Project Title : Solar Based Oxygen syntheses

Sanctioned Year : 2016

Amount : 2,50,000/-

Funding/Sanction agency : NAIN Sponsored

Status :

Principal Investigators : Dr.Sangamesh Sakri

Project Title : E-ticketing System using QF

Sanctioned Year : 2017

Amount : Rs.3,00,000/-

Funding/Sanction agency : NAIN Sponsored

Status :

Principal Investigators : Prof.Chandrasekhara S

Project Title : Facility for Study and Development of Protection Schemes for Power System

Sanctioned Year : 2023

Amount : 2000000

Funding/Sanction agency : VGST

Status :

Principal Investigators : Dr. Sangamesh Sakri (Principal Investigator) Prof. Priya Patil (Co-Principal Investigator)

Student Name : Somashekhar Swamy

Title of the Paper : Review on Significance Research on Enhancing the Quality of the Brain Image Using Neuro-Fuzzy System

Research Description : Computed tomography (CT) is the technique that uses X-ray equipment to produce detailed images of section inside the animal or human body. CT scanning is very beneficial for detecting the diseases like cancers, tumors in human body. But noise and Blur are the major factors that degrade the quality of CT images and makes difficult to diagnose. Reconstructing the CT images is the way to overcome these problems. This paper presents the study of various techniques that are suitable to reconstruct the CT images, along with the performance evaluation of various techniques. This survey mainly focuses on most prominent techniques like filtering technique (discrete wavelets, complex wavelets, and median filter), artificial neural network approach for image reconstruction and Median filtering technique, also discussed other denoising and deblurring techniques. The aim of this survey is to provide useful information on various deblurring and denoising techniques and to motivate research to implement the effective algorithm.

Faculty Coordinator : P.K.Kulkarni

Student Name : Somashekhar Swamy

Title of the Paper : A Comprehensive Study on Significance of Image De-Noising and De-Blurring Applications in Computed Tomography

Research Description : De-noising and De-blurring is the technique that uses procedures to subsidize and remove the strange and useless content in the images De-noising and de-blurring is very beneficial for detecting the diseases like cancers, tumors in human body. But noise and Blur are the major factors that degrade the quality of images and makes difficult to diagnose. Reconstructing the images is the way to overcome these problems. This paper presents the study of various applications that are suitable to reconstruct the images. This survey mainly focuses on most prominent applications in the field of image processing and how its applications are critical enough to avoid dissatisfied results. The aim of this survey is to provide useful information on various de-blurring and de-noising applications and to motivate research to implement the effective algorithm.

Faculty Coordinator : P.K.Kulkarni

Student Name : Somashekhar Swamy

Title of the Paper : Image Processing for Identifying Brain Tumor using Intelligent System

Research Description : MRI is the technique that uses X-ray equipment to produce detailed images of section inside the human body. MRI Scanning is very beneficial for detecting the diseases like cancer, tumors etc. in human body. But, the system noise during testing, the effect of spatial similar regions in the image makes it difficult to detect the actual region for diagnosis leading to miscalculations. The objective of the proposed research is to develop a robust recognition system, developing a new algorithmic approach to over come the effect of de noising and de blurring of MRI images, increase the sensitivity and reduce the time for analysis.

Faculty Coordinator : P.K.Kulkarni

Student Name : Somashekhar Swamy

Title of the Paper : ADAPTIVE APPROACH TO IMAGE CODING FOR TUMOR DETECTION IN MRI IMAGES

Research Description : In the process of image coding, images are processed to retrieve important information’s from a given sample to achieve the objective of information retrieval in image coding. In the application towards processing of medical image data, images are processed for filtration and segmentation to retrieve proper regions for effective region detection. Towards this approach, the conventional coding uses median filtration and region based segmentation approach to localize effective regions. However, this coding approach is observed to be erroneous under dynamic conditions, which leads to misclassification of image data in adaptive manner. To develop an approach for improvising the region localization, in this paper an adaptive filtration approach with content based segmentation model is proposed. The suggestive approach observed to be improved for filtration and region localization tested over MRI samples, as compared to the conventional segmentation system.

