dharmendra Roy

Effective integration of the internet of things and ensemble learning approaches for enhancing sudden topple recognition assisted by cloud computing technology

MATEC web of conferences, 2024

Falls provide a significant public health hazard globally for the senior population. Untreated Su... more Falls provide a significant public health hazard globally for the senior population. Untreated Sudden Topple in the elderly leads to functional loss and a notable decline in mobility, autonomy, and quality of life. Early identification of Sudden Topple is essential for a person's wellbeing or to provide needed care. Telehealth data centers need scalable processing and storing resources to accommodate the increasing number of individuals. Specialized methods that enable the transfer of just pertinent data are necessary. This study presents a Hybrid System composing Cloud Computing and the Internet of Things (IoT) (HS-CC-IoT) to monitor many elderly individuals, identify Sudden Topple, and alert caretakers. The experiments were conducted to reveal the necessary criteria for facilitating the operation of large-scale systems. The research assessed many machine learning algorithms for their appropriateness in detection. The experimental tests to identify sudden topples are in cloud-based data centers and on an Edge IoT gadget with an Ensemble Learning Algorithm. Experiments on the user-to-cloud data transfer showed that a substantial decrease in the quantity of saved and transferred data is possible when conducting Sudden Topple identification on the Edge.

Download

Inflate Agriculture Crop Yield Prophesy through Deep Neural Network Schemes

Genetic K-Means Clustering Algorithm for Mixed Numeric and Categorical Data Sets

International Journal of Artificial Intelligence & Applications, Apr 25, 2010

Clustering is one of the major data mining tasks and aims at grouping the data objects into meani... more Clustering is one of the major data mining tasks and aims at grouping the data objects into meaningful classes (clusters) such that the similarity of objects within clusters is maximized, and the similarity of objects from different clusters is minimized. In this paper we present a clustering algorithm based on Genetic k-means paradigm that works well for data with mixed numeric and categorical features. We propose a modified description of cluster center to overcome the numeric data only limitation of Genetic k-mean algorithm and provide a better characterization of clusters. The performance of this algorithm has been studied on benchmark data sets.

Download

Web Crawlers and Web Crawling Algorithms - A Review

International journal of scientific research in science, engineering and technology, Mar 30, 2016

As there is profound web development, there has been expanded enthusiasm for methods that help pr... more As there is profound web development, there has been expanded enthusiasm for methods that help productively find profound web interfaces. Because of accessibility of inexhaustible information on web, seeking has a noteworthy effect. On-going examines place accentuation on the pertinence and strength of the information found, as the found examples closeness is a long way from the investigated. Notwithstanding their importance pages for any inquiry subject, the outcomes are colossal to be investigated. One issue of pursuit on the Web is that internet searchers return huge hit records with low accuracy. Clients need to filter applicable reports from insignificant ones by physically bringing and skimming pages. Another debilitating viewpoint is that URLs or entire pages are returned as list items. It is likely that the response to a client question is just part of the page. Recovering the entire page really leaves the errand of inquiry inside a page to Web clients. With these two viewpoints staying unaltered, Web clients won't be liberated from the substantial weight of perusing pages and finding required data, and data got

Offline Signature Verification for Detecting Signature Forgery: A Comparative Study

International Journal of Computer Trends and Technology, Mar 25, 2015

As signature is generally used as a means of individual verification, there is a need for an auto... more As signature is generally used as a means of individual verification, there is a need for an automatic verification system. Signatures provide a safe means of verification and authorization in authorized documents. However one of the key challenges is the ability of the system to detect skilled and unskilled forgery. Many cases of bank cheque forgeries have been reported. Most of the offline signature verification system adopts recognition based technique where the system classifies a given signature sample as one of the samples from the database. However detection of a forgery in a given sample is challenging as the input sample looks alike to one of the samples in the database. A simple and a consistent system has to be designed which should identify various types of forgeries. Various approaches have been used to implement biometric signature verification some of which are dynamic time warping (DTW), Bayesian Learning, Template Matching Technique, Hidden Markov Model (HMM), Support Vector Machine (SVM) etc. This paper presents a comparative and qualitative study of these methods used for offline signature verification. Keywords—Skilled and Unskilled Forgery, Signature Verification, Forgery detection, Dynamic Time Warping, Bayesian Learning, Template Matching Technique, Hidden Markov Model (HMM), Support Vector Machine (SVM).

