MIT-World Peace University (MIT-WPU) is a top educational institute located in Pune, India. It is recognized within&nbsp;UGC under Government of Maharashtra Act No. XXXV 2017.&nbsp;

It is part of the MIT group of institutions (established in 1983) which covers 10+ campuses across India and has more than 50,000+ students enrolled across various disciplines at a given point of time. The institute houses 15 schools ranging from Engineering &amp; Technology, Design, Pharmacy, Commerce, Governance among others. It offers undergraduate, postgraduate and PhD programs among different disciplines.&nbsp;

&nbsp;

MIT World Peace University

School of Electronics & Communication Engineering

Traffic congestion has become a major issue nowadays. Not only in metro cities even in small cities this traffic congestion is a big problem. Therefore, we have required an intelligent traffic control system that can decrease traffic. Our current traffic control system is not flexible and adaptable, and it is time-based and is unaffected by heavy traffic. Because of its static nature, it is unreliable, unpredictable, and noisy. The proposed system is the approach for traffic control. In the proposed system the image processing and IR sensors are used. IR sensors are installed at the roadside from where they detect vehicles and send results to Arduino UNO. On the other side, image processing is used to count vehicles by using camera surveillance and sends the result to the microcontroller through a serial port. Arduino UNO then analysis both the results and update green path time according to the traffic density on the road. Using technologies separately may have its disadvantages which can be overcome by considering a combination of both results for a better result. This method increases the accuracy and productivity of the system. Apart from this, the project gives prioritization to the emergency vehicles like fire trucks, ambulances. Also, this project saves basic information about vehicles, i.e. speed, vehicle colour, type, etc. This information can be used for further analysis, and as a result, it can forecast traffic congestion and road condition at various times. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

World University root node

Engineering and Technology

Lecture Notes in Electrical Engineering

Traffic Management System Using AI and IoT

According to earlier studies, chest x-ray (CXR) pictures might be used as a preventative measure for artificial intelligence (AI) to identify corona pneumonia. Unfortunately, problems with the collections and research approaches from a scientific and medical standpoint have also surfaced, and we also have concerns about the resilience and susceptibility of AI systems. In this work, we solve these concerns by creating our original information through some kind of retroactive preclinical studies to supplement the data gathered from external factors, leading to a more accurate generation of AI-driven coronavirus detection techniques. In order to objectively analyse research strategy, we created design techniques by modifying datasets, optimised five learning algorithms, and added a variety of sensing circumstances to assess the resilience and analytical effectiveness of the methods. In a 3-class identification situation as opposed to a 4-class research design, the study demonstrates superior overall efficiency of 91-96 percent Sn (sensibility), 94-98 percent Sp (precision), and 90-96 percent PPV. With a reliability score of F1-measure, as well as a g-average of 96% in the 3 class instances of identification, InceptionV3has the best overall result. At the same time, InceptionV3exhibited the best results for COVID-19 pneumonia identification with 86 percent Sn, 99 percent Sp, and 91 percent PPV with an AUC of 0.99 in separating pneumonia with regular CXR. It achieved 0.98 AUC and a stutter stepping mean of 0.99 for those other categories for its capability to distinguish between COVID-19 pneumonia and non-COVID-19 pneumonia. © 2022 IEEE.

2022 IEEE 6th Conference on Information and Communication Technology, CICT 2022

Artificial Intelligence for the Detection of Covid-19 Pneumonia on Chest X-Ray

Navigation is how everything moves around globally, and electronics have provided better solutions for appliances to work seamlessly. There are two broad navigation options for finding your location. One is a Global Positioning System (GPS) system, and the other one is by using the Inertial Measurement Unit (IMU). The IMU is nothing but a combination of accelerometers and gyroscopes. This paper puts light on the Inertial Navigation System (INS), which uses GPS and IMU to get navigation data. GPS and INS give us real-time velocity and position data required for navigation. However, GPS has a slow update rate, up to 1-10Hz, while IMU performs far better at gaining navigation data with an update rate up to 1KHz. IMU accumulates errors and drifts over time while GPS has a low update rate. These drawbacks make both systems unreliable when used alone. The proposed work talks more about the use of both sensors, and Raspberry Pi Zero using a sensor fusion technology. Removing drawbacks of both the sensors and the use of the Kalman Filter are discussed in the paper. © 2022 IEEE.