Faculty Coordinator : P.K.Kulkarni

Student Name : Somashekhar Swamy

Title of the Paper : An automated detection and classification approach in MRI tumor diagnosis

Research Description : In the process of image coding, external noises impact a lot in processing efficiency. In the application of medical image processing, this effect is more, important due to its finer content details. It is required to minimize the noise effect with preserving the image content information, without losing the image generality. Towards the objective of image denoising, in this work, a dynamic block coding approach for noise minimization in medical image processing is presented. The second observing factor in region segmentation is the marking of small region patterns which are derived due to misclassification of actual and detected regions. the complexity of detection logic, due to recurrent coding is an additional factor to observe. In this paper, a new recurrent coding approach of region segmentation is proposed, overcoming the issue of region marking, discontinuity issue and small region miss-classification. The suggested approach is a simpler and robust to region detection, test over different MRI samples.

Faculty Coordinator : P.K.Kulkarni

Student Name : Somashekhar Swamy

Title of the Paper : Region Localization and Segmentation using Recurrent Morphological Coding

Research Description : In the process if region localization, images are processed in multiple orders to finalize the segmentation region. In progress to region segmentation, the evolved regions are majorly been discarded due to loss of intermediate information‟s. These information losses are generated due to trace pixel enlargement or natural discontinuity. The second observing factor in region segmentation is the marking of small region patterns which are derived due to misclassification of actual and detected regions. the complexity of detection logic, due to recurrent coding is an additional factor to observe. In this paper, a new recurrent coding approach of region segmentation is proposed, overcoming the issue of region marking, discontinuity issue and small region miss-classification. The suggested approach is a simpler and robust to region detection, test over different MRI samples.

Faculty Coordinator : P.K.Kulkarni

Student Name : Somashekhar Swamy

Title of the Paper : An Adaptive Learning and Classifier Model in MRI Tumor Detection

Research Description : In the process of image coding, external noises impact a lot in processing efficiency. In the application of medical image processing, this effect is more, important due to its finer content details. It is required to minimize the noise effect with preserving the image content information, without losing the image generality. Towards the objective of image denoising, in this work, a dynamic block coding approach for noise minimization in medical image processing is presented. The filtration approach is an enhancement to the objective of noise elimination using median filtration. The suggested approach, improves the retrieval accuracy more effectively under variant noise condition in consideration to conventional filtration approach.

Faculty Coordinator : P.K.Kulkarni

Student Name : Somashekhar Swamy

Title of the Paper : Density Driven Image Coding for Tumor Detection in mri Image

Research Description : The significant of multi spectral band resolution is explored towards selection of feature coefficients based on its energy density. Toward the feature representiaon in transformed domain, multi wavelet transformations were used for finer spectral representation. However, due to a large feature count these features are not optimal under low resource computing system. In the recognition units, running with low resources a new coding approach of feature selection, considering the band spectral density is developed. The effective selection of feature element, based on its spectral density achieve two objective of pattern recognition, the feature coefficient representiaon is minimized, hence leading to lower resource requirement, and dominant feature representation, resulting in higher retrieval performance.

Faculty Coordinator : P.K.Kulkarni

Student Name : Virupakshappa

Title of the Paper : Computer Based Diagnosis System for Tumor Detection &Classification: A Hybrid Approach

Research Description : Brain tumor is one among the most dangerous diseases in the world, patient’s life can be saved if the brain tumor is detected and diagnosed properly in its earliest stages. Since brain has the most complex structure in which tissues are interconnected rigorously. Thus makes the brain tumor detection a challenging task. Brain tumor detection and classification requires clinical experts to meet the standard level of accuracy. This limitation is overcome by the use of Computer Aided Diagnosis Systems (CAD Systems) in the diagnosis of brain tumors. In this paper we propose an efficient method for brain tumor detection and classification using hybrid method in which segmentation is carried out using Spatial Fuzzy Clustering, texture features are extracted using Gabor feature extraction method and finally classification using Artificial Neural Network (ANN) classifier. The system performance is examined with 40 trained images with 60 tested MRI scanned images. The comparative analysis in terms of accuracy with reference to the confusion matrix is presented in result section. From the experimental results we were able to achieve proposed system’s accuracy level up to 92.5%.

