Jadhav D.V.,Vishwakarma Institute of Technology |
Holambe R.S.,SGGSIE and T
Pattern Recognition Letters | Year: 2010
This paper presents an in-plane rotation (tilt), illumination invariant pattern recognition framework based on the combination of the features extracted using Radon and discrete cosine transforms and kernel based learning for face recognition. The use of Radon transform enhances the low frequency components, which are useful for face recognition and that of DCT yields low dimensional feature vector. The proposed technique computes Radon projections in different orientations and captures the directional features of the face images. DCT applied on Radon projections provides frequency features. Further, polynomial kernel Fisher discriminant analysis implemented on these features enhances discrimination capability of these features. The technique is also robust to zero mean white noise. The feasibility of the proposed technique has been evaluated using FERET, ORL, and Yale databases. © 2009 Elsevier B.V. All rights reserved.
Sridharan B.,Vishwakarma Institute of Technology
2010 IEEE 2nd International Advance Computing Conference, IACC 2010 | Year: 2010
This paper describes a novel approach towards the modification of Genetic Algorithms. The novelty of the modified Genetic Algorithm lies in the addition of a new parameter called the age of the chromosome that would select its ability to reproduce. Also, the concept of dynamic population and elitism size has been introduced. The modified Genetic Algorithm converges to the near optimum value at a faster rate, i.e. lesser number of generations are required for the convergence and due to the concept of dynamic population size the results obtained are more accurate. Thus, the modified algorithm is observed to be computationally more efficient. The algorithm was tested for some standard functions and curves and the results were found to be highly satisfactory. ©2010 IEEE.
Hulwan D.B.,Vishwakarma Institute of Technology |
Joshi S.V.,Vishwakarma Institute of Technology
Applied Energy | Year: 2011
Feasibility of using high percentage of ethanol in diesel-ethanol blends, with biodiesel as a co-solvent and properties enhancer has been investigated. The blends tested are D70/E20/B10 (blend A), D50/E30/B20 (blend B) D50/E40/B10 (blend C), and Diesel (D100). The blends are prepared to get maximum percentage of oxygen content but keeping important properties such as density, viscosity and Cetane index within acceptable limits. Experiments are conducted on a multicylinder, DI diesel engine, whose original injection timing was 13° CA BTDC. The engine did not run on blends B and C at this injection timing and it was required to advance timing to 18° and 21° CA BTDC to enable the use of blends B and C respectively. However advancing injection timing almost doubled the NO emissions and increased peak firing pressure. The P-θ and net heat release diagrams shows that the combustion process of these blends delayed at low loads but approaches to the diesel fuel at high loads. The comparison of blend results with baseline diesel showed that brake specific fuel consumption increased considerably, thermal efficiency improved slightly, smoke opacity reduced remarkably at high loads. NO variation depends on operating conditions while CO emissions drastically increased at low loads. Blend B which replaced 50% diesel and having oxygen content up to 12.21% by weight has given satisfactory performance for steady state running mode up to 1600 RPM however, it does not showed any benefit on peak smoke emission during free acceleration test. © 2011 Elsevier Ltd.
Maghade D.K.,Vishwakarma Institute of Technology |
Patre B.M.,Shri Guru Gobind Singhji Institute of Engineering and Technology
ISA Transactions | Year: 2012
In this paper, a decentralized PI/PID controller design method based on gain and phase margin specifications for two-input-two-output (TITO) interactive processes is proposed. The decouplers are designed for systems to minimize the interaction between the loops, and the first order plus dead time (FOPDT) model is achieved for each decoupled subsystem based on the frequency response fitting. An independent PI/PID controller is designed for each reduced order decoupled subsystem to obtain the desired gain and phase margins, and the performance is verified on the original interactive system to show the effectiveness of the proposed design method for the general class of TITO systems. Simulation examples are incorporated to validate the usefulness of the presented algorithm. An experimentation is performed on the Level-Temperature reactor process to show the practical applicability of the proposed method for the interactive system. © 2012 ISA. Published by Elsevier Ltd. All rights reserved.
Shahni F.,Vishwakarma Institute of Technology |
Malwatkar G.M.,Vishwakarma Institute of Technology
Journal of Process Control | Year: 2011
Abstract: In this paper, a simple method to find the minimum output variance (MOV) with proportional-integral-derivative (PID) controller is proposed. An assessment of improved achievable PID performance is important as PID controllers are the most commonly used in industries. The restriction on the controller structure e.g. PID (the case in this paper) results worst output variance compared to minimum variance (MV) benchmark. The problem in assessing the PID-MOV rises a non-convex problem and for non-convex problem, no direct and simple solution is possible. In this paper the non-convex problem is solved with simple ring of iterations. Several simulation examples are incorporated to show the usefulness of proposal algorithm. © 2011 Elsevier Ltd. All rights reserved.
