Isha Kansal - Academia.edu (original) (raw)

Papers by Isha Kansal

Research paper thumbnail of Weighted image de-fogging using luminance dark prior

Journal of Modern Optics, 2017

In this work, the weighted image de-fogging process based upon dark channel prior is modified by ... more In this work, the weighted image de-fogging process based upon dark channel prior is modified by using luminance dark prior. Dark channel prior estimates the transmission by using three colour channels whereas luminance dark prior does the same by making use of only Y component of YUV colour space. For each pixel in a patch of n × n size, the luminance dark prior uses n 2 pixels, rather than 3 × n 2 pixels used in DCP technique, which speeds up the de-fogging process. To estimate the transmission map, weighted approach based upon difference prior is used which mitigates halo artefacts at the time of transmission estimation. The major drawback of weighted technique is that it does not maintain the constancy of the transmission in a local patch even if there are no significant depth disruptions, due to which the de-fogged image looks over smooth and has low contrast. Apart from this, in some images, weighted transmission still carries less visible halo artefacts. Therefore, Gaussian filter is used to blur the estimated weighted transmission map which enhances the contrast of de-fogged images. In addition to this, a novel approach is proposed to remove the pixels belonging to bright light source(s) during the atmospheric light estimation process based upon histogram of YUV colour space. To show the effectiveness, the proposed technique is compared with existing techniques. This comparison shows that the proposed technique performs better than the existing techniques.

Research paper thumbnail of ROAD: Robotics-Assisted Onsite Data Collection and Deep Learning Enabled Robotic Vision System for Identification of Cracks on Diverse Surfaces

Sustainability

Crack detection on roads is essential nowadays because it has a significant impact on ensuring th... more Crack detection on roads is essential nowadays because it has a significant impact on ensuring the safety and reliability of road infrastructure. Thus, it is necessary to create more effective and precise crack detection techniques. A safer road network and a better driving experience for all road users can result from the implementation of the ROAD (Robotics-Assisted Onsite Data Collecting) system for spotting road cracks using deep learning and robots. The suggested solution makes use of a robot vision system’s capabilities to gather high-quality data about the road and incorporates deep learning methods for automatically identifying cracks. Among the tested algorithms, Xception stands out as the most accurate and predictive model, with an accuracy of over 90% during the validation process and a mean square error of only 0.03. In contrast, other deep neural networks, such as DenseNet201, InceptionResNetV2, MobileNetV2, VGG16, and VGG19, result in inferior accuracy and higher losse...

Research paper thumbnail of Face mask detection in foggy weather from digital images using transfer learning

The Imaging Science Journal

Research paper thumbnail of IoT-Fog-enabled robotics-based robust classification of hazy and normal season agricultural images for weed detection

Paladyn, Journal of Behavioral Robotics

The mechanization of farming is currently the most pressing problem facing humanity and a burgeon... more The mechanization of farming is currently the most pressing problem facing humanity and a burgeoning academic field. Over the last decade, there has been an explosion of Internet of Things (IoT) application growth in agriculture. Agricultural robotics is bringing about a new era of farming because they are growing more intelligent, recognizing causes of variation on the farm, consuming fewer resources, and optimizing their efficiency to more flexible jobs. The purpose of this article is to construct an IoT-Fog computing equipped robotic system for the categorization of weeds and soy plants during both the hazy season and the normal season. The used dataset in this article included four classes: soil, soybean, grass, and weeds. A two-dimensional Convolutional Neural Network (2D-CNN)-based deep learning (DL) approach was implemented for data image classification with dataset of height and width of 150 × 150 and of three channels. The overall proposed system is considered an IoT-connec...

