Automated evaluation metrics for video captioning hold an important place in computer vision. Alt... more Automated evaluation metrics for video captioning hold an important place in computer vision. Although some research has looked up in this direction, they produce subpar results or highlight the significance of a dataset with a certain domain. So far, evaluation metrics used in machine translation/ image captioning have been adapted for video captioning. At the same time, they don't give the desired result. This work intends to improve the performances of existing methods by proposing a new evaluation metric for the video captioning models. The proposed metric, Video-Captioning Evaluation Metric for Segments (VEMS), accounts for the main event, sub-actions, and background details. Thus VEMS evaluates captions at the segment level. A novel dataset structure called MSVD-S, a modified version of the MSVD dataset consisting of captions for multiple segments in a single video has also been proposed. Based on the MSVD-S dataset with weighted frames, VEMS captions to video in a frame-wise manner. Detailed performance analysis of the corner cases like the omission of actions, ignorance of sub-events, and lack of details has been done. The proposed metric provided improved performance as an evaluation score closer to an accuracy of 34.2% has been achieved.
International Journal of Network Security, Jul 1, 2020
In this paper, we present a novel (2, 2) reversible secret image sharing scheme. Our scheme permi... more In this paper, we present a novel (2, 2) reversible secret image sharing scheme. Our scheme permits secret messages to be shared with two participants by splitting the marked encrypted image into two shadows. The secret messages can be reconstructed if two participants collaborate with each other. The proposed scheme chooses suitable binary blocks of a cover image in which to embed the secret message and divides those blocks into two shadow blocks by executing a logical operation with all of the other binary blocks, thereby producing two shadows. In the data extraction procedure, the secret messages and the cover image can be reconstructed by the logical operation of the corresponding binary blocks of the two shadows. A practical application is demonstrated by modeling our scheme as a reversible watermarking scheme in the Cloud. The experimental results indicated that the proposed method is reversible and that it can restore the image and watermark properly.
Pedestrian crossings have also been highlighted as one of the most dangerous locations in the tra... more Pedestrian crossings have also been highlighted as one of the most dangerous locations in the transportation field. Because people and vehicles share the road, a crosswalk improves the road’s efficiency in a densely populated region. However, as the population grows, more accidents and serious injuries occur, and as a result, nationalities are attempting to reduce these incidents through marketing and legal fines. Various architectures and developmental models have been proposed by authors focusing on the safeguarding of pedestrians crossing the intersections and vehicles passing by. Few proposed machine learning and deep learning-based solutions to the pedestrian lanes; others provided an Internet of Things- (IoT-) based solution to the situation. Various challenges are left unresolved, such as evidence recording, image capturing, and recognition in case of an emergency. In the proposed scenario, an IoT-based technology is utilized to assist the vehicles passing by to act over the ...
Basically, it is hard for endeavors to recognize plant leaf images by a layman due to the varieti... more Basically, it is hard for endeavors to recognize plant leaf images by a layman due to the varieties in some plant leaves and the extensive information collected for investigation. It is hard to build an automated recognition framework that can handle massive data and give an intermediate analysis. Image examination and order and pattern recognition are some issues that are effectively connected to the existing methods. This paper focuses on designing an automated plant recognition system based on the best recognition algorithm and the Google platform to locate all plant locations on a map. A case study of India, which has huge biodiversity, is illustrated. The proposed system can show the detailed location of that particular species, where they can be found, and the shortest distance from the current location.
Widespread fear and panic has emerged about COVID-19 on social media platforms which are often su... more Widespread fear and panic has emerged about COVID-19 on social media platforms which are often supported by falsified and altered content. This mass hysteria creates public anxiety due to misinformation, misunderstandings, and ignorance of the impact of COVID-19. To assist health professionals in addressing this epidemic more appropriately at the onset, sentiment analysis can potentially help the authorities for devising appropriate strategies. This study analyzes tweets related to COVID-19 using a machine learning approach and offers a high-accuracy solution. Experiments are performed involving different machine and deep learning models along with various features such as Word2vec, term-frequency, term-frequency document frequency, and feature fusion of both feature-generating approaches. The proposed approach combines the extra tree classifier and convolutional neural network and uses feature fusion to achieve the highest accuracy score of 99%. The proposed approach obtains far be...
