Prashantha HS - Academia.edu (original) (raw)
Papers by Prashantha HS
International Journal of Creative Research Thoughts (IJCRT), 2023
Machine learning (ML) is a subset of artificial intelligence (AI) where computer is trained to ma... more Machine learning (ML) is a subset of artificial intelligence (AI) where computer is trained to make decisions like a human based on different characteristics from a set of data. Machine Learning is a subset of artificial intelligence (AI) which allows the software applications to predict the outcomes by considering the historical data as input. The paper presented discusses the detailed understanding and the contribution of different researchers in the area of Biomedical Signal processing.
International Journal of Creative Research Thoughts (IJCRT), 2023
Digital Image processing is an algorithm used to perform operations on a digital image, in order ... more Digital Image processing is an algorithm used to perform operations on a digital image, in order to extract some useful information or process images to enhance information from it. Face recognition is a method of identifying or verifying the identity of an individual using their face. Face recognition systems can be used to identify people in photos, video in real-time or offline. Machine learning algorithms can be used to identify, collect, store and evaluate face characteristics so that they can be matched to photos of people in a database. The paper presents the contributions made by different researchers to understand the methodology, tools and applications of face recognition.
International Journal of Creative Research Thoughts (IJCRT), 2023
Speech recognition is the process of converting human sound signals into words or instructions. T... more Speech recognition is the process of converting human sound signals into words or instructions. The research of speech recognition involves many subject areas such as computer technology, artificial intelligence, digital signal processing, pattern recognition, acoustics, linguistics, and cognitive science.Our speech is made up of many frequencies at the same time. The actual signal is really a sum of all those frequencies stuck together. The conversation or speech that is captured by a microphone or a telephone is converted from acoustic signal to a set of words in speech recognition.
International Journal of Creative Research Thoughts (IJCRT), 2023
Perfect weather predictions are needed for a plethora of activities and it was one of the main ch... more Perfect weather predictions are needed for a plethora of activities and it was one of the main challenging problems the entire world faced because of its multiple dimensions and non-linear trends. Weather depends on multiple climatic conditions like temperature, air pressure, humidity, wind flow speed and direction, cloud height and density, and rainfall. The most common phrases/keywords in weather prediction related articles were 'wind', 'precipitation', 'climate change', 'wind forecasting' and 'ensemble prediction'. The most common countries in which surveys were done are China, USA, Australia, India and Germany. Extreme meteorological events are often related to the occurrence of weather fronts.
International Journal of Creative Research Thoughts (IJCRT), 2023
In machine learning, classification is a supervised learning concept which basically categorizes ... more In machine learning, classification is a supervised learning concept which basically categorizes a set of data into classes. The most common classification problems arespeech recognition, face detection, handwriting recognition, and image classification. Image Classification is one of the most fundamental tasks in computer vision. It has revolutionized and propelled technological advancements in the most prominent fields, including the automobile industry, healthcare, manufacturing, and more. Image Classification (often referred to as Image Recognition) is the task of associating one (single-label classification) or more (multilabel classification) labels to a given image. Image classification is the process of predicting a specific class, or label, for something that is defined by a set of data points. Image classification is a subset of the classification problem, where an entire image is assigned a label. Grouping images into semantically meaningful categories using low-level visual features is a challenging and important problem in contentbased image retrieval. Based on these groupings, effective indices can be built for an image database
International Journal of Creative Research Thoughts (IJCRT), 2023
Stock market prediction and analysis are some of the most difficult jobs to complete. There are n... more Stock market prediction and analysis are some of the most difficult jobs to complete. There are numerous causes for this, including market volatility and a variety of other dependent and independent variables that influence the value of a certain stock in the market. These variables make it extremely difficult for any stock market expert to anticipate the rise and fall of the market with great precision. However, with the introduction of Machine Learning and its strong algorithms, the most recent market research and Stock Market Prediction advancements have begun to include such approaches in analysing stock market data. The Opening Value of the stock, the Highest and Lowest values of that stock on the same days, as well as the Closing Value at the end of the day, are all indicated for each date. Furthermore, the total volume of the stocks in the market is provided, with this information, it is up to the job of a Machine Learning Data Scientist to look at the data and develop different algorithms that may help in finding appropriate stocks values.