Research paper thumbnail of A Comprehensive Review on Small Satellites: Services and Applications

Lecture notes in electrical engineering, 2023

Research paper thumbnail of A Comprehensive Review on Fake Images/Videos Detection Techniques

2022 10th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO)

Research paper thumbnail of Hybrid optimization for fault‐tolerant and accurate localization in mobility assisted underwater wireless sensor networks

International Journal of Communication Systems

Research paper thumbnail of Fusion based fast de-fogging for foggy images

AIP Conference Proceedings

Research paper thumbnail of Effect of non uniform illumination compensation on dehazing/de-fogging techniques

AIP Conference Proceedings

Research paper thumbnail of Digital Image Processing and IoT in Smart Health Care -A review

2022 International Conference on Emerging Smart Computing and Informatics (ESCI)

Health care and well-being are concerned with the upkeep or maintenance of humans through prevent... more Health care and well-being are concerned with the upkeep or maintenance of humans through preventative medicine, diagnosis, therapies, regeneration, or prevention of disease, ailment, injury, and other health-related conditions in people. Healthcare is unique in comparison to other industries. It is an elevated segment, and people expect the best possible care and services at all costs. Through continuous integration and resource optimization, the use of IoT technology in health applications enables the health care industry to improve care quality while lowering costs. The IoT in diagnostic imaging enables real-time identification and correction of imaging apparatus parameters due to the ease with which imaging apparatus parameters can be auto-analyzed. This paper discusses the impact of online image processing methods in IoT-based health care, which can be beneficial in the health sector for predicting some major human diseases. Due to individuality, image complex nature, extensive variation between interpreters, and fatigue, human experts' ability to interpret images is quite limited. We focus on the role of Digital Image Processing in disease detection, Image Dataset Preparation for Machine and Deep Learning, the role of Digital Image Processing in IOT based applications of health care, a case study of IoT-based healthcare application of disease classification.

Research paper thumbnail of Reverse Engineering-A Method for Analyzing Malicious Code Behavior

2021 International Conference on Advances in Computing, Communication, and Control (ICAC3), 2021

Millions of dangerous programs are encountered in everyday life as a result of increasing compute... more Millions of dangerous programs are encountered in everyday life as a result of increasing computer science advances. Malware is any harmful code that affects hardware, software, a user, or a network. Various malware analysis methods assume that some of the malware's broken code is accessible, which is not necessarily the case. Malware-based cyber-attacks are reported on a daily basis by security specialists all around the world. Malware analysis is the process of cleaning target PCs and devices to remove malware's destructive capability. Software reverse engineering is the process of assessing a software system in its entirety or in part in order to extract design, knowledge, code, and implementation information about the program without having to look at the source code. In this research paper, we present the results of malware analysis on infected binary files using static and dynamic analysis.

Research paper thumbnail of CheXImageNet: a novel architecture for accurate classification of Covid-19 with chest x-ray digital images using deep convolutional neural networks

Health and Technology, 2022

Many countries around the world have been influenced by Covid-19 which is a serious virus as it g... more Many countries around the world have been influenced by Covid-19 which is a serious virus as it gets transmitted by human communication. Although, its syndrome is quite similar to the ordinary flu. The critical step involved in Covid-19 is the initial screening or testing of the infected patients. As there are no special detection tools, the demand for such diagnostic tools has been increasing continuously. So, it is eminently admissible to find out positive cases of this disease at the earliest so that the spreading of this dangerous virus can be controlled. Although, some methods for the detection of Covid-19 patients are available, which are performed upon respiratory based samples and among them, a critical approach for treatment is radiologic imaging or X-ray imaging. The latest conclusions obtained from X-ray digital imaging based algorithms and techniques recommend that such type of digital images may consist of significant facts regarding the SARS-CoV-2 virus. The utilization of Deep Neural Networks based methodologies clubbed with digital radiological imaging has been proved useful for accurately identifying this disease. This could also be adjuvant in conquering the problem of dearth of competent physicians in far-flung areas. In this paper, a CheXImageNet model has been introduced for detecting Covid-19 disease by using digital images of Chest X-ray with the help of an openly accessible dataset. Experiments for both binary class and multi-class have been performed in this work for benchmarking the effectiveness of the proposed work. An accuracy of 100% is reported for both binary classification (having cases of Covid-19 and Normal X-Ray) and classification for three classes (including cases of Covid-19, Normal X-Ray and, cases of Pneumonia disease) respectively.