In the paper, the authors investigated and predicted the future environmental circumstances of a ... more In the paper, the authors investigated and predicted the future environmental circumstances of a COVID-19 to minimize its effects using artificial intelligence techniques. The experimental investigation of COVID-19 instances has been performed in ten countries, including India,
Early and precise detection of diabetic retinopathy prevents vision impairments through computer-... more Early and precise detection of diabetic retinopathy prevents vision impairments through computer-aided clinical procedures. Identifying the symptoms and processing those by using sophisticated clinical procedures reduces hemorrhage kind of risks. The input diabetic retinopathy images are influenced by using computer vision-based processes for segmentation and classification through feature extractions. In this article, a delimiting segmentation using knowledge learning (DS-KL) is introduced for classifying and detecting exudate regions by using varying histograms. The input image is identified for its histogram changes from the feature-dependent segmentation process. Depending on the training knowledge from multiple inputs with different exudate regions, the segmentation is performed. This segmentation identifies infected and noninfected regions across the delimiting pixel boundaries. The knowledge-learning process stores the newly identified exudate region for training and pixel co...
Journal of Ambient Intelligence and Humanized Computing
protective steps were put in place right away to stop the virus from spreading across the air as ... more protective steps were put in place right away to stop the virus from spreading across the air as presented in Leung et al. (2020). Citizens around the world were urged to take 1 Introduction When the SARS-CoV-2 pandemic broke out last 2019,
Technological advancements have made it possible to monitor, diagnose, and treat patients remotel... more Technological advancements have made it possible to monitor, diagnose, and treat patients remotely. The vital signs of patients can now be collected with the help of Internet of Things (IoT)-based wearable sensor devices and then uploaded on to a fog server for processing and access by physicians for recommending prescriptions and treating patients through the Internet of Medical Things (IoMT) devices. This research presents the outcome of a survey conducted on healthcare integrated with fog computing and IoT to help researchers understand the techniques, technologies and performance parameters. A comparison of existing research focusing on technologies, procedures, and findings has been presented to investigate several aspects of fog computing in healthcare IoT-based systems, such as increased temporal complexity, storage capacity, scalability, bandwidth, and latency. Additionally, strategies, tools, and sensors used in various diseases such as heart disease, chronic disease, chiku...
Over the last few years, different contaminants have posed a danger to the quality of the water. ... more Over the last few years, different contaminants have posed a danger to the quality of the water. Hence modelling and forecasting water quality are very important in the management of water contamination. The paper proposes an ensemble machine learning-based model for assessing water quality. The results of the proposed model are compared with several machine learning models, including k-nearest neighbour, NaĂ¯ve Bayes, support vector machine, and decision tree. The considered dataset contains seven statistically important parameters: pH, conductivity, dissolved oxygen, Biochemical Oxygen Demand, nitrate, total coliform, and fecal coliform. The water quality index is calculated for assessing water quality. To utilize an ensemble approach, a voting classifier has been designed with hard voting. The highest prediction accuracy of 99.5% of the water quality index is presented by the voting classifier as compared to the prediction accuracy of 99.2%, 90%, 79%, and 99% presented through k-n...
This research proposed an improved transfer-learning bird classification framework to achieve a m... more This research proposed an improved transfer-learning bird classification framework to achieve a more precise classification of Protected Indonesia Birds (PIB) which have been identified as the endangered bird species. The framework takes advantage of using the proposed sequence of Batch Normalization Dropout Fully-Connected (BNDFC) layers to enhance the baseline model of transfer learning. The main contribution of this work is the proposed sequence of BNDFC that can be applied to any Convolutional Neural Network (CNN) based model to improve the classification accuracy, especially for image-based species classification problems. The experiment results show that the proposed sequence of BNDFC layers outperform other combination of BNDFC. The addition of BNDFC can improve the model's performance across ten different CNN-based models. On average, BNDFC can improve by approximately 19.88% in Accuracy, 24.43% in F-measure, 17.93% in G-mean, 23.41% in Sensitivity, and 18.76% in Precision. Moreover, applying fine-tuning (FT) is able to enhance the accuracy by 0.85% with a smaller validation loss of 18.33% improvement. In addition, MobileNetV2 was observed to be the best baseline model with the lightest size of 35.9 MB and the highest accuracy of 88.07% in the validation set.
The forecasting of bus passenger flow is important to the bus transit system's operation. Because... more The forecasting of bus passenger flow is important to the bus transit system's operation. Because of the complicated structure of the bus operation system, it's difficult to explain how passengers travel along different routes. Due to the huge number of passengers at the bus stop, bus delays, and irregularity, people are experiencing difficulties of using buses nowadays. It is important to determine the passenger flow in each station, and the transportation department may utilize this information to schedule buses for each region. In Our proposed system we are using an approach called the deep learning method with long short-term memory, recurrent neural network, and greedy layer-wise algorithm are used to predict the Karnataka State Road Transport Corporation (KSRTC) passenger flow. In the dataset, some of the parameters are considered for prediction are bus id, bus type, source, destination, passenger count, slot number, and revenue These parameters are processed in a greedy layer-wise algorithm to make it has cluster data into regions after cluster data move to the long short-term memory model to remove redundant data in the obtained data and recurrent neural network it gives the prediction result based on the iteration factors of the data. These algorithms are more accurate in predicting bus passengers. This technique handles the problem of passenger flow forecasting in Karnataka State Road Transport Corporation Bus Rapid Transit (KSRTCBRT) transportation, and the framework provides resource planning and revenue estimation predictions for the KSRTCBRT.
