Search

This paper presents the hybrid method of medical image fusion, and a hybrid method is a combination of PCNN and particle swarm optimization. The pulse coupled neural network is a dynamic neural network model, and the rate of recognition is very high. The particle swarm optimization is a swarm-based algorithm. The nature of the algorithm is iterative and reduces the unwanted noise factor during the process of fusion. For the extract of feature, coefficient applied contourlet transform, contourlet transform is rich feature extraction process in image processing. The method of contourlet transform is subband decomposition and directional feature coefficient with Laplacian pyramidal transform. The proposed algorithm increases the value of mutual information of the fused image. The experimental analysis process uses two types of images: computed tomography (CT) and magnetic resonance imaging (MRI). The proposed algorithm simulated in MATLAB software and tested with reputed medical image dataset. The process of evaluation measure standard parameters and compare with CTGA, and PCNN algorithm. Our experimental results show better performance instead of CTGA and PCNN.

copy>978-1-7281-7741-0/20/$31.00 ©2021 IEEEDuring this Pandemic time a day without Internet is unimaginable. At present the most widely used non-cellular wireless technology is Wi-Fi (Wireless Fidelity). It is based on radio wireless local area networking of devices based on IEEE 802.11 standards operating at the frequency range of 2.4GHz (12cm) UHF (Ultra High Frequency) and 5.8GHz (5cm) SHF (Super High Frequency) is probably the most popular used non-cellular wireless technology. Now imagine a world where each LED Light in and around Hospitals can connect us to high speed wireless internet with a simple flickering of light which is Li-Fi or Light Fidelity which is a new way of sending data wireless. As Doctors are the front line soldiers against Covid-19, This paper which explains about the prototype for transmitting the data by using different body sensors such as Heart Beat Sensors, Temperature Sensors or Pressure sensors from Li-Fi Transmitter to Li-Fi Receiver by using simple LED light as a platform to provide higher speed, secure, safer and greener light based radiation-free communication network compared to the existing modes of Wi-Fi Communication or Telecom tower- based Communication or LAN network for Covid-19 patients and the Medical Professionals.

With the advancement in technology and digitalization of society, the number of smartphone users has increased drastically. According to [1], there will be 5.3 billion Internet users by 2023. Due to this, Wi-Fi penetration rate will also increase. For tracking smartphones present inside the vehicles, the uniqueness of Wi-Fi Medium Access Control (MAC) addresses helps a lot. Road traffic monitoring systems can make use of this information for estimating travel time, average speed, and traffic flow in the areas of interest. This article focuses on utilizing Wi-Fi based road-side sniffers to listen to Wi-Fi probe signals broadcasted by the smartphones present inside vehicles for estimating traffic stats of road segments. Hopping Interval (HI) is the rate at which Wi-Fi radio of the sniffer is changed to one of the available channels in ISM band. A channel hopping algorithm is proposed, which adjusts HI of the sniffer according to the current traffic conditions. The proposed algorithm reduces the estimation error and increases the robustness and efficiency of the traffic monitoring system. To validate the algorithm, simulation experiments have been performed in Indian traffic scenarios, and results show that the proposed algorithm is better than static channel assignment on the sniffer with an improvement of 4%. Also, estimated travel time and the ground truth values are found out to be closely related, which assures the suitability of the Wi-Fi based road-side sniffers for traffic monitoring.

Today’s world is exploding with new music and attributing to its diversity, it has become the most popular past time ever. Even though there are over ninety seven million songs on the internet, a descriptive study of emotions and trends in music development is still missing. Trends in music development represent how the music industry has grown throughout the years. In this paper, we use the million songs dataset, a publicly available dataset that contains audio features and metadata about a million songs, to dig deeper into the musical trends our world has experienced in the past few decades. By finding out the most popular genres and how their popularity has changed throughout time, we seek to provide concrete evidence of influence of technological advancements on music sector. Sentiments and emotions in songs are analyzed to see which genre promotes more positivity or negativity in its music, providing data to aid current music recommendation systems. Furthermore, we also analyzed the business values of various genres of music to shed some light on which genres are most profitable.