Research paper thumbnail of A Systematic Review on Techniques of Facemask Detection using Digital Images

2021 9th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO), 2021

The hasty spread of Coronavirus around the entire world prompted a number of red alerts in variou... more The hasty spread of Coronavirus around the entire world prompted a number of red alerts in various areas including human health, humanity and day to day living styles of human beings. During this pandemic situation, face masks have been used worldwide for protecting humans against the disease of Coronavirus. Face masks are the best and most effective known method to control the dissemination of Covid-19. It is really very difficult to correctly monitor whether the individual have worn face mask or not in order to notify the victim in crowded and hustled places. The face mask detection process generally uses a camera to capture live streaming videos of crowd places and converting them into images. The systems for mask detection are developed with a toll-way gate which allows only if the person crossing it has a face mask worn on his/her face or it does not allow the person in. This paper aims to review various techniques proposed for identifying whether the individuals are wearing facemask or not to protect themselves and others from this deadly virus.

Research paper thumbnail of Comparative Analysis of various Machine and Deep Learning Models for Face Mask Detection using Digital Images

2021 9th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO), 2021

COVID-19 has majorly influenced life of all human beings worldwide. It is also affecting human&#3... more COVID-19 has majorly influenced life of all human beings worldwide. It is also affecting human's daily life style. During this pandemic situation, face masks have been used worldwide for protection against the disease of Coronavirus. Face masks are the most accepted and operative way to prevent the spread of Covid-19 cases. It is really very difficult to correctly monitor whether the individual have worn face mask or not in order to notify the victim in crowded and hustled places. The face mask detection process generally uses a camera to capture live streaming videos of crowd places and converting them into images. These images acts as input for facial mask recognition or detection systems. The systems for mask detection are developed with a toll-way gate which allows only if the person crossing it has a face mask worn on his/her face or it does not allow the person in. This paper intends to compare various machine and deep learning based models for binary classification of digital images into two classes of images with and without mask. For this, various parameters including Accuracy, Precision, Fl-Score and Recall are obtained for various models. In the experimental results and analysis section, it has been shown that CNN model outperforms the other state-of-the-art models in terms of above parameters.

Research paper thumbnail of Machine Learning Based Security Solutions in MANETs: State of the art approaches

Journal of Physics: Conference Series, 2021

Machine learning (ML) techniques provide the learning capability to a system and encourage adapta... more Machine learning (ML) techniques provide the learning capability to a system and encourage adaptation into the environment, based upon many logical and statistical operations. The prime goal of ML is to recognize the complex patterns and make decisions based on the results. There are various ML algorithms which are implemented to secure the mobile ad-hoc networks. The infrastructure-less environment of MANETs poses a great challenge in implementation of the security systems. The security approaches in MANETs mainly focus on intrusion detection, malicious attacks mitigation, elimination of outlier/misbehavior/selfish nodes and securing routing paths. The researchers have been using cutting edge technologies for providing efficient security solutions by taking into the consideration of dynamic environment of MANETs. These technologies include machine learning, Artificial Intelligence (AI), Genetic Algorithms based methods, biological-inspired algorithms and so on. This paper presents ...

Research paper thumbnail of Real-Time Data Transfer in Marine Environment Monitoring Applications

Energy-Efficient Underwater Wireless Communications and Networking, 2021

Due to immense application scenarios emerged by underwater communication, the researchers have sh... more Due to immense application scenarios emerged by underwater communication, the researchers have shown their ardent interest in this field. Underwater wireless sensor network (UWSN) provides a promising approach for realism of such type of applications such as pollution monitoring, data agglomeration, natural calamities prediction, etc. In UWSN, sensor nodes are deployed at a number of different points of depths based upon the application requirements. Nodes at different locations collect the information and forward it to the sink node at surface. As a node cannot directly communicate to the sink node at surface, the multi hop communication takes place which poses great challenges underwater such as limited resource, bandwidth unavailability, end-to-end delays, temperature, water pressure, etc. However, congestion control is considered to be fundamental issue to be solved for effective data transfer underwater. This chapter presents the importance of real-time data in marine environme...