In a network architecture, an intrusion detection system (IDS) is one of the most commonly used a... more In a network architecture, an intrusion detection system (IDS) is one of the most commonly used approaches to secure the integrity and availability of critical assets in protected systems. Many existing network intrusion detection systems (NIDS) utilize stand-alone classifier models to classify network traffic as an attack or as normal. Due to the vast data volume, these stand-alone models struggle to reach higher intrusion detection rates with low false alarm rates( FAR). Additionally, irrelevant features in datasets can also increase the running time required to develop a model. However, data can be reduced effectively to an optimal feature set without information loss by employing a dimensionality reduction method, which a classification model then uses for accurate predictions of the various network intrusions. In this study, we propose a novel feature-driven intrusion detection system, namely χ2-BidLSTM, that integrates a χ2 statistical model and bidirectional long short-term m...
A smart grid is a promising use-case of AIoT (AI (artificial intelligence) across IoT (internet o... more A smart grid is a promising use-case of AIoT (AI (artificial intelligence) across IoT (internet of things)) that enables bidirectional communication among utilities that arises with demand response (DR) schemes for demand-side management (DSM) and consumers that manage their power demands according to received DR signals. Disaggregating composite electric energy consumption data from a single minimal set of plug-panel current and voltage sensors installed at the electric panel in a practical field of interest, nonintrusive appliance load monitoring (NIALM), a cost-effective load disaggregation approach for (residential) DSM, is able to discern individual electrical appliances concerned without accessing each of them by individual plug-load power meters (smart plugs) deployed intrusively. The most common load disaggregation approaches are based on machine learning algorithms such as artificial neural networks, while approaches based on evolutionary computing, metaheuristic algorithms...
Multidimensional Systems and Signal Processing, 2021
This paper is mainly aimed at the decomposition of image quality assessment study by using Three ... more This paper is mainly aimed at the decomposition of image quality assessment study by using Three Parameter Logistic Mixture Model and k-means clustering (TPLMM-k). This method is mainly used for the analysis of various images which were related to several real time applications and for medical disease detection and diagnosis with the help of the digital images which were generated by digital microscopic camera. Several algorithms and distribution models had been developed and proposed for the segmentation of the images. Among several methods developed and proposed, the Gaussian Mixture Model (GMM) was one of the highly used models. One can say that almost the GMM was playing the key role in most of the image segmentation research works so far noticed in the literature. The main drawback with the distribution model was that this GMM model will be best fitted with a kind of data in the dataset. To overcome this problem, the TPLMM-k algorithm is proposed. The image decomposition process used in the proposed algorithm had been analyzed and its performance was analyzed with the help of various performance metrics like the Variance of Information (VOI), Global Consistency Error (GCE) and Probabilistic
Lung nodules are abnormal growths and lesions may exist. Both lungs may have nodules. Most lung n... more Lung nodules are abnormal growths and lesions may exist. Both lungs may have nodules. Most lung nodules are harmless (not cancerous/malignant). Pulmonary nodules are rare in lung cancer. X-rays and CT scans identify the lung nodules. Doctors may term the growth a lung spot, coin lesion, or shadow. It is necessary to obtain properly computed tomography (CT) scans of the lungs to get an accurate diagnosis and a good estimate of the severity of lung cancer. This study aims to design and evaluate a deep learning (DL) algorithm for identifying pulmonary nodules (PNs) using the LUNA-16 dataset and examine the prevalence of PNs using DB-Net. The paper states that a new, resource-efficient deep learning architecture is called for, and it has been given the name of DB-NET. When a physician orders a CT scan, they need to employ an accurate and efficient lung nodule segmentation method because they need to detect lung cancer at an early stage. However, segmentation of lung nodules is a difficult task because of the nodules' characteristics on the CT image as well as the nodules' concealed shape, visual quality, and context. The DB-NET model architecture is presented as a resource-efficient deep learning solution for handling the challenge at hand in this paper. Furthermore, it incorporates the Mish nonlinearity function and the mask class weights to improve segmentation effectiveness. In addition to the LUNA-16 dataset, which contained 1200 lung nodules collected during the LUNA-16 test, the LUNA-16 dataset was extensively used to train and assess the proposed model. The DB-NET architecture surpasses the existing U-NET model by a dice coefficient index of 88.89%, and it also achieves a similar level of accuracy to that of human experts.
