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

World University root node

School of Computer Engineering & Technology

Engineering and Technology

Lecture Notes in Networks and Systems

Machine Learning Prediction if the Patient is at Risk of Undergoing Surgery Based on Preoperative Medical Reports

Ocular disease is an eye disease that reduces the eye’s ability to work normally. Early ocular disease detection is important to avoid blindness caused by some of the diseases like cataracts, glaucoma, diabetes, age-related macular degeneration (AMD), etc. Artificial intelligence (AI) techniques have been used to build systems for the speedy diagnosis of such diseases. In recent years, the deep neural network (DNN) has shown remarkable success in this area. But the black box nature of such systems has created questions on the use of DNN in a high-risk system like health care. Explainable AI (XAI) is a suite of methods and techniques that provides explanations of predictions made by AI systems. This helps to achieve accountability, transparency, and debugging of the model in the healthcare domain. In this paper, we have proposed to develop an ocular disease classification model and an XAI method that can be used to explain the classification of eye diseases, from eye fundus images.

Explainability of Deep Learning-Based System in Health Care

Annotation of images is a time-consuming task. In recent times captioning of images is required in many Artificial Intelligence applications. Captioning Images intends to produce a sound and thorough portrayal that sums up the contents of a picture. Traditional approaches cannot handle the complexity and difficulties of image captioning as well as deep learning-based approaches. In this paper an encoder-decoder version with an attention mechanism integrating the picture feature that generates a semantic and syntactically sound description. One of the available benchmark datasets, Flickr8k is used to evaluate the model. The system incorporates an EfficientNet B3 encoder that draws out features from an image. It uses the mapped caption of images to produce the corpus. For the caption generation process, the model implements the attention and Gated Recurrent Unit (GRU) model to effectively bring about the image description. The various cosine similarity and N-Gram feature extraction means have been used for generating the blue score for the entire testing dataset and show the effectiveness of the proposed system.

Image Captioning using Deep Learning based System

Image Captioning using Deep Learning based System

Deep learning in computer vision has shown remarkable success in the performance of detection systems for plant diseases. However, due to the complexity and deeply nested structure of these models, these are still considered as black-box and explanations are not intuitive for human users. Many researchers have developed deep neural architectures for plant disease detection but have not provided classification explanations. To be used in practical applications, our model needs to explain why the model classified a given image. Explainable Artificial Intelligence (XAI) provides algorithms that can generate human-understandable explanations of AI decisions. In this paper, we summarize recent developments in XAI techniques, develop a plant disease detection system, and most importantly an explainable AI method named Gradient-weighted Class Activation Mapping ++ (GradCAM++) is used to locate the disease and highlight the most important regions on the leaves contributing towards the classification.

Explainable AI for Deep Learning Based Disease Detection

Image Captioning is one of the emerging topics of research in the field of AI. It uses a combination of Computer Vision (CV) and Natural Language Processing (NLP) to derive features from the image, use this information to identify objects, actions, their relationships, and generate a description for the image. It is most important concept in artificial intelligence applied in the fields like aid to the blind, self-driving cars, and many more. This paper we demonstrates a concise state of art image captioning and its method for caption generation using deep learning concepts. We also determine the approach for image caption generation using Convolutional Neural Network (CNN) and Generative Adversarial Network (GAN) model in deep learning framework. Using this approach system intelligent enough to create sentences for images. It uses the encoder-decoder architecture, where CNN is used for image vector generation and LSTM is used for the generation of a logical sentence using the NLP concepts. Finally, we evaluate the proposed system experimental analysis with numerous existing systems and show the effeteness of system. © 2021 IEEE.

2021 International Conference on Emerging Smart Computing and Informatics, ESCI 2021

Image captioning methods and metrics

In urban areas the water supply to residence and commercial establishments are provided at a fixed flow rate. There are incidents of excess water drawn by certain customers/users by connecting motor-pump sets to the waterlines which is considered as water theft. In this project it is proposed to develop an embedded based remote water monitoring and theft prevention system by recording the flow rates at the consumer/user end. In order to implement the proposed water supply system, each consumer should be provided with an embedded based water flow monitoring system consisting of a microcontroller to record the flow rate using a flow sensor and to transmit the same to a remote monitoring station using wireless transmitter and it is also provided with an electrically operated solenoid valve to supply water to the consumers. The valve turns on/off to stop the water supply whenever the flow rate exceed a predefined limit. The solenoid valves are also controlled using real time clock to control flow of water accordingly for a fixed duration of time. It is proposed to employ a gsm modem for wireless communication so that the information can be passed to particular responsible officer‟s cell phone for immediate action.One of the world's major problems is the scarcity of water, mainly due to the unmeasured consumption and the waste of this vital element. This project describes a modular water system, which helps not only in monitoring consumption but also in finding water pollution. In the past water resources were very limited & nobody can afford its wastage. For managing and providing water so many workers were required. Also to switch on valve & distribute water bill by going to home so many workers were required. To overcome this problem we have decided to work on automatic water management using microcontroller (atmega 328p). Also this project deals with problem of distributing water bill in different area by going to every user’s home. We have decided to use GSM module for communication. In this project we will make a PYTHON based management software which will handle user account details.