Faculty Coordinator : Dr. Basavaraj Amarapur

Student Name : Virupakshappa

Title of the Paper : Brain MRI segmentation using initial contour KPCM and optimal speed function for improved level set method

Research Description : Brain tumors are most aggressive kind of diseases, if left untreated may lead to very short life expectancy. Assessment of these tumors is usually done by Magnetic Resonance Imaging (MRI), but MRI produces large amount of data which rule out the manual segmentation in the stipulated time, also restricts the use of accurate quantitative evaluation in the clinical practice. So a reliable and automatic segmentation approach is required. Tumor segmentation in MRI brain image is a basic task in many computer vision problems. A typical methodology is to utilize Fuzzy iterative clustering algorithms that segregates the pixels into a given number of clusters. Nevertheless, the majority of these computations pose a few disadvantages which includes, they are time consuming, sensitive to noise and requires intialization. To overcome these problems, a novel segmentation method based on Particle Swarm Optimization (PSO) and rejection of outliers combined with level set method is developed. So as to enhance the segmentation result obtained from the previous research an Optimized Kernel Possibilistic C-Means(OKPCM) algorithm is proposed. Generally, in KPCM algorithm, an initial value of cluster centers is chosen randomly, but in our proposed method we change the existing KPCM algorithm by taking the cluster center initialization in to account. With the assistance of PSO method the cluster centers are picked ideally and the subsequent fuzzy clustering is utilized to specify an initial level set counter in the proposed improved level set based segmentation. The new improved level set based method has new speed function which efficiently removes boundary (contour) leakage problem is designed. The experiment is carried out on BRATS 2015 database and results are analyzed. The experimental results show that the proposed approach achieved segmentation accuracy of 96.83%.

Faculty Coordinator : Amarapur Basavaraj

Student Name : Virupakshappa

Title of the Paper : MRI Brain Tumor Segmentation: A Comparative Study and Analysis over Various Level Set Segmentation Algorithms

Research Description : Brain tumor is one of the deadliest cancers for the human community and requires image/signal processing approaches to record and analyze the disease-affected regions. Tumor segmentation in Magnetic Resonance images (MRI) of brain is a challenging task. With manual techniques for identifying the tumor through brain MRI is too time consuming. Therefore, the entire work is towards the development of computer aided diagnostic tool for brain tumor segmentation. The level set method has advantages over MRI Brain tumor segmentation; however, it often leads to either a complete breakdown or a premature termination of the curve evolution process and sometimes slower convergence, resulting in unsatisfactory results. By modifying the conventional Level Set method, we had proposed four methods earlier, so the main objective of this paper is to analyze those fourenhanced level set based segmentation methods. The proposed methodology is implemented in the working platform of MATLAB and the results were analyzed for the BraTS 2015 database images.

Faculty Coordinator : Dr. Basavaraj Amarapur

Student Name : Virupakshappa

Title of the Paper : Cognition-based MRI brain tumor segmentation technique using modified level set method

Research Description : Gliomas are the most common types of brain tumors seen in adults. Generally, it starts from glioma cells and afects the adja- cent tissues. Even though the analysis of glioma has well developed, the identifcation is still poor. In this paper, we propose an efcient modifed level set method for brain tumor segmentation, in which we preprocess the image to remove the noise and then accurately segment the magnetic resonance images (MRI). Therefore, this document anticipated an innovative level set algorithm for segmenting gliomas from the MRI brain images where the segmentation is made automatically by means of selecting the initial contour automatically from the maximum intensity pixel computed from the histogram intensity plots. The proposed methodology is implemented in the working platform of MATLAB to produce 99% accuracy, and the results are analyzed by the existing methods.

Faculty Coordinator : Basavaraj Amarapur

Student Name : Virupakshappa

Title of the Paper : Computer-aided diagnosis applied to MRI images of braintumor using cognition based modified level set and optimized ANN classifier

Research Description : MRI image segmentation and classification is one of the important tasks in medical image analysis and visualization, despite occurrence of noise makes it tough to segment the region of interest. In this paper, the MRI images are pre-processed and segmentation is carried out using modified Level set method for the tumor segmentation. Also, it is important to extract the useful features to predict the image class accurately. The proposed method operates Multi-Level wavelet decomposition features and for the wavelet coefficients modified chief descriptions like Grey Level Co-Occurrence Matrix (GLCM), Gabor and moment invariant features are extracted. The classification is carried out using the Adaptive Artificial Neuralm Network (AANN) methodology. In the adaptive ANN, the layer neurons are optimized using Whale Optimization Algorithm (WOA). The adaptive neural network optimizes the network structure to increase the classification accuracy and thus gives better classification results of tumors based on the segmented images. The proposed method will be executed in the working platform of MATLAB and the results are compared with the previous state of the art techniques. Finally, the proposed method results in classification accuracy of 98%.