Future wireless networks promise immense increases on data rate and energy efficiency while overc... more Future wireless networks promise immense increases on data rate and energy efficiency while overcoming the difficulties of charging the wireless stations or devices in the Internet of Things (IoT) with the capability of simultaneous wireless information and power transfer (SWIPT). For such networks, jointly optimizing beamforming, power control, and energy harvesting to enhance the communication performance from the base stations (BSs) (or access points (APs)) to the mobile nodes (MNs) served would be a real challenge. In this work, we formulate the joint optimization as a mixed integer nonlinear programming (MINLP) problem, which can be also realized as a complex multiple resource allocation (MRA) optimization problem subject to different allocation constraints. By means of deep reinforcement learning to estimate future rewards of actions based on the reported information from the users served by the networks, we introduce single-layer MRA algorithms based on deep Q-learning (DQN) ...
The aim of 5G wireless networks to provide Mbps and Gbps data rates to end users is expected to b... more The aim of 5G wireless networks to provide Mbps and Gbps data rates to end users is expected to be fulfilled by the advanced technologies such as multi-input multi-output (MIMO), carrier aggregation (CA), inter/intra-cell communication, and adaptive modulation and coding techniques, which would be all realized in the Long Term Evolution-Advanced (LTE-A) heterogeneous network constituted by macrocells (MCs) and small cells (SCs) adopting these 5G advanced techniques. Given the potential of significantly increasing the network performance, the resource allocation (RA) problem involved becomes harder than ever especially when MIMO and CA are included in the RA problem involving multiple types of resources to be concurrently determined for the global optimization. Facing this challenge, we develop a framework to jointly optimize energy efficiency (EE), spectrum efficiency (SE), and queue length for downlink transmissions with an overall and comprehensive consideration of dynamically allocating resource blocks (RBs), component carriers (CCs), modulation and coding schemes (MCSs), and deciding user association (UA) with a power control (PC) mechanism on discrete power levels (PLs) in the heterogeneous LTE-based MIMO wireless networks. Specially, for the complex joint RA, UA, and PC problem, we conduct a mixed integer programming model to accommodate the stochastic optimization problem involved with the drift-plus-penalty (DPP) approach for Lyapunov opportunistic optimization. In particular, although it involves a nondeterministic polynomial time (NP) problem, we can still show a reduced problem to be solved easily through linear relaxation when its coefficient matrix is totally unimodular (TUM), and to be solved efficiently as well even when the TUM property is not guaranteed. Based on the reduction, we further develop a distributed or semi-distributed algorithm operated on two levels to approach the optimal results with lower complexity if the UA requirement can be relaxed. Finally, apart from exhibiting its performance on the weighting parameters, the numerical experiments also show our approach to make a good tradeoff among SE, EE, and queue length, and outperform the greedy-based state-of-the-art algorithms. INDEX TERMS LTE-A heterogeneous wireless networks, MIMO, carrier aggregation, multi-resource allocation, user association, power control.
Security with Intelligent Computing and Big-data Services, 2018
In this paper, we proposed a joint lossless index coding and data hiding technique for the palett... more In this paper, we proposed a joint lossless index coding and data hiding technique for the palette images. The palette image is the compressed image of the color image quantization technique. The compressed codes of the palette image consist of the index table and the color palette. In the proposed technique, a three-category lossless index coding method is employed. The secret data is embedded into the encoded index table during the index coding process is executed. From the results, it is shown that good hiding capacity is obtained in the proposed technique while keeping a good bit rate.
With the rapid development of computing technologies, human-centered computing (HCC) has become a... more With the rapid development of computing technologies, human-centered computing (HCC) has become an emerging multidisciplinary area which integrates ubiquitous computing, wearable computing, and so on. When large amounts of information has to be mutually analyzed, HCC aims to bridge the gap between people and computing systems. However, ensuring a secure transmission becomes challenging in HCC because malicious hackers tend to tamper with the Internet data. The massive amount of counterfeited or tampered data on the Internet damage data trustworthiness and impede the progress of HCC. To address this problem, this study presented an image authentication method based on the residual histogram shifting technique. In this technique, an image histogram is generated using block-based processing. Then, the histogram is modified to achieve a high embedding capacity. By manipulating the image histogram, secret or authentication codes can be embedded in the image itself to prevent private info...