Research paper thumbnail of Classification and recognition of online hand-written alphabets using Machine Learning Methods

IOP Conference Series: Materials Science and Engineering, 2021

The hand-written alphabet recognition and classification plays an important role in pattern recog... more The hand-written alphabet recognition and classification plays an important role in pattern recognition, computer vision as well as image processing. In last few decades, a plethora of applications based on this area are developed such as sign identification, multi lingual learning systems etc. This paper classifies samples of hand-written alphabets into different classes using various machine learning methods. The challenging factor in hand written alphabets recognition lie in variations of style, shape and size of the letters. In this paper a simplified and accurate methodology is proposed based upon engineered features which are evaluated and tested using MatLab tool in comparison to other existing methods. The proposed system achieves a substantial amount of accuracy of 98% as compared to the state of the art approaches.

Research paper thumbnail of Minimum preserving subsampling-based fast image de-fogging

Journal of Modern Optics, 2018

Dark channel prior based techniques have been widely used in image and video de-fogging which pro... more Dark channel prior based techniques have been widely used in image and video de-fogging which produce real and impressive results. Their major limitation is the large computational cost of dark channel estimation. For the image of size M × N, to find n × n dark channel, 3 × n 2 × M × N operations are required, which increase its computational complexity. In this work, a novel approach of image subsampling is proposed, which preserves the value of local minimum in a patch. This subsampled image is used to construct the dark channel to improve the computational efficiency. Transmission map is refined using fast guided filter to remove blocking artifacts. Atmospheric light is calculated by ignoring pixels of bright light sources. To make the results uniformly bright, adaptive post processing is performed on de-fogging results. The image de-fogging technique is further extended for videos. It is demonstrated that proposed technique produces better results than existing state-of-the-art techniques while achieving real-time processing speed.

Research paper thumbnail of Shade dispersion methodologies for performance improvement of classical total cross‐tied photovoltaic array configuration under partial shading conditions

IET Renewable Power Generation, 2021

Large size photovoltaic (PV) systems face a large number of issues based on malfunction and unfav... more Large size photovoltaic (PV) systems face a large number of issues based on malfunction and unfavourable climatic conditions such as partial shading conditions (PSCs). These PSCs are the major causes of PV systems' performance degradation. In this paper, a symmetric matrix (SM) game puzzle is used to reconfigure the electrical connections of the PV array system. Present shade dispersion methodology is based on the 'physical reallocation of PV module-fixed electrical connections' principle. Modification in the electrical connections of conventional total cross-tied (TCT) PV array configuration introduces a new 'SM-TCT' configuration. An extensive comparative study of conventional TCT and novel-TCT (NTCT) configurations with proposed SM-TCT configuration prove the effectiveness to achieve higher performance. The MATLAB/Simulink results are obtained on the basis of the non-linear nature of current-voltage and power-voltage characteristics. SM-TCT, Shape-do-Ku, NTCT and TCT configurations are examined under three realistic PSCs in terms of power and voltage at global maximum power point, improved fill factor, reduced power losses, performance ratio and power enhancement. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Research paper thumbnail of Mathematical puzzle based PV array configuration for GMP enhancement under non-uniform irradiation

EAI Endorsed Transactions on Energy Web, 2018

This paper proposes a fast non uniform illumination compensation technique for different type of ... more This paper proposes a fast non uniform illumination compensation technique for different type of fog recovered images including satellite images. Till now the research of image dehazing/de-fogging is restricted to the consideration of fog in homogeneous environment. But generally the atmosphere is not homogeneous for example, there may be non uniform brightness, or external light sources may be present etc. Therefore, when the de-hazed or de-fogged image is recovered with the help of homogeneous physical image degradation model, it generally suffers from non uniform illumination or brightness. Therefore, the fog recovered images are further processed with the help of proposed technique and the final images produced are free from non uniform brightness or darkness. For improving the computational complexity of the de-fogging technique, image sub sampling process has been used in different steps of the proposed technique.