Faculty Coordinator : Basavaraj Amarapur

Student Name : Virupakshappa

Title of the Paper : A NEW APPROACH OF BRAIN TUMOR SEGMENTATION USING FAST CONVERGENCE LEVEL SET

Research Description : Segmentation of region of interest has a critical task in image processing applications. The accuracy of Segmentation is based on processing methodology and limiting value used. In this paper, an enhanced approach of region segmentation using level set (LS) method is proposed, which is achieved by using cross over point in the valley point as a new dynamic stopping criterion in the level set segmentation. The proposed method has been tested with developed database of MR Images. From the test results, it is found that proposed method improves the convergence performance such as complexity in terms of number of iterations, delay and resource overhead as compared to conventional level set based segmentation approach.

Faculty Coordinator : Basavaraj Amarapur

Student Name : Virupakshappa

Title of the Paper : A Segmentation Approach Using Level Set Coding for Region Detection in MRI Images

Research Description : Computer-aided diagnosis (CAD) systems for identifying brain tumor region in medical study have been investigated by various methods. This paper introduces an approach in computer-aided diagnosis for identification of brain tumor in early stages using level set segmentation method. The skull stripping and histogram equalization techniques are used as the processing techniques for the acquired image. The preprocessed image is used to segment region of interest using level set approach. The segmented image is fine-tuned by applying morphological operators. The proposed method gives better Mean Opinion Score (MOS) as compared to conventional level set method.

Faculty Coordinator : Basavaraj Amarapur

Student Name : Virupakshappa

Title of the Paper : An Automated Approach for Brain Tumor Identification using ANN Classifier

Research Description : Detection of the tumor and separating it from the background MRI image is most important task in brain image analysis. It needs clinical experts to meet the standard level of accuracy. This limitation is overcome by the application of computer aided technology in medical field for tumor identification and segmentation. In this paper we proposed an efficient tumor segmentation model by using Fuzzy-C-Mean (FCM) clustering, multiple feature extraction using Gabor Wavelets and artificial neural network classifier. The proposed system performance is examined with 40 trained images with 60 tested MRI scanned medical dataset. The proposed system performance is examined in term of accuracy with respect to the confusion matrix. From the result section we proved that we meet required system accuracy level upto 85%.

Faculty Coordinator : Dr. Basavaraj Amarapur

Student Name : Virupakshappa

Title of the Paper : TAXONOMY OF BRAIN TUMOR CLASSIFICATION TECHNIQUES: A SYSTEMATIC REVIEW

Research Description : The use of digital image processing has become very demanding in various areas including medical applications. There are many applications where image processing is used to understand, analyze, interpret and make decisions. The main purpose of image processing is to improve the quality of the images for human/machine perception. The image processing techniques implemented for the detection of tumor from MRI images consist of image pre-processing, segmentation, feature extraction and classification steps. In this paper we have analyzed existing brain tumor detection and classification techniques. Brain image classification is very important because it provides anatomical structure information, which is necessary for planning of the treatment and patient follow-up. Thus various methods are surveyed in order to get better classification accuracy in terms of specificity, sensitivity and accuracy. This survey serves to classify the brain MRI images into normal, benign and malignant tumor. Classification of tumor is done with various techniques like Artificial Neural Networks (ANN), Deep Neural Networks (DNN), K- Nearest Neighbor (KNN), Support Vector Machine (SVM), Sequential Minimal Optimization (SMO) etc.

Faculty Coordinator : Dr. Basavaraj Amarapur

Student Name : Virupakshappa

Title of the Paper : Facial Image Segmentation by Integration of Level Set and Neural Network Optimization with Hybrid Filter Pre-processing Model

Research Description : Face segmentation is the process of segmenting the visible parts of the face excluding the neck, ears, hair, and beards. In this field, several methods have been developed, but none of them have been effective in providing optimal face segmentation. Hence, we proposed a novel face segmentation method known as level-set-based neural network (NN) algorithm. This method exploits a hybrid filter for the pre-processing of images, which eliminates the unwanted noises and blurring effect from the images. The hybrid filter is the combination of Median, Mean, and Gaussian filters and effectively removes the unwanted noises. Hence the images are segmented by utilizing level-set-based NN algorithm which is commonly based on the population set and effectively reduces the gap between the predicted and expected outcomes. The proposed method is compared with state-of-art methods such as Fully convolution network (FCN), Gabor filter(GF), multi-class semantic face segmentation(MSFS), and genetic algorithms (GA). From the experimental analysis, it is evident that the proposed work achieved better results comparing to other approaches.