International Journal of Creative Research Thoughts (IJCRT), 2023
Many deep learning research works have shown successful results, primarily focusing on three data... more Many deep learning research works have shown successful results, primarily focusing on three data types: images, speech, and text. In addition, deep learning has also been successfully applied to communication signals/packets. Widely used applications of these kind of data are image classification, speech recognition, regression problem, pattern recognition, and text sentiment classification. The most fascinating of all is video data. Video data is also interesting for research from the perspective of its big size and dimension. Millions of video data are uploaded every day on YouTube; thus, it becomes a rich repository and empowered artificial intelligence (AI) research
International Journal of Creative Research Thoughts (IJCRT), 2023
Many deep learning research works have shown successful results, primarily focusing on three data... more Many deep learning research works have shown successful results, primarily focusing on three data types: images, speech, and text. In addition, deep learning has also been successfully applied to communication signals/packets. Widely used applications of these kind of data are image classification, speech recognition, regression problem, pattern recognition, and text sentiment classification. The most fascinating of all is video data. Video data is also interesting for research from the perspective of its big size and dimension. Millions of video data are uploaded every day on YouTube; thus, it becomes a rich repository and empowered artificial intelligence (AI) research
International Journal of Creative Research Thoughts (IJCRT), 2023
The goal of this proposed work is to create an Android-based image text-to-speech (ITTS) applicat... more The goal of this proposed work is to create an Android-based image text-to-speech (ITTS) application that enables users to translate text in photographs into spoken in formation in the language of their choice. The ability for users to customize the language in which the synthesized voice is produced is one of the application's standout features. Because of its user-friendly interface, a wide audience can access the Android application. Performance of an Android application is evaluated by precision, reactivity, and ability to customize language. This proposed workcan serve a variety of user demands, such as language learners, visually impaired people, and people looking for portable, effective tools for information consumption.
International Journal of Creative Research Thoughts (IJCRT), 2023
The human visual system effortlessly recognizes and identifies objects within its field of view, ... more The human visual system effortlessly recognizes and identifies objects within its field of view, showcasing remarkable speed and accuracy in tasks such as object detection and identification. Valuable clues regarding the origin and evolution of the universe are intricately woven into the shapes and formations of galaxies. However, automating the classification of galaxies from their images poses challenges due to the faintness of the galaxy images, interference from bright background stars, and inherent image noise. To address these complexities, we propose a method for the automatic detection and classification of galaxies. This method incorporates a data augmentation procedure, enhancing the robustness of trained models against data variations arising from different instruments and contrast-stretching functions. This innovative approach is a pivotal component of an expanding open-source computer vision repository dedicated to processing and analyzing extensive galactical datasets. The repository integrates high-performance deep learning algorithms, specifically leveraging the You Only Look Once (YOLO) model and other advanced techniques in image processing and computer vision. The underlying model, trained through deep learning methodologies, exemplifies the intersection of cutting-edge technology and the exploration of the cosmos.
International Journal on Computer Science and Engineering , 2010
Image segmentation is an essential but critical component in low level vision image analysis, pat... more Image segmentation is an essential but critical component in low level vision image analysis, pattern recognition, and in robotic systems. It is one of the most difficult and challenging tasks in image processing which determines the quality of the final result of the image analysis. Image segmentation is the process of dividing an image into different regions such that each region is homogeneous. Various image segmentation algorithms are discussed. Some examples in different image formats are presented and overall results discussed and compared considering different parameters.