Journal of Emerging Technologies and Innovative Research

Automatic Water Distribution System

Data Hiding or Steganography has been an important communicated network since people started communicating in writing. The main aim in steganography is to hide a secret message within cover media in such a way that an observer cannot detect the presence of contents of the hidden message. Cover images are original images without secret data and after embedding secret data they are called as stego images. Steganography and cryptography are counter parts in digital security the obvious advantage of steganography over Cryptography is that messages do not attract attention to themselves, to messengers, or to recipients. The progress in steganography has also led to many serious problems such as hacking, compression, reformate, etc. Steganography finds its role in attempt to address these growing concerns. We know that, with the use of steganographic techniques, it is possible to hide secret information within images, Audio and video files which is statistically undetectable. This paper proposes a new data embedding technique which uses two pixel sequentially one after another which is kwon as pixel pair matching (PPM) technique. The basic idea of PPM is to use the values of pixel pair as a reference coordinate, and search a coordinate in the neighbourhood set of this pixel pair according to a given message digit. The two pixel is then replaced by the searched coordinate to store the digit. The APPM method offers lower distortion than DE by providing more compact neighbourhood sets and allowing embedded digits in any notational system.

International Journal of Science and Research

Data Hiding Method Using Adaptive Pixel Pair Matching

For rapid development of network technology data is needed to be communicated over the network. Although network transmission is convenient and fast, the data is often attacked. Steganography is used to hide the data in the images. Images used for carrying data are called cover images and images with embedded data are called stego images. Data is embedded in the images in such a way that it should be detected by human eyes and it should be robust so that it offers resistance to various image processing methods and compression. In this survey paper our focus is on development and current status on data hiding techniques in spatial domain of images. Keywords-Steganography, stego image,least significant bit(LSB), optimal pixel adjustment process (OPAP), diamond encoding(DE).

International Journal of Modern Trends in Engineering and Research

Data Hiding Techniques: Survey and current status

Fault diagnosis is a central aspect of network fault management. Since faults are unavoidable in communication systems, their quick detection and isolation is essential for the robustness, reliability, and accessibility of a system. Fault localization is done using probing, monitoring, with agents, agent less, through WMI etc. Probing technique for fault localization involves placement of probe stations (Probe stations are specially instrumented nodes from where probes can be sent to monitor the network) which affects the diagnosis capability of the probes sent by the probe stations. Probe station locations affect probing efficiency, monitoring capability, and deployment cost. At the same time, probe station selection affects the selection of probe set for fault localization. We are presenting a novel approach for minimal probe station selection with an objective to reduce probe set size that will be used for fault localization. The probe station selected through this approach results into reduced deployment cost as well as it helps in producing smaller probe set for fault localization reducing overall operation cost. We provide experimental evaluation of the proposed algorithm and provided analysis of same with Shadow Node Reduction (SNR) algorithm through simulation results. © 2013 IEEE.

IEEE International Conference on Networks, ICON

Probe station placement algorithm for probe set reduction in network fault detection and localization

The number of malware samples intercepted and analyzed by antivirus providers has increased considerably in recent years. However, much of this software is essentially a repackaged version of malware that has already been identified. Consequently, assessing whether a piece of malware belongs to a known family or exhibits previously identified behavior that requires additional examination has become crucial. Random forest and Decision tree algorithms, as well as hybrid models of both algorithms, have been employed in past studies and research papers. We attempted to introduce an additional prediction technique known as SGD, which delivers good results when a dataset has over 100k variables (In our case 130k). As a result, SGD is one of our study paper’s distinguishing characteristics. Our approach has also been tested on both packed and obfuscated malware samples, ensuring that it is both reliable and scalable.

Malware Analysis Using Machine Learning Techniques

Yoga’s popularity is increasing daily. Doing Yoga regularly helps physically, mentally, and spiritually. Many people practice yoga without any formal training. However, practicing yoga incorrectly or even without proper supervision can lead to serious health problems such as stroke, nerve damage, and so on. As a result, there is a need for scientific posture analysis ensuring appropriate yoga posture is an essential aspect to consider. Pose identification strategies are effective in identifying posture and assisting people in performing yoga more effectively. Due to the paucity of data and real-time positioning, position detection is a difficult task. In this paper, we have done an extensive literature survey and offer a method for real-time pose estimation that uses the deep neural network(DNN) model to detect and fix errors in a person’s stance.