Faculty Coordinator : Nagabhushan Patil

Student Name : Virupakshappa

Title of the Paper : Brain Tumor Classification using Fuzzy Level Set and Soft Computing

Research Description : As brain tumor is increasing day by day in youths in large in number. Identifying the brain tumor as early as possible and providing the proper treatment is necessary. The brain tumor classification includes certain steps which are extracting the tumor region from the MRI brain image and extract the features and then decision is taken some techniques. This paper briefs out how the tumor region is extracted from MRI image using segmentation methods, gives the idea about features of image and also about the soft computing. The method uses median filter as noise removal method and fuzzy level set as segmentation method for extracting the tumor region.

Faculty Coordinator : Basavaraj Amarapur

Student Name : Virupakshappa

Title of the Paper : AN IMPROVED SEGMENTATION APPROACH USING LEVEL SET METHOD WITH DYNAMIC THRESHOLDING FOR TUMOR DETECTION IN MRI IMAGES

Research Description : In this paper an improved approach for segmentation of brain tumor using MRI images has been presented. This approach consists of three steps, namely: preprocessing, segmentation and post processing. The preprocessing has been carried out using skull stripping and histogram equalization techniques. The region of interest is segmented by applying conventional Level Set method to the preprocessed image. The segmented image consists of distortions at the boundaries which lead to boundary leakage problem. This can be minimized by applying dynamic Thresholding technique and region localization method. The proposed method achieved better accuracy of segmentation as compared to the conventional level set segmentation method.

Faculty Coordinator : Basavaraj Amarapur

Student Name : Virupakshappa, Deepak S. Uplaonkar

Title of the Paper : Ultrasound liver tumor segmentation using adaptively regularized kernel-based fuzzy C means with enhanced level set algorithm

Research Description : Purpose – The purpose of this study is to develop a hybrid algorithm for segmenting tumor from ultrasound images of the liver. Design/methodology/approach – After collecting the ultrasound images, contrast-limited adaptive histogram equalization approach (CLAHE) is applied as preprocessing, in order to enhance the visual quality of the images that helps in better segmentation. Then, adaptively regularized kernel-based fuzzy C means (ARKFCM) is used to segment tumor from the enhanced image along with local ternary pattern combined with selective level set approaches. Findings – The proposed segmentation algorithm precisely segments the tumor portions from the enhanced images with lower computation cost. The proposed segmentation algorithm is compared with the existing algorithms and ground truth values in terms of Jaccard coefficient, dice coefficient, precision, Matthews correlation coefficient, f-score and accuracy. The experimental analysis shows that the proposed algorithm achieved 99.18% of accuracy and 92.17% of f-score value, which is better than the existing algorithms. Practical implications – From the experimental analysis, the proposed ARKFCM with enhanced level set algorithm obtained better performance in ultrasound liver tumor segmentation related to graph-based algorithm. However, the proposed algorithm showed 3.11% improvement in dice coefficient compared to graph-based algorithm. Originality/value – The image preprocessing is carried out using CLAHE algorithm. The preprocessed image is segmented by employing selective level set model and Local Ternary Pattern in ARKFCM algorithm. In this research, the proposed algorithm has advantages such as independence of clustering parameters, robustness in preserving the image details and optimal in finding the threshold value that effectively reduces the computational cost.

Faculty Coordinator : Nagabhushan Patil

Student Name : Virupakshappa

Title of the Paper : An enhanced segmentation technique and improved support vector machine classifier for facial image recognition

Research Description : Purpose – One of the challenging issues in computer vision and pattern recognition is face image recognition. Several studies based on face recognition were introduced in the past decades, but it has few classification issues in terms of poor performances. Hence, the authors proposed a novel model for face recognition. Design/methodology/approach – The proposed method consists of four major sections such as data acquisition, segmentation, feature extraction and recognition. Initially, the images are transferred into grayscale images, and they pose issues that are eliminated by resizing the input images. The contrast limited adaptive histogram equalization (CLAHE) utilizes the image preprocessing step, thereby eliminating unwanted noise and improving the image contrast level. Second, the active contour and level set-based segmentation (ALS) with neural network (NN) or ALS with NN algorithm is used for facial image segmentation. Next, the four major kinds of feature descriptors are dominant color structure descriptors, scale-invariant feature transform descriptors, improved center-symmetric local binary patterns (ICSLBP) and histograms of gradients (HOG) are based on clour and texture features. Finally, the support vector machine (SVM) with modified random forest (MRF) model for facial image recognition. Findings – Experimentally, the proposed method performance is evaluated using different kinds of evaluation criterions such as accuracy, similarity index, dice similarity coefficient, precision, recall and F-score results. However, the proposed method offers superior recognition performances than other state-of-art methods. Further face recognition was analyzed with the metrics such as accuracy, precision, recall and F-score and attained 99.2, 96, 98 and 96%, respectively. Originality/value – The good facial recognition method is proposed in this research work to overcome threat to privacy, violation of rights and provide better security of data.