Journal of Xi'an University of Architecture & Technology, 2022
The most recent advancements in optics and photonics have produced a Hyperspectral image sensor w... more The most recent advancements in optics and photonics have produced a Hyperspectral image sensor with improved spectral and spatial resolution. To identify the materials and things on the earth's surface, geographical and spectral data are effectively used. The spectral signatures are modeled such that they can distinguish between different objects and materials. The classification of image pixels according to their spectral qualities can be seen as a similar challenge to the identification of various materials, objects, and surface ground cover classes based on their reflectance properties. Numerous fields, including astronomy, environmental research, surveillance, medicinal imaging, and agriculture have exploited classification of Hyperspectral imagery. The identification of items on the surface of the earth is well known to be possible with hyper spectral images. To perform classification and identify the various items on the image, the majorities of classifiers rely on spectral data and do not take spatial variables into account. Using the Spectral Angle Mapper Classification (SAM), Support Vector Machine algorithm, and Neural Net technique, the hyper spectral image is classified in this study based on spectral and spatial attributes. The hyper spectral image is segmented into a few pieces. Each patch's high level spectral and spatial properties are constructed using CNN, and multilayer perception aids in the categorization of picture data into distinct classes. The outcomes of the simulations reveal that SVM archives have the highest classification accuracy.
sian Conference on Innovation in Technology (ASIANCON) , 2021
Novel Coronavirus is an infectious disease, whose outbreak has brought human life to a standstill... more Novel Coronavirus is an infectious disease, whose outbreak has brought human life to a standstill. This virus has surged throughout the planet, affecting oodles of lives. In order to test an infected patient, throat swabs are collected and diagnosed with the help of RT-PCR testing kits, risking the lives of healthcare workers and the people coming in contact with the infected. The brisk upsurge in the quantum of positive cases with new variants emerging worldwide, it has become a serious concern to come up with an alternative detection of the virus. This paper aims to specifically discuss the deep learning techniques and different Convolution Neural Networks using medical modalities such as chest X-rays for the detection of the virus by training the pre-trained models. The different models are compared and different parameters are evaluated.
International Journal of Health Sciences,, 2022
practise since it allows these trainees to completely portray and envisage the existing problems ... more practise since it allows these trainees to completely portray and envisage the existing problems and the repercussions faced in an ideal manner.
International Journal of Soft Computing and Engineering (IJSCE), 2021
Hyperspectral Image (HSI) processing is the new advancement in image / signal processing field. T... more Hyperspectral Image (HSI) processing is the new advancement in image / signal processing field. The growth over the years is appreciable. The main reason behind the successful growth of the Hyperspectral imaging field is due to the enormous amount of spectral and spatial information that the imagery contains. The spectral band that the HSI which contains is also more in number. When an image is captured through the HSI cameras, it contains around 200-250 images of the same scene. Nowadays HSI is used extensively in the fields of environmental monitoring, Crop-Field monitoring, Classification, Identification, Remote sensing applications, Surveillance etc. The spectral and spatial information content present in Hyperspectral images are with high resolutions.Hyperspectral imaging has shown significant growth and widely used in most of the remote sensing applications due to its presence of information of a scene over hundreds of contiguous bands In. Hyperspectral Image Classification of materials is the critical application of HSI using Hyperspectral sensors. It collects hundreds of spectrum channels, where each channel consists of a sharp point of Electromagnetic Spectrum. The paper mainly focuses on Deep Learning techniques such as Convolutional Neural Network (CNN), Artificial Neural Network (ANN), and Support Vector machines (SVM), K-Nearest Neighbour (KNN) for the accuracy in classification. Finally in the summary the current state-of-the-art scheme, a critical discussion after reviewing the research work by other professionals and organizing it into review-based paper, also implying about the present status on classification accuracy using neural networks is carried out.
International Journal of Electrical and Computer Engineering (IJECE), 2022
Compressive sensing is receiving a lot of attention from the image processing research community ... more Compressive sensing is receiving a lot of attention from the image processing research community as a promising technique for image recovery from very few samples. The modality of compressive sensing technique is very useful in the applications where it is not feasible to acquire many samples. It is also prominently useful in satellite imaging applications since it drastically reduces the number of input samples thereby reducing the storage and communication bandwidth required to store and transmit the data into the ground station. In this paper, an interior point-based method is used to recover the entire satellite image from compressive sensing samples. The compression results obtained are compared with the compression results from conventional satellite image compression algorithms. The results demonstrate the increase in reconstruction accuracy as well as higher compression rate in case of compressive sensing-based compression technique.