2022 6th International Conference on Computing, Communication, Control and Automation, ICCUBEA 2022

SENSOR FUSION: An Advance Inertial Navigation System using GPS and IMU

Health monitoring and technologies related to health monitoring is an appealing area of research. The electrocardiogram (ECG) has constantly been the mainstream estimation plan to evaluate and analyse cardiovascular diseases. Heart health is important for everyone. Heart needs to be monitored regularly and early warning can prevent the permanent heart damage. Also, heart diseases are the leading cause of deaths worldwide. Hence, this work presents a design of a mini wearable ECG system and is interfacing with the Android application. This framework is created to show and analyse the ECG signal obtained from the ECG wearable system. The ECG signals will be shipped off an Android application via Bluetooth device and Wi-Fi. This system will automatically alert the user through notification. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Lecture Notes on Data Engineering and Communications Technologies

ECG Acquisition Analysis on Smartphone Through Bluetooth and Wireless Communication

Visual Pattern Recognition Networks (VPRNs) delivers remarkable performance using Deep Neural Networks (DNN). Various DNN architectures are available such as Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN) and its variants. Continuous emergence of novel architectures, availability of big data and enormous computing powers keeps this area flourishing. However, such DNN based VPRN models are plunged with enormous computational complexities, intense memory requirements, and high power consumptions which restricts its use in edgeAI environments. For instance, a typical pre-trained VGG-16 model needs 500MB memory, has over 138 million parameters and involves 15.5 billion Floating Point Operations (FLOPs). Such overheads demand for compression of VPRN models while retaining its performance. In this regard, various compression methods have been reported in literature. This paper presents experimentation on hybrid compression method which combines pruning and quantization to achieve higher compression ratio for nonsequential as well as sequential DNN based VPRNs. © 2022 IEEE.

2022 International Conference on Signal and Information Processing, IConSIP 2022

Model Compression for Non-Sequential and Sequential Visual Pattern Recognition Networks ― A Hybrid Approach

Lung cancer, also known as malignant tumours, is one of the deadly diseases that are prevalent worldwide. It is brought on by the lung tissues' hesitant cell growth. Only when the disease is in its early stages can treatment be begun to cure it. The results of blood tests and computed tomography (CT) scanning are used to find this. Emphasis on the tumours area allows for the detection of the aberrant picture. The sample, which is in jpg format, consists of CT scan images. Convolutional neural networks are used to train the developed framework (CNN). The benefit of using CNNs formed with fuzzy correlation maps is that they do not require considerable image thresholding or the scarce supply of specialized training data. The pre - trained models AlexNet CNN algorithm is presented the output of the fuzzy logic. A deep CNN requires a significant amount of data to analyse in order to get an accurate output. To handle the problem of the restricted size of the CNN set of data, a fuzzy logic-based fuzzy association map is proposed.This work analyses fuzzy-based CNN image classification models and proposes a hybrid CNN fuzzy logic.The use of present technology would increase as a result of this early cancer detection strategy, and cancer prevention would advance. © 2022 IEEE.

International Interdisciplinary Humanitarian Conference for Sustainability, IIHC 2022 - Proceedings

CNN and Fuzzy logic based hybrid approach for lung cancer detection and report generation

This work provides a novel segmentation and classification of diseases from the GI tract using WCE images. First, the input WCE images are collected from the KID Atlas data set, and they are preprocessed using filtering and contrast enhancement techniques. Then, the modified U-Net with parameter tuning by the Deer Hunting with Distance-based Solution Update (DH-DSU) is adopted for performing the lesion or abnormality segmentation. From the segmented images, the feature extraction is employed by a set of approaches such as ‘Gray Level Co-occurrence Matrix (GLCM), Scale-Invariant Feature Transform (SIFT), Histogram of Oriented Gradients (HOG), Local Binary Pattern (LBP) and Canny edge detection’. These features are concatenated and given to the improved Deep Neural Network (DNN) with DH-DSU-based hidden neuron optimisation. This provides the classification outcome in terms of three diseases of the GI tract such as inflammatory, polypoid and ulcer. Through the performance analysis, MCC of DH-DSU-IDNN is 8.49%, 1.73%, 18.46% and 17.23% more than DHOA-IDNN, WOA-IDNN, GWO-IDNN and PSO-IDNN, respectively. At the end, the offered approach has been validated on collected data sets, and the achievement of findings is adequate when compared with the baseline approaches. © 2022 Informa UK Limited, trading as Taylor & Francis Group.

