2 steel. D2 Steel is generally used for stamping or forming dies, punches, forming rolls, knives, slitters, shear blades, tools, scrap choppers, tyre shredders etc. Surface roughness is one of the main indices that determines the quality of machined products and is influenced by various cutting parameters. In machining operations, achieving desired surface quality by optimization of machining parameters, is a challenging job. In case of mating components the surface roughness become more essential and is influenced by the cutting parameters, because, these quality structures are highly correlated and are expected to be influenced directly or indirectly by the direct effect of process parameters or their interactive effects (i.e. on process environment). In this work, the effects of selected process parameters on surface roughness and subsequent setting of parameters with the levels have been accomplished by Taguchi’s parameter design approach. The experiments have been performed as per the combination of levels of different process parameters suggested by L9 orthogonal array. Experimental investigation of the end milling of AISI D2 steel with carbide tool by varying feed, speed and depth of cut and the surface roughness has been measured using surface roughness tester. Analyses of variance have been performed for mean and signal-to-noise ratio to estimate the contribution of the different process parameters on the process.

Statistical Optimization of Enzymatic Hydrolysis of Potato (Solanum tuberosum) Starch by Immobilized α-amylase

Enzymatic hydrolysis of starch from natural sources finds potential application in commercial production of alcoholic beverage and bioethanol. In this study the effect of starch concentration, temperature, time and enzyme concentration were studied and optimized for hydrolysis of Potato starch powder (of mesh 80/120) into glucose syrup by immobilized (using Sodium arginate) α-amylase using central composite design. The experimental result on enzymatic hydrolysis of Potato starch was subjected to multiple linear regression analysis using MINITAB 14 software. Positive linear effect of starch concentration, enzyme concentration and time was observed on hydrolysis of Potato starch by α-amylase. The statistical significance of the model was validated by F-test for analysis of variance (p ≤ 0.01). The optimum value of starch concentration, enzyme concentration, temperature, time and were found to be 6% (w/v), 2% (w/v), 40°C and 80min respectively. The maximum glucose yield at optimum condition was 2.34 mg/mL.

Friction Stir Welding of Aluminum Alloys: A Review

Friction stir welding is a solid state joining process. High strength aluminum alloys are widely used in aircraft and marine industries. Generally, the mechanical properties of fusion welded aluminum joints are poor. As friction stir welding occurs in solid state, no solidification structures are created thereby eliminating the brittle and eutectic phases common in fusion welding of high strength aluminum alloys. In this review the process parameters, microstructural evolution, and effect of friction stir welding on the properties of weld specific to aluminum alloys have been discussed.

Multicast Optimization Techniques using Best Effort Genetic Algorithms

Multicast Network Technology has pervaded our lives-a few examples of the Networking Techniques and also for the improvement of various routing devices we use. As we know the Multicast Data is a technology offers many applications to the user such as high speed voice, high speed data services, which is presently dominated by the Normal networking and the cable system and digital subscriber line (DSL) technologies. Advantages of Multi cast Broadcast such as over other routing techniques. Usually QoS (Quality of Service) Guarantees are required in most of Multicast applications. The bandwidth-delay constrained optimization and we use a multi objective model and routing approach based on genetic algorithm that optimizes multiple QoS parameters simultaneously. The proposed approach is non-dominated routes and the performance with high efficiency of GA. Its betterment and high optimization has been verified. We have also introduced and correlate the result of multicast GA with the Broadband wireless to minimize the delay in the path.

An Experimental Comparison of Unsupervised Learning Techniques for Face Recognition

Face Recognition has always been a fascinating research area. It has drawn the attention of many researchers because of its various potential applications such as security systems, entertainment, criminal identification etc. Many supervised and unsupervised learning techniques have been reported so far. Principal Component Analysis (PCA), Self Organizing Maps (SOM) and Independent Component Analysis (ICA) are the three techniques among many others as proposed by different researchers for Face Recognition, known as the unsupervised techniques. This paper proposes integration of the two techniques, SOM and PCA, for dimensionality reduction and feature selection. Simulation results show that, though, the individual techniques SOM and PCA itself give excellent performance but the combination of these two can also be utilized for face recognition. Experimental results also indicate that for the given face database and the classifier used, SOM performs better as compared to other unsupervised learning techniques. A comparison of two proposed methodologies of SOM, Local and Global processing, shows the superiority of the later but at the cost of more computational time.