Design Engineering, 2021
In medical Imaging, one of the most important aspects to be kept in mind is the preprocessing of... more In medical Imaging, one of the most important aspects to be kept in mind is the
preprocessing of the image before it is being deployed on different architectures for
analysis. In recent past, it has been observed that in medical image diagnosis, deep
learning models especially Convolution Networks have been declared as an efficient
technique by most researchers. Since most of the times in biomedical imaging, the choice
of the dataset and the region of interest plays a crucial role in diagnosis of the tumorous
cell in human brain, an attempt has been made here to find the best variant of U-Net deep
learning model for segmentation of the brain image as against the manual and automated
approach of diagnosis of tumor in human brain. This paper investigates the different
variants of the U-Net model before concluding the lighter and the best variant for the
segmentation by taking into consideration fine-tuning of different hyper parameters
involved in decision making. During the experimentation, the best model is selected not
just, because one model outperformed the other in giving better accuracy, instead care
has been ensured in selecting the model that has acceptable higher accuracy performance
while performing fewer computations but giving out faster inference. In this work with
the aid of transfer learning a block wise fine tuning of hyper parameters is carried out to
derive a model with 99.23% of accuracy on BraTs 2019 FLAIR dataset.
International Journal of Engineering Trends and Technology, 2021
In medical imaging, one of the tough tasks is the classification of tumors present in the human b... more In medical imaging, one of the tough tasks is the classification of tumors present in the human brain. The work concentrates on the detection of the exact infected location in the human brain that consists of tumor and provide suitable techniques to administer treatment for the same. To achieve this objective, although there are various deep learning techniques employed by different researchers, an attempt has been made in this work to diagnose the existence of tumors using transfer learning. The experimentation has been carried out on the standard benchmark dataset-BraTs 2018. A hybrid model has been developed in this work for classifying the tumor as benign or malignant. The hybrid model is developed using depth-wise convolutions instead of the traditional approach in combination with the residual block being introduced in the final layer into the modified U-Net model deployed using a pre-trained VGG-16 model for classification. This hybrid model was then fine-tuned by varying certain vital hyperparameters to obtain an accuracy of about 92.30%.
Journal Of Critical Reviews, 2021
Pneumonia has been an illness which affects millions of people worldwide. With the pandemic Covid... more Pneumonia has been an illness which affects millions of people worldwide. With the pandemic Covid around worldwide, pneumonia has been one of the symptoms which affects humans and can lead to death if left untreated. There is a need to diagnose Pneumonia at its very early stage so that person can be treated without many complications. Chest X rays are the cheapest and easiest method to detect Pneumonia.Pneumonia detection using machine learning can help radiologistto diagnose it and thus avoid misdiagnosis. This paper gives an insight to the various techniques that are used to detect Pneumonia from chest X rays.
HTL Journal, 2022
Hyperspectral Imaging is an essential technique to deep explore surfaces, which provides more det... more Hyperspectral Imaging is an essential technique to deep explore surfaces, which provides more detailed information than the single point spectroscopy. Over the past decade there has been many devices which were invented for the Hyperspectral Image (HSI) acquisition .The complexity lies in the image dataset dimension, as the HSI data sets area difficult to manage due to its large dimensionality. Along with this the datasets contains problems related to calibration, the calibration with respect to geometrical correction, radiometric and atmospheric correction has to be performed to the data acquired. The preprocessing steps have to be carefully performed on the acquired Hyperspectral data as to carry forward the further processes of Image classification [1] , segmentation etc. Various pre-processing techniques are made use in the literature surveys. This paper focuses on the prominent pre-processing techniques available. The atmospheric correction using FLAASH carried out showed better corrected results.
International Journal of Creative Research Thoughts (IJCRT), 2023
Machine learning (ML) is a subset of artificial intelligence (AI) where computer is trained to ma... more Machine learning (ML) is a subset of artificial intelligence (AI) where computer is trained to make decisions like a human based on different characteristics from a set of data. Machine Learning is a subset of artificial intelligence (AI) which allows the software applications to predict the outcomes by considering the historical data as input. The paper presented discusses the detailed understanding and the contribution of different researchers in the area of Biomedical Signal processing.