Research paper thumbnail of Weighted image de-fogging using luminance dark prior

Journal of Modern Optics, 2017

In this work, the weighted image de-fogging process based upon dark channel prior is modified by ... more In this work, the weighted image de-fogging process based upon dark channel prior is modified by using luminance dark prior. Dark channel prior estimates the transmission by using three colour channels whereas luminance dark prior does the same by making use of only Y component of YUV colour space. For each pixel in a patch of n × n size, the luminance dark prior uses n 2 pixels, rather than 3 × n 2 pixels used in DCP technique, which speeds up the de-fogging process. To estimate the transmission map, weighted approach based upon difference prior is used which mitigates halo artefacts at the time of transmission estimation. The major drawback of weighted technique is that it does not maintain the constancy of the transmission in a local patch even if there are no significant depth disruptions, due to which the de-fogged image looks over smooth and has low contrast. Apart from this, in some images, weighted transmission still carries less visible halo artefacts. Therefore, Gaussian filter is used to blur the estimated weighted transmission map which enhances the contrast of de-fogged images. In addition to this, a novel approach is proposed to remove the pixels belonging to bright light source(s) during the atmospheric light estimation process based upon histogram of YUV colour space. To show the effectiveness, the proposed technique is compared with existing techniques. This comparison shows that the proposed technique performs better than the existing techniques.

Research paper thumbnail of ROAD: Robotics-Assisted Onsite Data Collection and Deep Learning Enabled Robotic Vision System for Identification of Cracks on Diverse Surfaces

Sustainability

Crack detection on roads is essential nowadays because it has a significant impact on ensuring th... more Crack detection on roads is essential nowadays because it has a significant impact on ensuring the safety and reliability of road infrastructure. Thus, it is necessary to create more effective and precise crack detection techniques. A safer road network and a better driving experience for all road users can result from the implementation of the ROAD (Robotics-Assisted Onsite Data Collecting) system for spotting road cracks using deep learning and robots. The suggested solution makes use of a robot vision system’s capabilities to gather high-quality data about the road and incorporates deep learning methods for automatically identifying cracks. Among the tested algorithms, Xception stands out as the most accurate and predictive model, with an accuracy of over 90% during the validation process and a mean square error of only 0.03. In contrast, other deep neural networks, such as DenseNet201, InceptionResNetV2, MobileNetV2, VGG16, and VGG19, result in inferior accuracy and higher losse...

Research paper thumbnail of Face mask detection in foggy weather from digital images using transfer learning

The Imaging Science Journal

Research paper thumbnail of IoT-Fog-enabled robotics-based robust classification of hazy and normal season agricultural images for weed detection

Paladyn, Journal of Behavioral Robotics

The mechanization of farming is currently the most pressing problem facing humanity and a burgeon... more The mechanization of farming is currently the most pressing problem facing humanity and a burgeoning academic field. Over the last decade, there has been an explosion of Internet of Things (IoT) application growth in agriculture. Agricultural robotics is bringing about a new era of farming because they are growing more intelligent, recognizing causes of variation on the farm, consuming fewer resources, and optimizing their efficiency to more flexible jobs. The purpose of this article is to construct an IoT-Fog computing equipped robotic system for the categorization of weeds and soy plants during both the hazy season and the normal season. The used dataset in this article included four classes: soil, soybean, grass, and weeds. A two-dimensional Convolutional Neural Network (2D-CNN)-based deep learning (DL) approach was implemented for data image classification with dataset of height and width of 150 × 150 and of three channels. The overall proposed system is considered an IoT-connec...