Electrocardiogram(ECG) is a valuable clinical signal, which is widely used to identify the cardiovascular diseases. However, it remains a cumbersome process to manually evaluate the ECG signals because of smaller variations in its physiological features in normal and abnormal cases that too when there are a huge number of cardiac patients to examine. In such a scenario, automatic classification of ECG signals can provide an ease to the doctors to make a correct diagnosis of a particular disease. This work proposes a classification model to classify the ECG in five different classes based on their morphological features. Instead of using manually designed features as most of the existing ECG classification works do, we have extracted data-driven non-linear features using convolutional neural network. The 1D-CNN model architecture is based on three convolutional, max pooling and dense layers which automatically extracts distinguishable nonlinear features from the ECG signals and automatically classify them into five different classes: Non-ectopic beats (Normal Beat), Supraventricular ectopic beats, Ventricular ectopic beats, Fusion Beats and Unknown Beats. The proposed algorithm was assessed using open-source database of MIT-BIH, which is based on 47 subjects. After 5-fold cross-validation, the presented algorithm achieves an accuracy of 97.36% and f1 score of 99.83%. It is a simple yet fast performing model that is implementable on e-healthcare-based devices for remote heart diagnosis of patients.

This paper presents an ultra-low-power solar energy harvesting system with a power management module (SEHS-PMU) for the Internet-of-Things (IoT) applications. The voltages generated are suitable for the IoT sensors used for smart cities as well as Internet-of-Medical-Things (IoMT) based biomedical applications. The SEHS-PMU is harvesting energy from the solar cell, and it can provide regulated voltages of 3.3 V, 1.8 V, 1 V, and 0.5 V. The harvesting mechanism leads to battery-less IoT, and is safe for the humanity and environment. Charge pumps (CP) are used as a boost and buck converter, which is suitable for monolithic integration. Low drop-out regulators (LDOs) are used for regulating the loads. A digital controller controls the selection of loads as per requirement for saving the power. The SEHS-PMU is designed in CMOS 180 nm technology library.

This paper presents a multiband microstrip antenna that can be used in L, S, C and X band application of microwave frequencies. The multiband effect brought by the use of closed C shape conductive strip around the patch. Microwave substrate of size 30*30mm having dielectric constant 4.4 is used for an antenna design. At resonating frequency 5.3GHz the No. of bands obtain is 6 with a gain of 23.8 dB.

The objective of the paper is to design an antenna with multiple resonant frequencies. The proposed antenna resonates at 12 different frequencies with a centre frequency at 5.3GHz. Microstrip patch antenna with conducting stips is used to achieved this purpose. Microwave substrate of size 30mm×30mm having a dielectric constant of 4.4 is used to design the antenna. The desired frequencies are obtained by cutting slots in the patch and impedance matching is done by providing defective ground. The multiband antenna is designed using high frequency structural simulator (HFSS) software with return loss of about −14.34dB and excellent gain of 20.51dB at 5.3GHz.

In this paper, a time-optimal feedback solution to the game of two cars is obtained for the case where the pursuer is faster and more agile than the evader. It is shown that the saddle point strategies for both the pursuer and the evader are coincident with one of the common tangents from the minimum radius turning circles of the pursuer to the minimum radius turning circles of the evader when the evader is distant enough from the pursuer. Four valid tangents are identified using geometry and a $2\times 2$ matrix game is formulated corresponding to these tangents. The solution of this game, at each instant, provides feedback strategies for the pursuer and the evader. The computations required to implement this law are insignificant and hence the proposed feedback law is suitable for real time implementation.

In various fields of technology, computation techniques are pushing further the limits of human outreach. One such field is the character recognition commonly known as Optical Character Recognition (OCR). This paper is about taking offline handwritten Assamese alphabets as input using modern input devices like camera or scanner and tried to convert it to digital form to recognize the alphabets. Template Matching technique is anticipated for recognizing handwritten characters, including the schemes of Grayscale, Binary, Dilation, Erosion conversion or transformation of image, detection of ROI and comparing with dataset. The efficiency or the accuracy is calculated for every character which is shown in table 1 with a number of testing cases. With this proposed template matching technique we have been able to get accuracy more than 80% for maximum Assamese handwritten alphabets.

A Spanish fan is a hand held traditional fan which is used as an accessory and also by Flamenco dancers. The manipulation of the fan is quite difficult as it involves dynamic motion which includes opening, flapping and closing the fan along a pivotal point. The key points include the motion to be quick and the fan to be opened to the maximum degree possible without human intervention. A robotic arm with 7 Degrees of Freedom (DOF) is used to manipulate the autonomous motion. The fan placed on the table is localized and detected using a camera by background subtraction, masking and filtering; post which the contour of the fan is detected. The pixels obtained is then transformed into real life coordinates. The Dynamixel motors then traverses to the coordinates of the fan's position to grasp, open, flap, close and put the fan down.