International Journal of Creative Research Thoughts (IJCRT), 2023
Digital Image processing is an algorithm used to perform operations on a digital image, in order ... more Digital Image processing is an algorithm used to perform operations on a digital image, in order to extract some useful information or process images to enhance information from it. Face recognition is a method of identifying or verifying the identity of an individual using their face. Face recognition systems can be used to identify people in photos, video in real-time or offline. Machine learning algorithms can be used to identify, collect, store and evaluate face characteristics so that they can be matched to photos of people in a database. The paper presents the contributions made by different researchers to understand the methodology, tools and applications of face recognition.
International Journal of Creative Research Thoughts (IJCRT), 2023
Speech recognition is the process of converting human sound signals into words or instructions. T... more Speech recognition is the process of converting human sound signals into words or instructions. The research of speech recognition involves many subject areas such as computer technology, artificial intelligence, digital signal processing, pattern recognition, acoustics, linguistics, and cognitive science.Our speech is made up of many frequencies at the same time. The actual signal is really a sum of all those frequencies stuck together. The conversation or speech that is captured by a microphone or a telephone is converted from acoustic signal to a set of words in speech recognition.
International Journal of Creative Research Thoughts (IJCRT), 2023
Perfect weather predictions are needed for a plethora of activities and it was one of the main ch... more Perfect weather predictions are needed for a plethora of activities and it was one of the main challenging problems the entire world faced because of its multiple dimensions and non-linear trends. Weather depends on multiple climatic conditions like temperature, air pressure, humidity, wind flow speed and direction, cloud height and density, and rainfall. The most common phrases/keywords in weather prediction related articles were 'wind', 'precipitation', 'climate change', 'wind forecasting' and 'ensemble prediction'. The most common countries in which surveys were done are China, USA, Australia, India and Germany. Extreme meteorological events are often related to the occurrence of weather fronts.
International Journal of Creative Research Thoughts (IJCRT), 2023
In machine learning, classification is a supervised learning concept which basically categorizes ... more In machine learning, classification is a supervised learning concept which basically categorizes a set of data into classes. The most common classification problems arespeech recognition, face detection, handwriting recognition, and image classification. Image Classification is one of the most fundamental tasks in computer vision. It has revolutionized and propelled technological advancements in the most prominent fields, including the automobile industry, healthcare, manufacturing, and more. Image Classification (often referred to as Image Recognition) is the task of associating one (single-label classification) or more (multilabel classification) labels to a given image. Image classification is the process of predicting a specific class, or label, for something that is defined by a set of data points. Image classification is a subset of the classification problem, where an entire image is assigned a label. Grouping images into semantically meaningful categories using low-level visual features is a challenging and important problem in contentbased image retrieval. Based on these groupings, effective indices can be built for an image database
International Journal of Creative Research Thoughts (IJCRT), 2023
Stock market prediction and analysis are some of the most difficult jobs to complete. There are n... more Stock market prediction and analysis are some of the most difficult jobs to complete. There are numerous causes for this, including market volatility and a variety of other dependent and independent variables that influence the value of a certain stock in the market. These variables make it extremely difficult for any stock market expert to anticipate the rise and fall of the market with great precision. However, with the introduction of Machine Learning and its strong algorithms, the most recent market research and Stock Market Prediction advancements have begun to include such approaches in analysing stock market data. The Opening Value of the stock, the Highest and Lowest values of that stock on the same days, as well as the Closing Value at the end of the day, are all indicated for each date. Furthermore, the total volume of the stocks in the market is provided, with this information, it is up to the job of a Machine Learning Data Scientist to look at the data and develop different algorithms that may help in finding appropriate stocks values.