Automated evaluation metrics for video captioning hold an important place in computer vision. Alt... more Automated evaluation metrics for video captioning hold an important place in computer vision. Although some research has looked up in this direction, they produce subpar results or highlight the significance of a dataset with a certain domain. So far, evaluation metrics used in machine translation/ image captioning have been adapted for video captioning. At the same time, they don't give the desired result. This work intends to improve the performances of existing methods by proposing a new evaluation metric for the video captioning models. The proposed metric, Video-Captioning Evaluation Metric for Segments (VEMS), accounts for the main event, sub-actions, and background details. Thus VEMS evaluates captions at the segment level. A novel dataset structure called MSVD-S, a modified version of the MSVD dataset consisting of captions for multiple segments in a single video has also been proposed. Based on the MSVD-S dataset with weighted frames, VEMS captions to video in a frame-wise manner. Detailed performance analysis of the corner cases like the omission of actions, ignorance of sub-events, and lack of details has been done. The proposed metric provided improved performance as an evaluation score closer to an accuracy of 34.2% has been achieved.
International Journal of Network Security, Jul 1, 2020
In this paper, we present a novel (2, 2) reversible secret image sharing scheme. Our scheme permi... more In this paper, we present a novel (2, 2) reversible secret image sharing scheme. Our scheme permits secret messages to be shared with two participants by splitting the marked encrypted image into two shadows. The secret messages can be reconstructed if two participants collaborate with each other. The proposed scheme chooses suitable binary blocks of a cover image in which to embed the secret message and divides those blocks into two shadow blocks by executing a logical operation with all of the other binary blocks, thereby producing two shadows. In the data extraction procedure, the secret messages and the cover image can be reconstructed by the logical operation of the corresponding binary blocks of the two shadows. A practical application is demonstrated by modeling our scheme as a reversible watermarking scheme in the Cloud. The experimental results indicated that the proposed method is reversible and that it can restore the image and watermark properly.
Pedestrian crossings have also been highlighted as one of the most dangerous locations in the tra... more Pedestrian crossings have also been highlighted as one of the most dangerous locations in the transportation field. Because people and vehicles share the road, a crosswalk improves the road’s efficiency in a densely populated region. However, as the population grows, more accidents and serious injuries occur, and as a result, nationalities are attempting to reduce these incidents through marketing and legal fines. Various architectures and developmental models have been proposed by authors focusing on the safeguarding of pedestrians crossing the intersections and vehicles passing by. Few proposed machine learning and deep learning-based solutions to the pedestrian lanes; others provided an Internet of Things- (IoT-) based solution to the situation. Various challenges are left unresolved, such as evidence recording, image capturing, and recognition in case of an emergency. In the proposed scenario, an IoT-based technology is utilized to assist the vehicles passing by to act over the ...
Basically, it is hard for endeavors to recognize plant leaf images by a layman due to the varieti... more Basically, it is hard for endeavors to recognize plant leaf images by a layman due to the varieties in some plant leaves and the extensive information collected for investigation. It is hard to build an automated recognition framework that can handle massive data and give an intermediate analysis. Image examination and order and pattern recognition are some issues that are effectively connected to the existing methods. This paper focuses on designing an automated plant recognition system based on the best recognition algorithm and the Google platform to locate all plant locations on a map. A case study of India, which has huge biodiversity, is illustrated. The proposed system can show the detailed location of that particular species, where they can be found, and the shortest distance from the current location.
Widespread fear and panic has emerged about COVID-19 on social media platforms which are often su... more Widespread fear and panic has emerged about COVID-19 on social media platforms which are often supported by falsified and altered content. This mass hysteria creates public anxiety due to misinformation, misunderstandings, and ignorance of the impact of COVID-19. To assist health professionals in addressing this epidemic more appropriately at the onset, sentiment analysis can potentially help the authorities for devising appropriate strategies. This study analyzes tweets related to COVID-19 using a machine learning approach and offers a high-accuracy solution. Experiments are performed involving different machine and deep learning models along with various features such as Word2vec, term-frequency, term-frequency document frequency, and feature fusion of both feature-generating approaches. The proposed approach combines the extra tree classifier and convolutional neural network and uses feature fusion to achieve the highest accuracy score of 99%. The proposed approach obtains far be...