Computer Methods in Biomechanics and Biomedical Engineering: Imaging and Visualization

Gastrointestinal tract disease segmentation and classification in wireless capsule endoscopy using intelligent deep learning model

Remotely operated vehicles (ROVs) are underwaterrobots that are commanded from the surface by a person. This robot is made up of embedded systems and mechanical systems, and it is connected to the outside world via a series ofload-bearing umbilical cables that hold power, data, and communication cables. For the ROV, data transfer as well ascontrol signals are required. The navigation team was made up of working on the ROV's navigation system. It has a video camera and a ballast system for balance, movement propulsion system pumps, and IMU to locate the position of a vehicle microcontrollers to compute, process, and offer the navigation system, sensor for feedback data back to the lighting and components controlling and coordinating all of this is a challenge. A manipulator, subsea equipment, and devices to assess clarity, temperature, and depth are all included in the proposed task. The purpose of the proposed effort is to create such a ROV to execute a given mission involving several persons. In addition, a novel form of ROV attitude control is introduced, which makes use of floats to modify the ROV's centre of buoyancy. This is an interdisciplinary endeavour in which the members are from many fields. Together with the mechanical department, the electronics navigation team worked on this. © 2022 IEEE.

Proceedings - 2022 International Conference on Recent Trends in Microelectronics, Automation, Computing and Communications Systems, ICMACC 2022

Navigation of Underwater Remotely Operated Vehicle

Additive noise degrades speech quality and intelligibility. Speech enhancement reduces this noise to make speech more pleasant and intelligible. It plays a significant role in speech recognition or speech-operated systems. In this paper, we propose a single-channel speech enhancement method in which the log-minimum mean square error method (log-MMSE) and modified accelerated particle swarm optimization algorithm are used to design a filter for improving the quality and intelligibility of noisy speech. Accelerated particle swarm optimization (APSO) algorithm is modified in which a single dimension of particle position is changed in a single iteration while obtaining the particle's new position. Using this algorithm, a filter is designed with multiple passbands and notches for speech enhancement. The modified algorithm converges faster compared with standard particle swarm optimization algorithm (PSO) and APSO giving optimum filter coefficients. The designed filter is used to enhance the speech. The proposed speech enhancement method improves the perceptual estimation of speech quality (PESQ) by 17.05% for 5dB babble noise, 33.92 % for 5dB car noise, 14.96 % for 5dB airport noise, and 39.13 % for 5dB exhibition noise. The average output PESQ for these four types of noise is improved compared to conventional methods of speech enhancement. There is an average of 7.58 dB improvement in segmental SNR for these noise types. The proposed method improves speech intelligibility with minimum speech distortion. © 2022 Penerbit UTM Press. All rights reserved.

ASEAN Engineering Journal

SINGLE CHANNEL SPEECH ENHANCEMENT USING EVOLUTIONARY ALGORITHM WITH LOG-MMSE

This paper tactics to develop the novel Tanner-Whitehouse 3 (TW3)-based automated Bone Age Assessment (BAA) model for children with the assistance of Faster Region- Convolutional Neural Network (Faster R-CNN) and Optimal trained Recurrent Neural Network (RNN). The proposed model covers three main stages: (a) Segmentation, (b) Faster R-CNN-based feature learning, and (c) optimal classification. Further, the segmentation of those regions is performed by adaptive Otsu thresholding. A Faster R-CNN is built to learn the features from the segmented regions. Once the features are extracted from the pooling layers of Faster R-CNN, it is subjected to the RNN. As a modification to RNN, the training weight is optimized by the Average Fitness-based Sun Flower Optimization (AF-SFO), and the optimized network predicts the age of the bone. The experimental evaluation of the proposed model over a set of images collected manually and publically shows its superior performance when compared to the state-of-the-art techniques. © 2021 Elsevier Ltd

Biomedical Signal Processing and Control

Faster Region-Convolutional Neural network oriented feature learning with optimal trained Recurrent Neural Network for bone age assessment for pediatrics