International Journal of Creative Research Thoughts (IJCRT), 2023
Many deep learning research works have shown successful results, primarily focusing on three data... more Many deep learning research works have shown successful results, primarily focusing on three data types: images, speech, and text. In addition, deep learning has also been successfully applied to communication signals/packets. Widely used applications of these kind of data are image classification, speech recognition, regression problem, pattern recognition, and text sentiment classification. The most fascinating of all is video data. Video data is also interesting for research from the perspective of its big size and dimension. Millions of video data are uploaded every day on YouTube; thus, it becomes a rich repository and empowered artificial intelligence (AI) research
International Journal of Creative Research Thoughts (IJCRT), 2023
Many deep learning research works have shown successful results, primarily focusing on three data... more Many deep learning research works have shown successful results, primarily focusing on three data types: images, speech, and text. In addition, deep learning has also been successfully applied to communication signals/packets. Widely used applications of these kind of data are image classification, speech recognition, regression problem, pattern recognition, and text sentiment classification. The most fascinating of all is video data. Video data is also interesting for research from the perspective of its big size and dimension. Millions of video data are uploaded every day on YouTube; thus, it becomes a rich repository and empowered artificial intelligence (AI) research
International Journal of Creative Research Thoughts (IJCRT), 2023
The goal of this proposed work is to create an Android-based image text-to-speech (ITTS) applicat... more The goal of this proposed work is to create an Android-based image text-to-speech (ITTS) application that enables users to translate text in photographs into spoken in formation in the language of their choice. The ability for users to customize the language in which the synthesized voice is produced is one of the application's standout features. Because of its user-friendly interface, a wide audience can access the Android application. Performance of an Android application is evaluated by precision, reactivity, and ability to customize language. This proposed workcan serve a variety of user demands, such as language learners, visually impaired people, and people looking for portable, effective tools for information consumption.
International Journal of Creative Research Thoughts (IJCRT), 2023
The human visual system effortlessly recognizes and identifies objects within its field of view, ... more The human visual system effortlessly recognizes and identifies objects within its field of view, showcasing remarkable speed and accuracy in tasks such as object detection and identification. Valuable clues regarding the origin and evolution of the universe are intricately woven into the shapes and formations of galaxies. However, automating the classification of galaxies from their images poses challenges due to the faintness of the galaxy images, interference from bright background stars, and inherent image noise. To address these complexities, we propose a method for the automatic detection and classification of galaxies. This method incorporates a data augmentation procedure, enhancing the robustness of trained models against data variations arising from different instruments and contrast-stretching functions. This innovative approach is a pivotal component of an expanding open-source computer vision repository dedicated to processing and analyzing extensive galactical datasets. The repository integrates high-performance deep learning algorithms, specifically leveraging the You Only Look Once (YOLO) model and other advanced techniques in image processing and computer vision. The underlying model, trained through deep learning methodologies, exemplifies the intersection of cutting-edge technology and the exploration of the cosmos.
International Journal on Computer Science and Engineering , 2010
Image segmentation is an essential but critical component in low level vision image analysis, pat... more Image segmentation is an essential but critical component in low level vision image analysis, pattern recognition, and in robotic systems. It is one of the most difficult and challenging tasks in image processing which determines the quality of the final result of the image analysis. Image segmentation is the process of dividing an image into different regions such that each region is homogeneous. Various image segmentation algorithms are discussed. Some examples in different image formats are presented and overall results discussed and compared considering different parameters.
Journal of Xi'an University of Architecture & Technology, 2022
The most recent advancements in optics and photonics have produced a Hyperspectral image sensor w... more The most recent advancements in optics and photonics have produced a Hyperspectral image sensor with improved spectral and spatial resolution. To identify the materials and things on the earth's surface, geographical and spectral data are effectively used. The spectral signatures are modeled such that they can distinguish between different objects and materials. The classification of image pixels according to their spectral qualities can be seen as a similar challenge to the identification of various materials, objects, and surface ground cover classes based on their reflectance properties. Numerous fields, including astronomy, environmental research, surveillance, medicinal imaging, and agriculture have exploited classification of Hyperspectral imagery. The identification of items on the surface of the earth is well known to be possible with hyper spectral images. To perform classification and identify the various items on the image, the majorities of classifiers rely on spectral data and do not take spatial variables into account. Using the Spectral Angle Mapper Classification (SAM), Support Vector Machine algorithm, and Neural Net technique, the hyper spectral image is classified in this study based on spectral and spatial attributes. The hyper spectral image is segmented into a few pieces. Each patch's high level spectral and spatial properties are constructed using CNN, and multilayer perception aids in the categorization of picture data into distinct classes. The outcomes of the simulations reveal that SVM archives have the highest classification accuracy.