Research paper thumbnail of A Comprehensive Review on Small Satellites: Services and Applications

Lecture notes in electrical engineering, 2023

Research paper thumbnail of A Comprehensive Review on Fake Images/Videos Detection Techniques

2022 10th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO)

Research paper thumbnail of Hybrid optimization for fault‐tolerant and accurate localization in mobility assisted underwater wireless sensor networks

International Journal of Communication Systems

Research paper thumbnail of Fusion based fast de-fogging for foggy images

AIP Conference Proceedings

Research paper thumbnail of Effect of non uniform illumination compensation on dehazing/de-fogging techniques

AIP Conference Proceedings

Research paper thumbnail of Digital Image Processing and IoT in Smart Health Care -A review

2022 International Conference on Emerging Smart Computing and Informatics (ESCI)

Health care and well-being are concerned with the upkeep or maintenance of humans through prevent... more Health care and well-being are concerned with the upkeep or maintenance of humans through preventative medicine, diagnosis, therapies, regeneration, or prevention of disease, ailment, injury, and other health-related conditions in people. Healthcare is unique in comparison to other industries. It is an elevated segment, and people expect the best possible care and services at all costs. Through continuous integration and resource optimization, the use of IoT technology in health applications enables the health care industry to improve care quality while lowering costs. The IoT in diagnostic imaging enables real-time identification and correction of imaging apparatus parameters due to the ease with which imaging apparatus parameters can be auto-analyzed. This paper discusses the impact of online image processing methods in IoT-based health care, which can be beneficial in the health sector for predicting some major human diseases. Due to individuality, image complex nature, extensive variation between interpreters, and fatigue, human experts' ability to interpret images is quite limited. We focus on the role of Digital Image Processing in disease detection, Image Dataset Preparation for Machine and Deep Learning, the role of Digital Image Processing in IOT based applications of health care, a case study of IoT-based healthcare application of disease classification.

Research paper thumbnail of Reverse Engineering-A Method for Analyzing Malicious Code Behavior

2021 International Conference on Advances in Computing, Communication, and Control (ICAC3), 2021

Millions of dangerous programs are encountered in everyday life as a result of increasing compute... more Millions of dangerous programs are encountered in everyday life as a result of increasing computer science advances. Malware is any harmful code that affects hardware, software, a user, or a network. Various malware analysis methods assume that some of the malware's broken code is accessible, which is not necessarily the case. Malware-based cyber-attacks are reported on a daily basis by security specialists all around the world. Malware analysis is the process of cleaning target PCs and devices to remove malware's destructive capability. Software reverse engineering is the process of assessing a software system in its entirety or in part in order to extract design, knowledge, code, and implementation information about the program without having to look at the source code. In this research paper, we present the results of malware analysis on infected binary files using static and dynamic analysis.

Research paper thumbnail of CheXImageNet: a novel architecture for accurate classification of Covid-19 with chest x-ray digital images using deep convolutional neural networks

Health and Technology, 2022

Many countries around the world have been influenced by Covid-19 which is a serious virus as it g... more Many countries around the world have been influenced by Covid-19 which is a serious virus as it gets transmitted by human communication. Although, its syndrome is quite similar to the ordinary flu. The critical step involved in Covid-19 is the initial screening or testing of the infected patients. As there are no special detection tools, the demand for such diagnostic tools has been increasing continuously. So, it is eminently admissible to find out positive cases of this disease at the earliest so that the spreading of this dangerous virus can be controlled. Although, some methods for the detection of Covid-19 patients are available, which are performed upon respiratory based samples and among them, a critical approach for treatment is radiologic imaging or X-ray imaging. The latest conclusions obtained from X-ray digital imaging based algorithms and techniques recommend that such type of digital images may consist of significant facts regarding the SARS-CoV-2 virus. The utilization of Deep Neural Networks based methodologies clubbed with digital radiological imaging has been proved useful for accurately identifying this disease. This could also be adjuvant in conquering the problem of dearth of competent physicians in far-flung areas. In this paper, a CheXImageNet model has been introduced for detecting Covid-19 disease by using digital images of Chest X-ray with the help of an openly accessible dataset. Experiments for both binary class and multi-class have been performed in this work for benchmarking the effectiveness of the proposed work. An accuracy of 100% is reported for both binary classification (having cases of Covid-19 and Normal X-Ray) and classification for three classes (including cases of Covid-19, Normal X-Ray and, cases of Pneumonia disease) respectively.