Shilaskar S.,Vishwakarma Institute of Technology |
Ghatol A.,K Js Educational Institutes
Expert Systems with Applications | Year: 2013
Machine learning has emerged as an effective medical diagnostic support system. In a medical diagnosis problem, a set of features that are representative of all the variations of the disease are necessary. The objective of our work is to predict more accurately the presence of cardiovascular disease with reduced number of attributes. We investigate intelligent system to generate feature subset with improvement in diagnostic performance. Features ranked with distance measure are searched through forward inclusion, forward selection and backward elimination search techniques to find subset that gives improved classification result. We propose hybrid forward selection technique for cardiovascular disease diagnosis. Our experiment demonstrates that this approach finds smaller subsets and increases the accuracy of diagnosis compared to forward inclusion and back-elimination techniques. © 2012 Elsevier Ltd. All rights reserved.
Mangate L.D.,Vishwakarma Institute of Technology |
Chaudhari M.B.,Vishwakarma Institute of Technology
International Journal of Thermal Sciences | Year: 2015
Synthetic jet is a novel flow phenomenon potentially useful for cooling of electronic devices. The heat transfer and acoustic characteristics of impinging synthetic jet are investigated experimentally using different configurations of diamond and oval shape orifices with same hydraulic diameter. The experiments are conducted at an excitation frequency of 100-250 Hz. The ratio of axial distance between the heated copper block and orifice plate to orifice diameter (z/d) is in the range of 0-16. The average heat transfer coefficient and the sound pressure level (SPL) are obtained for different configurations of the orifice. The maximum value of the average heat transfer coefficient using diamond and oval shape orifices is found to be 17% and 7% higher respectively as compared to circular orifice of the same hydraulic diameter at 200 Hz. For all the configurations of oval shape orifice experimented in the present study, Nusselt number gets maximized at the same location. However, no such behaviour is observed in case of diamond shape orifice. The sound pressure level (SPL) obtained in case of diamond and oval shape orifices is less by 7 dB as compared to circular orifice at an excitation frequency of 200 Hz. The oval shape orifice is found to perform better in terms of heat transfer and acoustic aspect. SPL obtained for all the orifice shapes considered in this study is less than 40 dB at 100 Hz which is acceptable as per the United States Environmental Protection Agency (USEPA) and therefore synthetic jet with these orifice shapes can practically be employed for cooling applications without any further efforts for noise reduction. © 2014 Elsevier Masson SAS. All rights reserved.
Padalkar A.,Vishwakarma Institute of Technology
Proceedings - IEEE International Conference on Information Processing, ICIP 2015 | Year: 2015
In this research, a system was designed and implemented on hardware to control speed and angular position of shaft of the gear box with reduction ratio 10:1 attached to a BLDC motor of 3000 RPM by taking feedback from internal hall sensors of motor. A microcontroller processes feedback and drives motor to its target angular position which is fed to it by user via USART. By processing same hall feedback at same time, speed of motor can also be controlled. System can be used for replacing costly servo motors in low precision application in industry such as box picking and placing. Also stepper motor can be replaced by this system to achieve more accurate position control as the system involves feedback. © 2015 IEEE.
Mali S.N.,Vishwakarma Institute of Technology |
Patil P.M.,Vishwakarma Institute of Technology |
Jalnekar R.M.,Vishwakarma Institute of Technology
Digital Signal Processing: A Review Journal | Year: 2012
Rapid growth in the demand and consumption of digital information in past decade has led to valid concerns over issues such as content security, authenticity and digital right management. Imperceptible data hiding in digital images is an excellent example of demonstration of handling these issues. Classical Cryptography is related with concealing the content of messages, whereas, Steganography is related with concealing the existence of communication by hiding the messages in cover. This paper presents a robust and secured method of embedding high volume of text information in digital Cover-images without incurring any perceptual distortion. It is robust against intentional or unintentional attacks such as image compression, tampering, resizing, filtering and Additive White Gaussian Noise (AWGN). The schemes available in the literature can deal with these attacks individually, whereas the proposed work is a single methodology that can survive all these attacks. Image Adaptive Energy Thresholding (AET) is used while selecting the embedding locations in frequency domain. Coding framework with Class Dependent Coding Scheme (CDCS) along with redundancy and interleaving of embedded information gives enhancement in data hiding capacity. Perceptual quality of images after data hiding has been tested using Peak Signal to Noise Ratio (PSNR) whereas statistical variations in selected Image Quality Measures (IQMs) are observed with respect to Steganalysis. The results have been compared with existing algorithms like STOOL in spatial domain, COX in DCT domain and CDMA in DWT domain. © 2011 Elsevier Inc.
Wagle S.,Vishwakarma Institute of Technology
2016 International Conference on Internet of Things and Applications, IOTA 2016 | Year: 2016
The emergence of the paradigm of Internet of Things (IoT) has necessitated the development of machine-to-machine (M2M) protocols geared towards wireless sensor network interfacing to the Internet and implementing machine learning algorithms over the cloud. This paper discusses the viability of the MQ Telemetry Transport (MQTT) protocol for such applications. This paper introduces MQTT along with its merits and demerits and suitability towards IoT applications. Then it outlines an implementation of a typical IoT application involving ubiquitous sensing, M2M communication, cloud computing and semantic data extraction. The results of this experiment are then analyzed. Finally, the paper looks at future improvements in the proposed architecture for widespread use. © 2016 IEEE.