sian Conference on Innovation in Technology (ASIANCON) , 2021
Novel Coronavirus is an infectious disease, whose outbreak has brought human life to a standstill... more Novel Coronavirus is an infectious disease, whose outbreak has brought human life to a standstill. This virus has surged throughout the planet, affecting oodles of lives. In order to test an infected patient, throat swabs are collected and diagnosed with the help of RT-PCR testing kits, risking the lives of healthcare workers and the people coming in contact with the infected. The brisk upsurge in the quantum of positive cases with new variants emerging worldwide, it has become a serious concern to come up with an alternative detection of the virus. This paper aims to specifically discuss the deep learning techniques and different Convolution Neural Networks using medical modalities such as chest X-rays for the detection of the virus by training the pre-trained models. The different models are compared and different parameters are evaluated.
International Journal of Health Sciences,, 2022
practise since it allows these trainees to completely portray and envisage the existing problems ... more practise since it allows these trainees to completely portray and envisage the existing problems and the repercussions faced in an ideal manner.
International Journal of Soft Computing and Engineering (IJSCE), 2021
Hyperspectral Image (HSI) processing is the new advancement in image / signal processing field. T... more Hyperspectral Image (HSI) processing is the new advancement in image / signal processing field. The growth over the years is appreciable. The main reason behind the successful growth of the Hyperspectral imaging field is due to the enormous amount of spectral and spatial information that the imagery contains. The spectral band that the HSI which contains is also more in number. When an image is captured through the HSI cameras, it contains around 200-250 images of the same scene. Nowadays HSI is used extensively in the fields of environmental monitoring, Crop-Field monitoring, Classification, Identification, Remote sensing applications, Surveillance etc. The spectral and spatial information content present in Hyperspectral images are with high resolutions.Hyperspectral imaging has shown significant growth and widely used in most of the remote sensing applications due to its presence of information of a scene over hundreds of contiguous bands In. Hyperspectral Image Classification of materials is the critical application of HSI using Hyperspectral sensors. It collects hundreds of spectrum channels, where each channel consists of a sharp point of Electromagnetic Spectrum. The paper mainly focuses on Deep Learning techniques such as Convolutional Neural Network (CNN), Artificial Neural Network (ANN), and Support Vector machines (SVM), K-Nearest Neighbour (KNN) for the accuracy in classification. Finally in the summary the current state-of-the-art scheme, a critical discussion after reviewing the research work by other professionals and organizing it into review-based paper, also implying about the present status on classification accuracy using neural networks is carried out.
International Journal of Electrical and Computer Engineering (IJECE), 2022
Compressive sensing is receiving a lot of attention from the image processing research community ... more Compressive sensing is receiving a lot of attention from the image processing research community as a promising technique for image recovery from very few samples. The modality of compressive sensing technique is very useful in the applications where it is not feasible to acquire many samples. It is also prominently useful in satellite imaging applications since it drastically reduces the number of input samples thereby reducing the storage and communication bandwidth required to store and transmit the data into the ground station. In this paper, an interior point-based method is used to recover the entire satellite image from compressive sensing samples. The compression results obtained are compared with the compression results from conventional satellite image compression algorithms. The results demonstrate the increase in reconstruction accuracy as well as higher compression rate in case of compressive sensing-based compression technique.