Faculty Coordinator : Nagabhushan Patil

Student Name : Sushma J Patil, P.Ramesh, V Agalya, D Kodandapani

Title of the Paper : Fuzzy-PI Controlled Cascade H-Bridge Inverter Fed Single Phase Induction Motor Drive

Research Description : Now a day‘s multilevel inverters (MLI) are emerging in ac drive oriented applications. MLI topological structure that allows output voltage is to be integrated from isolated voltage sources. Various MLI topologies have been developed. The operation of the MLI is used to make a output ac voltage from variation in the dc source. This paper presents the detailed experimental study and effect of Proportional Integral controller (PI) and Fuzzy Logic Controller (FLC) for closed loop speed control of the 1-phase asymmetric cascade H Bridge (CHB) MLI fed induction motor drives. The FLC design and rule based implementation procedures are clarified. The FPGA based simulation and experimental results are tested and discussed.

Faculty Coordinator : M S. Aspalli

Student Name : Geeta

Title of the Paper : Proportional Resonant Controller For Semi Converter Three PhaseVSI Fed Induction Motor Drive to Enhance Time Responce.

Research Description : This effort deals with closed loop Semi Converter Three Phase Induction Motor Drive(SCTPIMD) using proportional integral(PI), Fractional order proportional integral derivative(FOPID) and Proportional resonant(PR) controller. This effort proposes PR controller for SCTPIMD. To create closed loop semi converter with three phase induction motor drive with enhanced dynamic response, PI,FOPID& PR controlled TPIMD systems are composed and simulated using MATLAB. The operation simulation results of SCTPIMD are examined. The simulation consequences of PI, FOPID& PR controlled SCTPIMD systems are analyzed interims of time domain parameters and comparison table has been exhibited. The outcomes show that the reaction with PR is better compared with the PI and FOPID controlled SCTPIMD systems.

Faculty Coordinator : M. S. Aspalli

Student Name : Geeta

Title of the Paper : Direct Torque Controlled Semi Converter -Three PhaseVSI Fed Induction Motor Drive with Enhanced Time Responce.

Research Description : This effort deals with closed loop Semi Converter Three Phase Induction Motor Drive(SCTPIMD) using Hysteresis controller(HC) and Direct Torque controller (DTC). This effort proposes DTC controller for SCTPIMD. To enhance dynamic response, closed loop semi converter with three phase induction motor drive with HC and DTCcontroller are composed and simulated using MATLAB. The operation &simulation results of SCTPIMD are examined. The simulation consequences of Hysteresis and DTC controlled SCTPIMD systems are analyzed interims of time domain parameters like settling time and steady state error and comparison table has been exhibited. The outcomes show that the reaction with DTC - SCTPIMD is better when compared with the hysteresis controlled SCTPIMD systems.

Faculty Coordinator : Dr. M.S.Aspalli

Student Name : Sushma J Patil, Nitin Kumar

Title of the Paper : A Switched Capacitor Multilevel Inverter with Voltage Boosting Ability

Research Description : This paper presents a Switched Capacitor multilevel inverter with voltage boosting ability to obtain fourfold boost with fewer components which can be used to boost and convert, the low output DC of the solar cells or renewable energy sources, into high AC output voltage thus the inverter can be used in applications like integration of solar PV/battery hybrid system for micro grid, hybrid electric vehicle, aerospace industries etc. The proposed topology with single dc source uses only eight switches to achieve a nine-level output, self-voltage balance of capacitors and fourfold boost, thus, the effective cost is reduced compared to other switched- capacitor multilevel inverters (SCMLIs). Like other SCMLIs, the proposed multilevel switched-capacitor inverters do not require a backend H-bridge, in which four witches must withstand the peak output voltage. Therefore, total standing voltage (TSV) can be lowered. The switched capacitor configuration is used to auto-balance the capacitor by supplying the switches with appropriate pulses using phase disposition pulse width modulation method.