Music Recommendation Based on Current Mood Using Ai &amp; Ml

Research Square (Research Square), Mar 24, 2023

Music always has a special connection to our emotions. It is a method of connecting people all ov... more Music always has a special connection to our emotions. It is a method of connecting people all over the world through pure emotions. When all of this is considered, it is exceedingly difficult to generalize music and state that everyone would enjoy the same type. Music recommendation based on mood is greatly needed because they will assist humans in relieving stress and listening to soothing music based on their current emotions. Its primary goal is to accurately predict the user's mood, and songs are played by the application based on the user's selection as well as his current mood. This bot recognizes human emotions from facial images using Human-Computer Interaction (HCI). The extraction of facial elements from the user's face is another critical factor. We used the CNN Algorithm to accurately detect the user's face in the live webcam feed and to detect emotion based on facial attributes such as the arrangement of the user's mouth and eyes. There is also a questionnaire where we can directly select a good option.

Download

Digital Image Line Extraction and Edge Smoothing and Quality Analysis

International Journal of Technology, Jan 8, 2013

Face Mask Detection using MobilenetV2 of Convolution Neural Network

International Research Journal on Advanced Science Hub, May 1, 2023

In light of the fact that the COVID-19 pandemic is spreading swiftly over the earth, it is crucia... more In light of the fact that the COVID-19 pandemic is spreading swiftly over the earth, it is crucial to create new technologies to study and resist the disease. Face masks and gloves are required for protection against the coronavirus, and scientists and doctors have advised everyone to wear the mask for the whole day. Thus, various procedures are accessible to different people wearing face masks. Masks are advised as a basic barrier to stop respirational beads from receiving into the air and against other people once someone is found to be using cover hacks. Additionally, this is known as source governance. This article is based on the current understanding of respiratory beads' function in the escalation of COVID-19 infection. In this problem, the face mask procedure was built using MobileNetV2. Compared to the current system, Mobile Net V2 can be used to identify face masks among individuals with greater accuracy. The input data file contains 500 images taken from the Kaggle face mask Detection Dataset. A scene with a mix of people donning masks and without mask. The output is a segmented picture of the same. Later, this process is improved by using a webcam to capture real-time video. The video is then segmented into the frame and resized as required, and the result is a video-segmented image. The model was then run to determine whether or not individuals were wearing masks after performing the pre-processing function. An accuracy of 80 was used to acquire the results.

Download

Air Quality Hotspot Monitoring with Trajectories of IoT in Smart City Implementation

2023 3rd International Conference on Pervasive Computing and Social Networking (ICPCSN)

Face Mask Detection using MobilenetV2 of Convolution Neural Network

International Research Journal on Advanced Science Hub

In light of the fact that the COVID-19 pandemic is spreading swiftly over the earth, it is crucia... more In light of the fact that the COVID-19 pandemic is spreading swiftly over the earth, it is crucial to create new technologies to study and resist the disease. Face masks and gloves are required for protection against the coronavirus, and scientists and doctors have advised everyone to wear the mask for the whole day. Thus, various procedures are accessible to different people wearing face masks. Masks are advised as a basic barrier to stop respirational beads from receiving into the air and against other people once someone is found to be using cover hacks. Additionally, this is known as source governance. This article is based on the current understanding of respiratory beads' function in the escalation of COVID-19 infection. In this problem, the face mask procedure was built using MobileNetV2. Compared to the current system, Mobile Net V2 can be used to identify face masks among individuals with greater accuracy. The input data file contains 500 images taken from the Kaggle face mask Detection Dataset. A scene with a mix of people donning masks and without mask. The output is a segmented picture of the same. Later, this process is improved by using a webcam to capture real-time video. The video is then segmented into the frame and resized as required, and the result is a video-segmented image. The model was then run to determine whether or not individuals were wearing masks after performing the pre-processing function. An accuracy of 80 was used to acquire the results.