Research paper thumbnail of A Systematic Review on Techniques of Facemask Detection using Digital Images

2021 9th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO), 2021

The hasty spread of Coronavirus around the entire world prompted a number of red alerts in variou... more The hasty spread of Coronavirus around the entire world prompted a number of red alerts in various areas including human health, humanity and day to day living styles of human beings. During this pandemic situation, face masks have been used worldwide for protecting humans against the disease of Coronavirus. Face masks are the best and most effective known method to control the dissemination of Covid-19. It is really very difficult to correctly monitor whether the individual have worn face mask or not in order to notify the victim in crowded and hustled places. The face mask detection process generally uses a camera to capture live streaming videos of crowd places and converting them into images. The systems for mask detection are developed with a toll-way gate which allows only if the person crossing it has a face mask worn on his/her face or it does not allow the person in. This paper aims to review various techniques proposed for identifying whether the individuals are wearing facemask or not to protect themselves and others from this deadly virus.

Research paper thumbnail of Comparative Analysis of various Machine and Deep Learning Models for Face Mask Detection using Digital Images

2021 9th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO), 2021

COVID-19 has majorly influenced life of all human beings worldwide. It is also affecting human&#3... more COVID-19 has majorly influenced life of all human beings worldwide. It is also affecting human's daily life style. During this pandemic situation, face masks have been used worldwide for protection against the disease of Coronavirus. Face masks are the most accepted and operative way to prevent the spread of Covid-19 cases. It is really very difficult to correctly monitor whether the individual have worn face mask or not in order to notify the victim in crowded and hustled places. The face mask detection process generally uses a camera to capture live streaming videos of crowd places and converting them into images. These images acts as input for facial mask recognition or detection systems. The systems for mask detection are developed with a toll-way gate which allows only if the person crossing it has a face mask worn on his/her face or it does not allow the person in. This paper intends to compare various machine and deep learning based models for binary classification of digital images into two classes of images with and without mask. For this, various parameters including Accuracy, Precision, Fl-Score and Recall are obtained for various models. In the experimental results and analysis section, it has been shown that CNN model outperforms the other state-of-the-art models in terms of above parameters.

Research paper thumbnail of Machine Learning Based Security Solutions in MANETs: State of the art approaches

Journal of Physics: Conference Series, 2021

Machine learning (ML) techniques provide the learning capability to a system and encourage adapta... more Machine learning (ML) techniques provide the learning capability to a system and encourage adaptation into the environment, based upon many logical and statistical operations. The prime goal of ML is to recognize the complex patterns and make decisions based on the results. There are various ML algorithms which are implemented to secure the mobile ad-hoc networks. The infrastructure-less environment of MANETs poses a great challenge in implementation of the security systems. The security approaches in MANETs mainly focus on intrusion detection, malicious attacks mitigation, elimination of outlier/misbehavior/selfish nodes and securing routing paths. The researchers have been using cutting edge technologies for providing efficient security solutions by taking into the consideration of dynamic environment of MANETs. These technologies include machine learning, Artificial Intelligence (AI), Genetic Algorithms based methods, biological-inspired algorithms and so on. This paper presents ...

Research paper thumbnail of Real-Time Data Transfer in Marine Environment Monitoring Applications

Energy-Efficient Underwater Wireless Communications and Networking, 2021

Due to immense application scenarios emerged by underwater communication, the researchers have sh... more Due to immense application scenarios emerged by underwater communication, the researchers have shown their ardent interest in this field. Underwater wireless sensor network (UWSN) provides a promising approach for realism of such type of applications such as pollution monitoring, data agglomeration, natural calamities prediction, etc. In UWSN, sensor nodes are deployed at a number of different points of depths based upon the application requirements. Nodes at different locations collect the information and forward it to the sink node at surface. As a node cannot directly communicate to the sink node at surface, the multi hop communication takes place which poses great challenges underwater such as limited resource, bandwidth unavailability, end-to-end delays, temperature, water pressure, etc. However, congestion control is considered to be fundamental issue to be solved for effective data transfer underwater. This chapter presents the importance of real-time data in marine environme...