In the paper, the authors investigated and predicted the future environmental circumstances of a ... more In the paper, the authors investigated and predicted the future environmental circumstances of a COVID-19 to minimize its effects using artificial intelligence techniques. The experimental investigation of COVID-19 instances has been performed in ten countries, including India,
Early and precise detection of diabetic retinopathy prevents vision impairments through computer-... more Early and precise detection of diabetic retinopathy prevents vision impairments through computer-aided clinical procedures. Identifying the symptoms and processing those by using sophisticated clinical procedures reduces hemorrhage kind of risks. The input diabetic retinopathy images are influenced by using computer vision-based processes for segmentation and classification through feature extractions. In this article, a delimiting segmentation using knowledge learning (DS-KL) is introduced for classifying and detecting exudate regions by using varying histograms. The input image is identified for its histogram changes from the feature-dependent segmentation process. Depending on the training knowledge from multiple inputs with different exudate regions, the segmentation is performed. This segmentation identifies infected and noninfected regions across the delimiting pixel boundaries. The knowledge-learning process stores the newly identified exudate region for training and pixel co...
Journal of Ambient Intelligence and Humanized Computing
protective steps were put in place right away to stop the virus from spreading across the air as ... more protective steps were put in place right away to stop the virus from spreading across the air as presented in Leung et al. (2020). Citizens around the world were urged to take 1 Introduction When the SARS-CoV-2 pandemic broke out last 2019,
Technological advancements have made it possible to monitor, diagnose, and treat patients remotel... more Technological advancements have made it possible to monitor, diagnose, and treat patients remotely. The vital signs of patients can now be collected with the help of Internet of Things (IoT)-based wearable sensor devices and then uploaded on to a fog server for processing and access by physicians for recommending prescriptions and treating patients through the Internet of Medical Things (IoMT) devices. This research presents the outcome of a survey conducted on healthcare integrated with fog computing and IoT to help researchers understand the techniques, technologies and performance parameters. A comparison of existing research focusing on technologies, procedures, and findings has been presented to investigate several aspects of fog computing in healthcare IoT-based systems, such as increased temporal complexity, storage capacity, scalability, bandwidth, and latency. Additionally, strategies, tools, and sensors used in various diseases such as heart disease, chronic disease, chiku...
Over the last few years, different contaminants have posed a danger to the quality of the water. ... more Over the last few years, different contaminants have posed a danger to the quality of the water. Hence modelling and forecasting water quality are very important in the management of water contamination. The paper proposes an ensemble machine learning-based model for assessing water quality. The results of the proposed model are compared with several machine learning models, including k-nearest neighbour, NaĂ¯ve Bayes, support vector machine, and decision tree. The considered dataset contains seven statistically important parameters: pH, conductivity, dissolved oxygen, Biochemical Oxygen Demand, nitrate, total coliform, and fecal coliform. The water quality index is calculated for assessing water quality. To utilize an ensemble approach, a voting classifier has been designed with hard voting. The highest prediction accuracy of 99.5% of the water quality index is presented by the voting classifier as compared to the prediction accuracy of 99.2%, 90%, 79%, and 99% presented through k-n...
This research proposed an improved transfer-learning bird classification framework to achieve a m... more This research proposed an improved transfer-learning bird classification framework to achieve a more precise classification of Protected Indonesia Birds (PIB) which have been identified as the endangered bird species. The framework takes advantage of using the proposed sequence of Batch Normalization Dropout Fully-Connected (BNDFC) layers to enhance the baseline model of transfer learning. The main contribution of this work is the proposed sequence of BNDFC that can be applied to any Convolutional Neural Network (CNN) based model to improve the classification accuracy, especially for image-based species classification problems. The experiment results show that the proposed sequence of BNDFC layers outperform other combination of BNDFC. The addition of BNDFC can improve the model's performance across ten different CNN-based models. On average, BNDFC can improve by approximately 19.88% in Accuracy, 24.43% in F-measure, 17.93% in G-mean, 23.41% in Sensitivity, and 18.76% in Precision. Moreover, applying fine-tuning (FT) is able to enhance the accuracy by 0.85% with a smaller validation loss of 18.33% improvement. In addition, MobileNetV2 was observed to be the best baseline model with the lightest size of 35.9 MB and the highest accuracy of 88.07% in the validation set.
The forecasting of bus passenger flow is important to the bus transit system's operation. Because... more The forecasting of bus passenger flow is important to the bus transit system's operation. Because of the complicated structure of the bus operation system, it's difficult to explain how passengers travel along different routes. Due to the huge number of passengers at the bus stop, bus delays, and irregularity, people are experiencing difficulties of using buses nowadays. It is important to determine the passenger flow in each station, and the transportation department may utilize this information to schedule buses for each region. In Our proposed system we are using an approach called the deep learning method with long short-term memory, recurrent neural network, and greedy layer-wise algorithm are used to predict the Karnataka State Road Transport Corporation (KSRTC) passenger flow. In the dataset, some of the parameters are considered for prediction are bus id, bus type, source, destination, passenger count, slot number, and revenue These parameters are processed in a greedy layer-wise algorithm to make it has cluster data into regions after cluster data move to the long short-term memory model to remove redundant data in the obtained data and recurrent neural network it gives the prediction result based on the iteration factors of the data. These algorithms are more accurate in predicting bus passengers. This technique handles the problem of passenger flow forecasting in Karnataka State Road Transport Corporation Bus Rapid Transit (KSRTCBRT) transportation, and the framework provides resource planning and revenue estimation predictions for the KSRTCBRT.