Download

Web Crawlers and Web Crawling Algorithms - A Review

International journal of scientific research in science, engineering and technology, Mar 30, 2016

As there is profound web development, there has been expanded enthusiasm for methods that help pr... more As there is profound web development, there has been expanded enthusiasm for methods that help productively find profound web interfaces. Because of accessibility of inexhaustible information on web, seeking has a noteworthy effect. On-going examines place accentuation on the pertinence and strength of the information found, as the found examples closeness is a long way from the investigated. Notwithstanding their importance pages for any inquiry subject, the outcomes are colossal to be investigated. One issue of pursuit on the Web is that internet searchers return huge hit records with low accuracy. Clients need to filter applicable reports from insignificant ones by physically bringing and skimming pages. Another debilitating viewpoint is that URLs or entire pages are returned as list items. It is likely that the response to a client question is just part of the page. Recovering the entire page really leaves the errand of inquiry inside a page to Web clients. With these two viewpoints staying unaltered, Web clients won't be liberated from the substantial weight of perusing pages and finding required data, and data got

Digital Image Line Extraction and Edge Smoothing and Quality Analysis

An Overly Architecture for throughput Optimal Routing using Cluster Based network

Cluster-based network control structures endorse more capable utilize of resources in controlling... more Cluster-based network control structures endorse more capable utilize of resources in controlling large dynamic networks. Therefore they are very good candidates for ad hoc networks. There are many types of cluster-based architectures proposed in the literature for definite purposes. Of them link-cluster architecture is specifically designed for transmission management in ad hoc networks. Bequest networks are frequently planned to operate with plain single-path routing, like the shortest path, which is known to be throughput suboptimal. On the other hand, earlier proposed throughput optimal policies (i.e., backpressure) need each machine in the network to make active routing decisions. We study overlay architecture for dynamic routing; only a subset of devices (overlay nodes) require to make the dynamic routing decisions. We conclude the necessary compilation of nodes that must bifurcate traffic for achieving the utmost multi-commodity network throughput. Keywords— Cluster based, ov...

Download

Segmentation of Liver using Hybrid K-means Clustering and Level Set

Liver Segmentation from Computed Tomography (CT) is a significant but challenging task. In the ab... more Liver Segmentation from Computed Tomography (CT) is a significant but challenging task. In the abdominal CT scan image the liver is seen clearly nevertheless the physician needs the accurate boundaries and correct size measurement for the liver transplantation surgery. The boundaries of the various organs are nor clearly visible as a result of complex structure of the human body. This paper presents an efficient and fully automated liver segmentation technique. The proposed method uses Ant Colony Optimization technique along side K-Means and Level set to locate the liver region. Prior to the liver transplantation surgery the livers that belong to the living donor and the recipient are evaluated to spot liver region and determine the size mismatch. This knowledge helps to decide if the donor and recipient is good match or not. The essential problem in manual segmentation is that boundaries of liver may be identified differently by different radiologists and even by the same radiologi...

Efficient Frequent Pattern Mining in Web Log

Proceedings of the International Conference on Computer Applications — Database Systems, 2010

ABSTRACT Semantic Web Mining aims at combining the two fast-developing research areas Semantic We... more ABSTRACT Semantic Web Mining aims at combining the two fast-developing research areas Semantic Web and Web Mining. The idea is to improve, on the one hand, the results of Web Mining by exploiting the new semantic structures in the Web; and to make use of Web Mining, on the other hand, for building up the Semantic Web. Frequent pattern mining is an important research area in the field of web data mining with wide range of applications. One of them is to use frequent pattern discovery methods in Web log data. The aim of discovering frequent patterns in Web log data is to obtain information about the navigational behaviour of the users. In this paper we give road map of semantic web and web mining and we present P and T tree based pattern mining approaches for the Web usage mining. Our experiment shows that P and T tree based pattern mining approaches is give better performance than existing one.