Research paper thumbnail of Classification and recognition of online hand-written alphabets using Machine Learning Methods

IOP Conference Series: Materials Science and Engineering, 2021

The hand-written alphabet recognition and classification plays an important role in pattern recog... more The hand-written alphabet recognition and classification plays an important role in pattern recognition, computer vision as well as image processing. In last few decades, a plethora of applications based on this area are developed such as sign identification, multi lingual learning systems etc. This paper classifies samples of hand-written alphabets into different classes using various machine learning methods. The challenging factor in hand written alphabets recognition lie in variations of style, shape and size of the letters. In this paper a simplified and accurate methodology is proposed based upon engineered features which are evaluated and tested using MatLab tool in comparison to other existing methods. The proposed system achieves a substantial amount of accuracy of 98% as compared to the state of the art approaches.

Research paper thumbnail of Minimum preserving subsampling-based fast image de-fogging

Journal of Modern Optics, 2018

Dark channel prior based techniques have been widely used in image and video de-fogging which pro... more Dark channel prior based techniques have been widely used in image and video de-fogging which produce real and impressive results. Their major limitation is the large computational cost of dark channel estimation. For the image of size M × N, to find n × n dark channel, 3 × n 2 × M × N operations are required, which increase its computational complexity. In this work, a novel approach of image subsampling is proposed, which preserves the value of local minimum in a patch. This subsampled image is used to construct the dark channel to improve the computational efficiency. Transmission map is refined using fast guided filter to remove blocking artifacts. Atmospheric light is calculated by ignoring pixels of bright light sources. To make the results uniformly bright, adaptive post processing is performed on de-fogging results. The image de-fogging technique is further extended for videos. It is demonstrated that proposed technique produces better results than existing state-of-the-art techniques while achieving real-time processing speed.

Research paper thumbnail of Shade dispersion methodologies for performance improvement of classical total cross‐tied photovoltaic array configuration under partial shading conditions

IET Renewable Power Generation, 2021

Large size photovoltaic (PV) systems face a large number of issues based on malfunction and unfav... more Large size photovoltaic (PV) systems face a large number of issues based on malfunction and unfavourable climatic conditions such as partial shading conditions (PSCs). These PSCs are the major causes of PV systems' performance degradation. In this paper, a symmetric matrix (SM) game puzzle is used to reconfigure the electrical connections of the PV array system. Present shade dispersion methodology is based on the 'physical reallocation of PV module-fixed electrical connections' principle. Modification in the electrical connections of conventional total cross-tied (TCT) PV array configuration introduces a new 'SM-TCT' configuration. An extensive comparative study of conventional TCT and novel-TCT (NTCT) configurations with proposed SM-TCT configuration prove the effectiveness to achieve higher performance. The MATLAB/Simulink results are obtained on the basis of the non-linear nature of current-voltage and power-voltage characteristics. SM-TCT, Shape-do-Ku, NTCT and TCT configurations are examined under three realistic PSCs in terms of power and voltage at global maximum power point, improved fill factor, reduced power losses, performance ratio and power enhancement. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Research paper thumbnail of Mathematical puzzle based PV array configuration for GMP enhancement under non-uniform irradiation

EAI Endorsed Transactions on Energy Web, 2018

This paper proposes a fast non uniform illumination compensation technique for different type of ... more This paper proposes a fast non uniform illumination compensation technique for different type of fog recovered images including satellite images. Till now the research of image dehazing/de-fogging is restricted to the consideration of fog in homogeneous environment. But generally the atmosphere is not homogeneous for example, there may be non uniform brightness, or external light sources may be present etc. Therefore, when the de-hazed or de-fogged image is recovered with the help of homogeneous physical image degradation model, it generally suffers from non uniform illumination or brightness. Therefore, the fog recovered images are further processed with the help of proposed technique and the final images produced are free from non uniform brightness or darkness. For improving the computational complexity of the de-fogging technique, image sub sampling process has been used in different steps of the proposed technique.