In a network architecture, an intrusion detection system (IDS) is one of the most commonly used a... more In a network architecture, an intrusion detection system (IDS) is one of the most commonly used approaches to secure the integrity and availability of critical assets in protected systems. Many existing network intrusion detection systems (NIDS) utilize stand-alone classifier models to classify network traffic as an attack or as normal. Due to the vast data volume, these stand-alone models struggle to reach higher intrusion detection rates with low false alarm rates( FAR). Additionally, irrelevant features in datasets can also increase the running time required to develop a model. However, data can be reduced effectively to an optimal feature set without information loss by employing a dimensionality reduction method, which a classification model then uses for accurate predictions of the various network intrusions. In this study, we propose a novel feature-driven intrusion detection system, namely χ2-BidLSTM, that integrates a χ2 statistical model and bidirectional long short-term m...
A smart grid is a promising use-case of AIoT (AI (artificial intelligence) across IoT (internet o... more A smart grid is a promising use-case of AIoT (AI (artificial intelligence) across IoT (internet of things)) that enables bidirectional communication among utilities that arises with demand response (DR) schemes for demand-side management (DSM) and consumers that manage their power demands according to received DR signals. Disaggregating composite electric energy consumption data from a single minimal set of plug-panel current and voltage sensors installed at the electric panel in a practical field of interest, nonintrusive appliance load monitoring (NIALM), a cost-effective load disaggregation approach for (residential) DSM, is able to discern individual electrical appliances concerned without accessing each of them by individual plug-load power meters (smart plugs) deployed intrusively. The most common load disaggregation approaches are based on machine learning algorithms such as artificial neural networks, while approaches based on evolutionary computing, metaheuristic algorithms...
Multidimensional Systems and Signal Processing, 2021
This paper is mainly aimed at the decomposition of image quality assessment study by using Three ... more This paper is mainly aimed at the decomposition of image quality assessment study by using Three Parameter Logistic Mixture Model and k-means clustering (TPLMM-k). This method is mainly used for the analysis of various images which were related to several real time applications and for medical disease detection and diagnosis with the help of the digital images which were generated by digital microscopic camera. Several algorithms and distribution models had been developed and proposed for the segmentation of the images. Among several methods developed and proposed, the Gaussian Mixture Model (GMM) was one of the highly used models. One can say that almost the GMM was playing the key role in most of the image segmentation research works so far noticed in the literature. The main drawback with the distribution model was that this GMM model will be best fitted with a kind of data in the dataset. To overcome this problem, the TPLMM-k algorithm is proposed. The image decomposition process used in the proposed algorithm had been analyzed and its performance was analyzed with the help of various performance metrics like the Variance of Information (VOI), Global Consistency Error (GCE) and Probabilistic
Lung nodules are abnormal growths and lesions may exist. Both lungs may have nodules. Most lung n... more Lung nodules are abnormal growths and lesions may exist. Both lungs may have nodules. Most lung nodules are harmless (not cancerous/malignant). Pulmonary nodules are rare in lung cancer. X-rays and CT scans identify the lung nodules. Doctors may term the growth a lung spot, coin lesion, or shadow. It is necessary to obtain properly computed tomography (CT) scans of the lungs to get an accurate diagnosis and a good estimate of the severity of lung cancer. This study aims to design and evaluate a deep learning (DL) algorithm for identifying pulmonary nodules (PNs) using the LUNA-16 dataset and examine the prevalence of PNs using DB-Net. The paper states that a new, resource-efficient deep learning architecture is called for, and it has been given the name of DB-NET. When a physician orders a CT scan, they need to employ an accurate and efficient lung nodule segmentation method because they need to detect lung cancer at an early stage. However, segmentation of lung nodules is a difficult task because of the nodules' characteristics on the CT image as well as the nodules' concealed shape, visual quality, and context. The DB-NET model architecture is presented as a resource-efficient deep learning solution for handling the challenge at hand in this paper. Furthermore, it incorporates the Mish nonlinearity function and the mask class weights to improve segmentation effectiveness. In addition to the LUNA-16 dataset, which contained 1200 lung nodules collected during the LUNA-16 test, the LUNA-16 dataset was extensively used to train and assess the proposed model. The DB-NET architecture surpasses the existing U-NET model by a dice coefficient index of 88.89%, and it also achieves a similar level of accuracy to that of human experts.