Continuous training is crucial for creating and maintaining the right skill-profile for the industrial organization's workforce. There is a tremendous variety in the available trainings within an organization: technical, project management, quality, leadership, domain-specific, soft-skills etc. Hence it is important to assist the employee in choosing the best trainings, which perfectly suits her background, project needs and career goals. In this paper, we focus on algorithms for training recommendation in an industrial setting. We formalize the problem of next training recommendation, taking into account the employee's training and work history. We present several new unsupervised sequence mining algorithms to mine the past trainings data from the organization for arriving at personalized next training recommendation. Using the real-life data about trainings of 118587 employees over 5019 distinct trainings from a large multi-national IT organization, we show that these algorithms outperform several standard recommendation engine algorithms as well as those based on standard sequence mining algorithms.

In this paper, we address the issue of type reduction of multi-dimensional interval type-2 (IT2) fuzzy sets (FSs). We utilize the Karnik-Mendel (KM) algorithm to estimate the centroid boundary of a multi-dimensional footprint of uncertainty (FOU). We deal with two-dimensional (2-D) fuzzy sets as we can visualize the FOU using 3-D plots, thus making the illustration of the methods simple. However, the basic idea can be extended to multiple dimensions. We give a formal definition of the centroid boundary of a 2-D IT2 fuzzy membership function (FMF) and propose two methods for its estimation. The first method computes embedded type-1 (T1) FSs whose centroids constitute the centroid boundary. We obtain the embedded sets by producing slices of the domain using different sets of parallel planes and then apply the KM algorithm over each slice, to obtain “embedded-curves.” For the second method, we approximate our first method by restricting embedded-curves to be “embedded-lines” thus enhancing computational speed. These type reduction techniques can be applied to applications involving multi-dimensional centroid estimation such as, clustering, support vector estimation for dimensionality reduction, fuzzy logic controllers, to mention a few.

In this paper, we propose a novel method for visualizing two-dimensional interval type-2 fuzzy membership functions (2-D IT2 FMFs) using one-dimensional general type-2 fuzzy membership functions (1-D GT2 FMFs), and also describe the procedure for extending our method to fuzzy sets representing higher dimensional data. Then we present a type reduction method for mapping 2-D IT2 fuzzy sets into 2-D type-1 fuzzy sets that uses alpha-plane representation of general fuzzy sets. We discuss the problem of “multiple membership values for the same element,” which violates set properties, in an IT2 Fuzzy C-means (FCM) algorithm for clustering and propose a solution that uses transformations in the visualization method. These techniques can be applied to applications involving fuzzy sets that represent multidimensional data for proper visualization and type reduction, such as image segmentation, classification and prediction, to name a few.

The MANET that is Mobile Ad hoc Network are forming a group of many nodes. They can interact with each other in limited area. All the Malicious nodes present in the MANET always disturb the usual performance of routing and that cause the degradation of dynamic performance of the network. Nodes which are malicious continuously try to stump the neighbor nodes during the process of routing as all neighbor nodes in the network merely forward the reply and response of neighboring. The intermediate nodes work is very responsible in routing procedure with continuous movement. During the work we have recommended one security scheme against the attack of packet dropping by malicious node in the network. The scheme which is recommended here will work to find attacker by using the concept of detection of link to forward the data or information between sender and receiver. The packet dropping on link, through node is detected and prevented by IDS security system. The scheme not only works to identify the nodes performing malicious activity however prevent them also. The identification of attacker is noticed by dropping of data packets in excsssessive quantity. The prevention of it can be done via choosing the alternate route somewhere the attacker performing malicious activity not available among the senders to receivers. The neighbor nodes or intermediary identify the malicious activity performer by the way of reply of malicious nodes which is confirmed. The recommended IDS system secures the network and also increases the performance after blocking malicious nodes that perform malicious activity in the network. The network performance measures in the presence of attack and secure IDS with the help of performance metrics like PDR, throughput etc. Planned secure routing improves data receiving and minimizes dropping data in network.

Active ECUs (Electronic Control Unit) in a hybrid car while it is charging leads to inefficient power usage and reduction in ECU lifetime. To solve this issue, AUTOSAR has suggested PN (partial networking). As per the AUTOSAR specification, ECUs which are not required to be kept active during battery charging must be shutdown. This project's aim is to analyze the robustness of the design suggested by AUTOSAR in certain situations explained by testcases mentioned in the specification with the intention of suggesting a solution to AUTOSAR in case any issue is found during this regression testing.