Design Engineering, 2021
In medical Imaging, one of the most important aspects to be kept in mind is the preprocessing of... more In medical Imaging, one of the most important aspects to be kept in mind is the
preprocessing of the image before it is being deployed on different architectures for
analysis. In recent past, it has been observed that in medical image diagnosis, deep
learning models especially Convolution Networks have been declared as an efficient
technique by most researchers. Since most of the times in biomedical imaging, the choice
of the dataset and the region of interest plays a crucial role in diagnosis of the tumorous
cell in human brain, an attempt has been made here to find the best variant of U-Net deep
learning model for segmentation of the brain image as against the manual and automated
approach of diagnosis of tumor in human brain. This paper investigates the different
variants of the U-Net model before concluding the lighter and the best variant for the
segmentation by taking into consideration fine-tuning of different hyper parameters
involved in decision making. During the experimentation, the best model is selected not
just, because one model outperformed the other in giving better accuracy, instead care
has been ensured in selecting the model that has acceptable higher accuracy performance
while performing fewer computations but giving out faster inference. In this work with
the aid of transfer learning a block wise fine tuning of hyper parameters is carried out to
derive a model with 99.23% of accuracy on BraTs 2019 FLAIR dataset.
International Journal of Engineering Trends and Technology, 2021
In medical imaging, one of the tough tasks is the classification of tumors present in the human b... more In medical imaging, one of the tough tasks is the classification of tumors present in the human brain. The work concentrates on the detection of the exact infected location in the human brain that consists of tumor and provide suitable techniques to administer treatment for the same. To achieve this objective, although there are various deep learning techniques employed by different researchers, an attempt has been made in this work to diagnose the existence of tumors using transfer learning. The experimentation has been carried out on the standard benchmark dataset-BraTs 2018. A hybrid model has been developed in this work for classifying the tumor as benign or malignant. The hybrid model is developed using depth-wise convolutions instead of the traditional approach in combination with the residual block being introduced in the final layer into the modified U-Net model deployed using a pre-trained VGG-16 model for classification. This hybrid model was then fine-tuned by varying certain vital hyperparameters to obtain an accuracy of about 92.30%.
Journal Of Critical Reviews, 2021
Pneumonia has been an illness which affects millions of people worldwide. With the pandemic Covid... more Pneumonia has been an illness which affects millions of people worldwide. With the pandemic Covid around worldwide, pneumonia has been one of the symptoms which affects humans and can lead to death if left untreated. There is a need to diagnose Pneumonia at its very early stage so that person can be treated without many complications. Chest X rays are the cheapest and easiest method to detect Pneumonia.Pneumonia detection using machine learning can help radiologistto diagnose it and thus avoid misdiagnosis. This paper gives an insight to the various techniques that are used to detect Pneumonia from chest X rays.
HTL Journal, 2022
Hyperspectral Imaging is an essential technique to deep explore surfaces, which provides more det... more Hyperspectral Imaging is an essential technique to deep explore surfaces, which provides more detailed information than the single point spectroscopy. Over the past decade there has been many devices which were invented for the Hyperspectral Image (HSI) acquisition .The complexity lies in the image dataset dimension, as the HSI data sets area difficult to manage due to its large dimensionality. Along with this the datasets contains problems related to calibration, the calibration with respect to geometrical correction, radiometric and atmospheric correction has to be performed to the data acquired. The preprocessing steps have to be carefully performed on the acquired Hyperspectral data as to carry forward the further processes of Image classification [1] , segmentation etc. Various pre-processing techniques are made use in the literature surveys. This paper focuses on the prominent pre-processing techniques available. The atmospheric correction using FLAASH carried out showed better corrected results.
[
Abstract— Image Compression plays a crucial role in multimedia revolution. This paper proposes em... more Abstract— Image Compression plays a crucial role in multimedia revolution. This paper proposes emerging hybrid algorithm approach for Image Compression, it used Haar Transform and the Singular Value Decomposition. Firstly, Haar Transform is applied on the Image with appropriate threshold, the DC coefficient where the whole energy is packed is stored separately. To the truncated matrix obtained, SVD is applied, the matrices obtained after applying SVD are again truncated with appropriate rank. The threshold and rank are varied accordingly to analyze various prospects of compression maintaining PSNR of 30 dB which ensures visual perception. The inverse process is performed to reconstruct the image, hence exploring various performance parameters of hybrid image compression.