Future wireless networks promise immense increases on data rate and energy efficiency while overc... more Future wireless networks promise immense increases on data rate and energy efficiency while overcoming the difficulties of charging the wireless stations or devices in the Internet of Things (IoT) with the capability of simultaneous wireless information and power transfer (SWIPT). For such networks, jointly optimizing beamforming, power control, and energy harvesting to enhance the communication performance from the base stations (BSs) (or access points (APs)) to the mobile nodes (MNs) served would be a real challenge. In this work, we formulate the joint optimization as a mixed integer nonlinear programming (MINLP) problem, which can be also realized as a complex multiple resource allocation (MRA) optimization problem subject to different allocation constraints. By means of deep reinforcement learning to estimate future rewards of actions based on the reported information from the users served by the networks, we introduce single-layer MRA algorithms based on deep Q-learning (DQN) ...
The aim of 5G wireless networks to provide Mbps and Gbps data rates to end users is expected to b... more The aim of 5G wireless networks to provide Mbps and Gbps data rates to end users is expected to be fulfilled by the advanced technologies such as multi-input multi-output (MIMO), carrier aggregation (CA), inter/intra-cell communication, and adaptive modulation and coding techniques, which would be all realized in the Long Term Evolution-Advanced (LTE-A) heterogeneous network constituted by macrocells (MCs) and small cells (SCs) adopting these 5G advanced techniques. Given the potential of significantly increasing the network performance, the resource allocation (RA) problem involved becomes harder than ever especially when MIMO and CA are included in the RA problem involving multiple types of resources to be concurrently determined for the global optimization. Facing this challenge, we develop a framework to jointly optimize energy efficiency (EE), spectrum efficiency (SE), and queue length for downlink transmissions with an overall and comprehensive consideration of dynamically allocating resource blocks (RBs), component carriers (CCs), modulation and coding schemes (MCSs), and deciding user association (UA) with a power control (PC) mechanism on discrete power levels (PLs) in the heterogeneous LTE-based MIMO wireless networks. Specially, for the complex joint RA, UA, and PC problem, we conduct a mixed integer programming model to accommodate the stochastic optimization problem involved with the drift-plus-penalty (DPP) approach for Lyapunov opportunistic optimization. In particular, although it involves a nondeterministic polynomial time (NP) problem, we can still show a reduced problem to be solved easily through linear relaxation when its coefficient matrix is totally unimodular (TUM), and to be solved efficiently as well even when the TUM property is not guaranteed. Based on the reduction, we further develop a distributed or semi-distributed algorithm operated on two levels to approach the optimal results with lower complexity if the UA requirement can be relaxed. Finally, apart from exhibiting its performance on the weighting parameters, the numerical experiments also show our approach to make a good tradeoff among SE, EE, and queue length, and outperform the greedy-based state-of-the-art algorithms. INDEX TERMS LTE-A heterogeneous wireless networks, MIMO, carrier aggregation, multi-resource allocation, user association, power control.
Security with Intelligent Computing and Big-data Services, 2018
In this paper, we proposed a joint lossless index coding and data hiding technique for the palett... more In this paper, we proposed a joint lossless index coding and data hiding technique for the palette images. The palette image is the compressed image of the color image quantization technique. The compressed codes of the palette image consist of the index table and the color palette. In the proposed technique, a three-category lossless index coding method is employed. The secret data is embedded into the encoded index table during the index coding process is executed. From the results, it is shown that good hiding capacity is obtained in the proposed technique while keeping a good bit rate.
With the rapid development of computing technologies, human-centered computing (HCC) has become a... more With the rapid development of computing technologies, human-centered computing (HCC) has become an emerging multidisciplinary area which integrates ubiquitous computing, wearable computing, and so on. When large amounts of information has to be mutually analyzed, HCC aims to bridge the gap between people and computing systems. However, ensuring a secure transmission becomes challenging in HCC because malicious hackers tend to tamper with the Internet data. The massive amount of counterfeited or tampered data on the Internet damage data trustworthiness and impede the progress of HCC. To address this problem, this study presented an image authentication method based on the residual histogram shifting technique. In this technique, an image histogram is generated using block-based processing. Then, the histogram is modified to achieve a high embedding capacity. By manipulating the image histogram, secret or authentication codes can be embedded in the image itself to prevent private info...
Uploads
Papers by Yu-Chen Hu