Faculty Coordinator : Dr. M.S.Aspalli

Student Name : Ranjeeta Sugandhi, Sushma J Patil

Title of the Paper : A Five-Level Cascaded H-Bridge Multilevel Inverter with Reduced Number of Switches

Research Description : This paper presents simulation of power losses in a 5-level H-bridge multilevel inverter with a focus to reduce the number of power switching devices in the path for the flow of current. The proposed topology utilizes a double DC sources with five switches. In order to achieve higher efficiency operation of power electronics devices (power converters), conduction and switching losses have to be reduced. Multilevel inverters are designed to produce desired output voltages from different DC sources. A comparative discussion of simulated power loss values is addressed based on how the reduction of power switches contributed to the decrease of conduction and switching losses.

Faculty Coordinator : Dr. M.S.Aspalli

Student Name : Jitendra Bakliwal

Title of the Paper : Buck Converter Based LED Driving Topology for Solid State Luminaire

Research Description : This Paper deals with the practical implementation of 20W LED power driving circuit for domestic lighting. In this work, buck converter topology is implemented using controller IC MT 7834. It has integrated MOSFET which is used to provide constant output current with Single-stage active power factor correction, and benefits like higher efficiency, overvoltage protection, overcurrent protection & overtemperature protection. The circuit is operated in Quasi-Resonant mode to achieve higher efficiency. This paper also consists of test report which is carried out over the actual hardware of 20W LED power driving circuit. Finally experimental results shows that proposed LED power driving circuit is best suited for the household lighting applications.

Faculty Coordinator : Dr. M.S.Aspalli

Student Name : Rangayya

Title of the Paper : An enhanced segmentation technique and improved support vector machine classifier for facial image recognition.

Research Description : Purpose: One of the challenging issues in computer vision and pattern recognition is face image recognition. Several studies based on face recognition were introduced in the past decades, but it has few classification issues in terms of poor performances. Hence, the authors proposed a novel model for face recognition.

Faculty Coordinator : Dr. Nagabhushan Patil

Student Name : Rangayya

Title of the Paper : Facial Image Segmentation by Integration of Level Set and Neural Network Optimization with Hybrid Filter Pre-processing Model.

Research Description : Face segmentation is the process of segmenting the visible parts of the face excluding the neck, ears, hair, and beards. In this field, several methods have been developed, but none of them have been effective in providing optimal face segmentation. Hence, we proposed a novel face segmentation method known as level-set-based neural network (NN) algorithm. This method exploits a hybrid filter for the pre-processing of images, which eliminates the unwanted noises and blurring effect from the images. The hybrid filter is the combination of Median, Mean, and Gaussian filters and effectively removes the unwanted noises. Hence the images are segmented by utilizing level-set-based NN algorithm which is commonly based on the population set and effectively reduces the gap between the predicted and expected outcomes. The proposed method is compared with state-of-art methods such as Fully convolution network (FCN), Gabor filter(GF), multi-class semantic face segmentation(MSFS), and genetic algorithms (GA). From the experimental analysis, it is evident that the proposed work achieved better results comparing to other approaches.

Faculty Coordinator : Dr. Nagabhushan Patil

Student Name : Rangayya

Title of the Paper : Improved face recognition method using SVM-MRF with KTBS based KCM segmentation approach.

Research Description : Digital image processing is a technique for visually analyzing images that can be manipulated using different approaches. It can also be used to improve image quality and group images by identifying different special features in the images. When it comes to face recognition, evaluating face images with different postures is challenging. As a result, different researchers use edge detection approaches to analyze the images, however, the classification accuracy is low due to the high computational complexity. In the meantime, assessing images with a lot of noise or poor contrast is necessary. As a result, we have presented a novel face recognition approach that uses Contrast limited adaptive histogram equalization (CLAHE) to preprocess the images and a kernelized total Bregman divergence-based K-Means Clustering algorithm to improve segmentation even for images with high noise levels. Using dominant color structure descriptors, SIFT descriptors, improved center-symmetric local binary patterns (ICSLBP), and histograms of gradients(HOG), the features are derived. After that, a support vector machine (SVM) with a modified random forest (MRF) model is used to do facial recognition. The results demonstrated that the proposed SVM-MRF method achieved better classification accuracy with low computational complexity. The main reason is due to the novel segmentation technique used. The experiments are conducted on AR, CMU-PIE, and YALE datasets and obtained better segmentation and classification accuracy than other approaches, and the feature extraction time is also reduced to a great extent.