Music Recommendation Based on Current Mood Using Ai & Ml

Music always has a special connection to our emotions. It is a method of connecting people all ov... more Music always has a special connection to our emotions. It is a method of connecting people all over the world through pure emotions. When all of this is considered, it is exceedingly difficult to generalize music and state that everyone would enjoy the same type. Music recommendation based on mood is greatly needed because they will assist humans in relieving stress and listening to soothing music based on their current emotions. Its primary goal is to accurately predict the user's mood, and songs are played by the application based on the user's selection as well as his current mood. This bot recognizes human emotions from facial images using Human-Computer Interaction (HCI). The extraction of facial elements from the user’s face is another critical factor. We used the CNN Algorithm to accurately detect the user's face in the live webcam feed and to detect emotion based on facial attributes such as the arrangement of the user's mouth and eyes. There is also a questi...

Digital Image Line Extraction Using Adaptive Filtering and Edge Smoothing Using Cluster Analysis

ion” IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS, VOL. 15, NO. 1, JANUARY/FEBRUARY 2... more ion” IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS, VOL. 15, NO. 1, JANUARY/FEBRUARY 2009 [6] John Canny, Member, IEEE“A Computational Approach to Edge Detection“, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. Pami-8, No. 6, November 1986 [7] Mohammadreza Asghari Oskoei and Huosheng Hu:“A Survey on Edge Detection Methods “ 29 February 2010, School of Computer Science & Electronic Engineering ,University of Essex,Colchester,United Kingdom [8] Saritha A K, MTech Student,MES College of Engineering Ameera P.M, Assistant Professor,MES College of Engineering “Image segmentation based on kernel fuzzy C means clustering using edge detection method on noisy images “International Journal of Advanced Research in Computer Engineering & Technology (IJARCET) Volume 2, Issue 2, February 2013 [9] V.C. Dinh ,R. P. W. Duin, Raimund Leitner, Pavel Paclik“Hyper spectral Image Analysis, Multi-dimensional Image Analysis, Edge Detection, Clustering Algorithm”,ASCI 2009

Object detection in low contrast image

Indian journal of applied research, 2015

Increase of scope in machine learning is always move around image processing. But processing whol... more Increase of scope in machine learning is always move around image processing. But processing whole image is very long-lasting task so segmentation of image into area of interest come into play, it reduces computational time as well as reduce the memory to store and reduces bandwidth in case of transmission of image features. If I say image is cluster of various object “having same pixel value” according to segmentation algorithm, we classify them in three parts: region based segmentation method, data clustering and edge based segmentation. In this paper we survey algorithms available for segmentation, their specialty and disadvantages.

A Novel Color Image Denoising Technique Using Window Based Soft Fuzzy Filter

The image corrupted by different kinds of noises is a frequently encountered problem in image acq... more The image corrupted by different kinds of noises is a frequently encountered problem in image acquisition and transmission. The noise comes from noisy channel transmission errors. The impulse noise (or salt and pepper noise) is caused by sharp, unexpected disturbances in the image signal; its appearance is randomly scattered white or black (or both) pixels over the image. Gaussian noise is an idealized form of white noise, which is caused by some random fluctuations in the signal. Speckle noise (or more simply just speckle) can be modelled by random values multiplied by pixel values; hence it is also called multiplicative noise. This work presents a novel technique for edge preserved color image denoising using window based soft fuzzy filter based on asymmetrical triangular membership function. However lots of techniques like median, mean and average filters are available for gray image denoising, but most of the time it is found that all these filters are capable to provide good no...

Automatic Number Plate Recognition System

Indian Journal of Science and Technology, 2017

Download

Uploads

Papers by dharmendra Roy

Log In