Faculty Coordinator : Dr. Nagabhushan Patil

Student Name : Deepak S. Uplaonkar

Title of the Paper : Ultrasound liver tumor segmentation using adaptively regularized kernel-based fuzzy C means with enhanced level set algorithm.

Research Description : Purpose The purpose of this study is to develop a hybrid algorithm for segmenting tumor from ultrasound images of the liver.

Faculty Coordinator : Dr. Nagabhushan Patil

Student Name : Deepak S. Uplaonkar

Title of the Paper : An Efficient Discrete Wavelet Transform Based Partial Hadamard Feature Extraction And Hybrid Neural Network Based Monarch Butterfly Optimization For Liver Tumor Classification.

Research Description : The liver tumor is one of the most widely occurring cancers nowadays. There are several forms of liver tumors, which are most often caused by hepatitis and cirrhosis. Furthermore, metastatic liver cancer may spread to other organs, posing a serious health risk. Hence it is ineluctable to diagnose this intimidating problem as early as possible. Liver tumour classification from ultrasound images is a challenging task since it is based on the structure and orientation of the liver tumour cells. To overcome this challenge, a novel hybrid artificial neural network-based monarch butterfly optimization algorithm is proposed for accurate liver tumour classification. Before the classification process, the liver tumor cells are preprocessed using different techniques such as adaptive filtering, median filtering, and color to greyscale transformation. Then the pre-processed images are segmented using the adaptively regularized kernel-based fuzzy C-means clustering algorithm and level enhanced segmentation which enhances the segmentation process and the same features are aligned in the same segment. Further, the feature vectors are extracted with the aid of the hybrid Discrete Wavelet Transform-based partial Hadamard transform method, and the same features are mapped as the same vector. The classification task is performed by a hybrid artificial neural network-based monarch butterfly optimization algorithm which enhances the classification accuracy. The comparative analyses with the state-of-art works show that the proposed work outperforms all the other approaches in terms of accuracy, specificity, sensitivity, precision, recall, and F1-score.

Faculty Coordinator : Dr. Nagabhushan Patil

Student Name : Deepak S. Uplaonkar1

Title of the Paper : Modified Otsu thresholding based level set and local directional ternary patter technique for liver tumor segmentation.

Research Description : In recent times, the liver tumors are one of the leading causes of death, hence automated segmentation of liver tumors helps physicians in early diagnosis and treatment options. In this paper, a novel segmentation technique is proposed for accurate segmentation of tumor regions from the liver Ultrasound images. Initially, liver Ultrasound images are collected from a real time dataset, which comprises of 105 liver metastases images. Then, label removal is accomplished by using binary thresholding and morphological operation to remove text from the liver Ultrasound images. Additionally, the quality of liver Ultrasound images is improved by applying contrast limited adaptive histogram equalization that improves original image contrast and preserves the image brightness. After image enhancement, Otsu thresholding based level set with enhanced edge indicator function and local directional ternary pattern technique is proposed for segmenting liver lesion/tumor region from the enhanced images. In the experimental phase, the proposed technique performance is validated in light of Matthews’s correlation coefficient, Jaccard coefficient, Dice coefficient, accuracy, precision and f-score. The simulation result showed that the proposed technique achieved 99.43% of segmentation accuracy, which is 5.43% higher than the existing graph based approach.

Faculty Coordinator : Dr. Nagabhushan Patil

Title : VI Solutions, Bangalore

Reference Files:

Title : Minds Solvit Pvt. Ltd.,

Description : Providing platform to students coding activities, conducting discussion on cutting edge technologies, students Summer training

Reference Files:

Title : Global Edge Software Ltd., Bangalore

Reference Files:

Title : Transneuron Technologies, Bangalore

Reference Files:

Title : Medini ISO 9001 – 2015 Certified Company, Bangalore

Description : Collaborative Research activities, Providing softwares for lab

Reference Files:

Title : The space warrior fellowship Kodihalli, Bangalore

Description : Organizing Students labs, Providing support for project execution,

Reference Files:

Title : Texas Instruments, Ed Gate Technologies Pvt Ltd.

Description : Providing support in development of curriculum, Aids in lab setups, Workshops, Training Programs

Reference Files:

Title : Canter CADD Technologies Pvt Ltd, Kalaburgi

Description : Workshops, Training Programs, Project Management.

Reference Files: