Search

In the Internet of Things (IoT) era, an immense amount of sensing devices are obtained and produce various sensory data over time for a wide range of disciplines and applications. These devices will result in significant, fast, and real-time data streams based on the utilization characteristics. Utilizing analytics over such data streams to identify new information, model future insights, and make control decisions is a necessary process that makes IoT a worthy paradigm for businesses and a quality-of-life improving technology. This paper presents a study of digital agriculture and its significance in terms of the application of an IoT-based device — a two-wheeled self-balancing robot — followed by a thorough procedural explanation of the development of the device, which begins with the mathematical modeling of the system through the Euler–Lagrange method to obtain the equations of motion for the same and linearize the equation to define the control method to be used to balance the robot structure, all based on the concept of the inverted pendulum. Then paper discusses the suitable and the most efficient control method, which is the linear quadratic regulator (LQR), for these robots. Then deep learning-based LQR (DL-LQR) method is implemented in the robots performing the algorithm to balance it successfully. [ABSTRACT FROM AUTHOR]

Nowadays, Biometric systems are prevalent for personal recognition. But due to pandemic COVID 19, it is difficult to pursue a touch-based biometric system. To encourage a touchless biometric system, a less constrained multimodal personal identification system using palmprint and dorsal hand vein is presented. Hand based Touchless recognition system gives a higher user-friendly system and avoids the spread of coronavirus. A method using Convolution Neural Networks(CNN) to extract discriminative features from the data samples is proposed. A pre-trained function PCANeT is used in the experiments to show the performance of the system in fusion scheme. This method doesn't require keeping the palm in a specific position or at a certain distance like most other papers. Different patches of ROI are used at two different layers of CNN. Fusion of palmprint and dorsal hand vein is done for final result matching. Both Feature level and score level fusion methods are compared. Results shows the accuracy of upto 98.55% and 98.86% and Equal error rate (EER) of upto 1.22% and 0.93% for score level fusion and feature level fusion, respectively. Our method gives higher accurate results in a less constrained environment. [ABSTRACT FROM AUTHOR]

The availability of techniques for driver distraction detection has been difficult to put to use because of delays caused due to lag in inferencing the model. Distractions caused due to handheld devices have been major causes of traffic accidents as they affect the decision-making capabilities of the driver and gives them less time to react to difficult situations. Often drivers try to multitask which reduces their reaction time leading to accidents, which can easily be avoided if they had been attentive. As such, problems related to the driver's negligence towards safety a possible solution is to monitor the driver and driving behavior and alerting them if they are distracted. In this paper, we propose a novel approach for detecting when a driver is distracted due to in hand electronic devices which is not only able to detect the distraction with high accuracy but also is energy and memory efficient. Our proposed compressed neural got an accuracy of 0.83 in comparison to 0.86 of heavyweight network. [ABSTRACT FROM AUTHOR]

A convolution neural network (CNN) based deep learning method has been proposed for automatic classification and localization of nonlinear loads present in an interconnected power system. The identification of nonlinear loads has been previously dealt with the use of Nonlinear Auto Regression neural network with eXogenous inputs (NARX), Backpropagation Neural Network (BPNN), Probabilistic Neural Network (PNN), Artificial Neural Networks (ANN) and Fuzzy Logic (FL). However, these techniques had not explored the area of classification of industrial and domestic nonlinear loads in an interconnected power system. Also, a Deep learning-based solution for identification of the type of nonlinear load has not been reported in the literature to date. Hence, to address these shortcomings, an IEEE-9 Bus system with industrial nonlinear loads has been used to obtain various current waveforms with distortions. The recorded current waveforms are transformed into a time-frequency (TF) domain plane, and the obtained images are then fed to the deep learning algorithm. The colored images of the TF plots of each type of nonlinear load in Red-Green-Blue (RGB) index provide the best visual features for extraction. The TF domain signatures of individual events are scaled to a standard size before feeding to the algorithm. Through these TF signatures, unique features were extracted with the deep learning algorithm, and then passed on to different stages of convolution and max-pooling with fully connected layers. The softmax classifier at the end classifies the input data into the type of nonlinear present in the power system. The algorithm, when run at different buses, also identifies the location of the nonlinear load. The proposed methodology avoids the usage of any additional fusion layer for obtaining unique features, reduces the training time and maintains the highest accuracy of 100%. [ABSTRACT FROM AUTHOR]

Person identification using biometric features is an effective method for recognizing and authenticating the identity of a person. Multimodal biometric systems combine different biometric modalities in order to make better predictions as well as for achieving increased robustness. This paper proposes a touchless multimodal person identification model using deep learning techniques by combining the gait and speech modalities. Separate pipelines for both the modalities were developed using Convolutional Neural Networks. The paper also explores various fusion strategies for combining the two pipelines and shows how various metrics get affected with different fusion strategies. Results show that weighted average and product fusion rules work best for the data used in the experiments. [ABSTRACT FROM AUTHOR]

Single biometric modalities like facial features and vein patterns despite being reliable characteristics show limitations that restrict them from offering high performance and robustness. Multimodal biometric systems have gained interest due to their ability to overcome the inherent limitations of the underlying single biometric modalities and generally have been shown to improve the overall performance for identification and recognition purposes. This paper proposes highly accurate and robust multimodal biometric identification as well as recognition systems based on fusion of face and finger vein modalities. The feature extraction for both face and finger vein is carried out by exploiting deep convolutional neural networks. The fusion process involves combining the extracted relevant features from the two modalities at score level. The experimental results over all considered public databases show a significant improvement in terms of identification and recognition accuracy as well as equal error rates. [ABSTRACT FROM AUTHOR]

Industrial Wireless Sensor Network (IWSN) includes numerous sensor nodes that collect data about target objects and transmit to sink nodes (SN). During data transmission among nodes, intrusion detection is carried to improve data security and privacy. Intrusion detection system (IDS) examines the network for intrusions based on user activities. Several works have been done in the field of intrusion detection and different measures are carried out to increase data security from the issues related to black hole, Sybil attack, Worm hole, identity replication attack and etc. In various existing approaches, secure data transmission is not achieved, therefore resulted in compromising the security and privacy of IWSNs. Accurate intrusion detection is still challenging task in terms of improving security and intrusion detection rate. In order to improve intrusion detection rate (IDR) with minimum time, generalized Frechet Hyperbolic Deep and Dirichlet Secured (FHD-DS) data communication model is introduced. At first, Frechet Hyperbolic Deep Traffic (FHDT) feature extraction method is designed to extract more relevant network activities and inherent traffic features. With the help of extracted features, anomalous or normal data is predicted. Followed by Statistical Dirichlet Anomaly-based Intrusion Detection model is applied to discover intrusion. Here, Dirichlet distribution is evaluated to attain secure data transmission and significantly detect intrusions in WSNs. Experimental evaluation is carried out with KDD cup 99 dataset on factors such as IDR, intrusion detection time (IDT) and data delivery rate (DDR). The observed results show that the generalized FHD-DS data communication method achieves higher IDR with minimum time. [ABSTRACT FROM AUTHOR]

In this paper, we present a method for generating bird's eye video from egocentric RGB videos. Working with egocentric views is tricky since such the view is highly warped and prone to occlusions. On the other hand, a bird's eye view has a consistent scaling in at least the two dimensions it shows. Moreover, most of the state-of-the-art systems for tasks such as path prediction are built for bird's eye views of the subjects. We present a deep learning-based approach that transfers the egocentric RGB images captured from a dashcam of a car to bird's eye view. This is a task of view translation, and we perform two experiments. The first one uses an image-to-image translation method, and the other uses a video-to-video translation. We compare the results of our work with homographic transformation, and our SSIM values are better by a margin of 77% and 14.4%, and the RMSE errors are lower by 40% and 14.6% for image-to-image translation and video-to-video translation, respectively. We also visually show the efficacy and limitations of each method with helpful insights for future research. Compared to previous works that use homography and LIDAR for 3D point clouds, our work is more generalizable and does not require any expensive equipment. [ABSTRACT FROM AUTHOR]

Almost all of the automatic accident detection (AAD) system suffers from the tradeoff between computational overhead and detection accuracy. Recent advances in detection and classification methodologies have shown phenomenal improvements in accuracy but these systems require a huge number of computational resources making them unviable for deployment requiring real-time feedback. This paper proposes a methodology to develop a reliable and computationally inexpensive real-time automatic accident detection system that can be deployed with minimum hardware requirements. Specifically, we split our AAD system into three major stages (Detection, Tracking and Classification) and propose algorithms for each stage with reduced computational need. For the detection stage, we propose Mini-YOLO, a deep learning model architecture trained using knowledge distillation that has comparable accuracy with its counterpart YOLO(You-Only-Look-Once) with reduced model size and computational overhead. Mini-YOLO achieves an average precision (AP) score of 34.2 on the MS-COCO dataset while outperforming all other detection algorithms in runtime complexity, achieving a staggering 28 frames per second on a low-end machine. For the tracking stage, we adopt SORT (Simple Online Real-time Tracking) and for classification stage, we compare multiple machine learning algorithms and show that a support vector machine with radial basis kernel performs the best with an area under the curve (AUC) score of 0.98, model size of 448 KB (kilobytes) and 12.73 ms (milliseconds) latency. [ABSTRACT FROM AUTHOR]

A cancer tumour consists of thousands of genetic mutations. Even after advancement in technology, the task of distinguishing genetic mutations, which act as driver for the growth of tumour with passengers (Neutral Genetic Mutations), is still being done manually. This is a time-consuming process where pathologists interpret every genetic mutation from the clinical evidence manually. These clinical shreds of evidence belong to a total of nine classes, but the criterion of classification is still unknown. The main aim of this research is to propose a multiclass classifier to classify the genetic mutations based on clinical evidence (i.e., the text description of these genetic mutations) using Natural Language Processing (NLP) techniques. The dataset for this research is taken from Kaggle and is provided by the Memorial Sloan Kettering Cancer Center (MSKCC). The world-class researchers and oncologists contribute the dataset. Three text transformation models, namely, CountVectorizer, TfidfVectorizer, and Word2Vec, are utilized for the conversion of text to a matrix of token counts. Three machine learning classification models, namely, Logistic Regression (LR), Random Forest (RF), and XGBoost (XGB), along with the Recurrent Neural Network (RNN) model of deep learning, are applied to the sparse matrix (keywords count representation) of text descriptions. The accuracy score of all the proposed classifiers is evaluated by using the confusion matrix. Finally, the empirical results show that the RNN model of deep learning has performed better than other proposed classifiers with the highest accuracy of 70%. [ABSTRACT FROM AUTHOR]

Through the commencement of the COVID-19 pandemic, the whole globe is in disarray and debating on unique approaches to stop this viral transmission. Masks are being worn by people all around the world as one of the preventative measures to avoid contracting this sickness. Although some people are following and adopting this precaution, others are not, despite official recommendations from the administration and public health organisations has been announced. In this paper DTLMV2 (Deep Transfer Learning MobileNetV2 for the objective of classification) is proposed - A face mask identification model that can reliably determine whether an individual is wearing a mask or not is suggested and implemented in this work. The model architecture employs the peruse of MobileNetV2, a lightweight Convolutional Neural Network (CNN) that requires less computing power and can be readily integrated into computer vision and mobile systems. The computer vision with MobileNet is required to formulate a low-cost mask detection system for a group of people in open spaces that can assist in determining whether a person is wearing a mask or not, as well as function as a surveillance system since it is effective on both real-time pictures and videos. The face recognition model obtained 97.01% accuracy on validation data, 98% accuracy on training data and 97.45% accuracy on testing data. [ABSTRACT FROM AUTHOR]

Virus diseases are a continued threat to human health in both community and healthcare settings. The current virus disease COVID-19 outbreak raises an unparalleled public health issue for the world at large. Wuhan is the city in China from where this virus came first and, after some time the whole world was affected by this severe disease. It is a challenge for every country's people and higher authorities to fight with this battle due to the insufficient number of resources. On-going assessment of the epidemiological features and future impacts of the COVID-19 disease is required to stay up-to-date of any changes to its spread dynamics and foresee needed resources and consequences in different aspects as social or economic ones. This paper proposes a prediction model of confirmed and death cases of COVID-19. The model is based on a deep learning algorithm with two long short-term memory (LSTM) layers. We consider the available infection cases of COVID-19 in India from January 22, 2020, till October 9, 2020, and parameterize the model. The proposed model is an inference to obtain predicted coronavirus cases and deaths for the next 30 days, taking the data of the previous 260 days of duration of the pandemic. The proposed deep learning model has been compared with other popular prediction methods (Support Vector Machine, Decision Tree and Random Forest) showing a lower normalized RMSE. This work also compares COVID-19 with other previous diseases (SARS, MERS, h1n1, Ebola, and 2019-nCoV). Based on the mortality rate and virus spread, this study concludes that the novel coronavirus (COVID-19) is more dangerous than other diseases. • A predictive model of confirmed and death cases of COVID-19 is proposed. • The predictive model is a deep learning model based on LSTM networks. • COVID-19 cases and deaths for the next 30 days are predicted. • The spread and mortality rates of five different virus diseases are compared. • It is concluded that COVID-19 is more dangerous than SARS, MERS, h1n1, and Ebola. [ABSTRACT FROM AUTHOR]

Multiple sclerosis (MS) is a chronic neurological disease of the central nervous system (CNS). Early diagnosis of MS is highly desirable as treatments are more effective in preventing MS-related disability when given in the early stages of the disease. The main aim of this research is to predict the occurrence of a second MS-related clinical event, which indicates the conversion of clinically isolated syndrome (CIS) to clinically definite MS (CDMS). In this study, we apply a branch of artificial intelligence known as deep learning and develop a fully automated algorithm primed with convolutional neural network (CNN) that has the ability to learn from MRI scan features. The basic architecture of our algorithm is that of the VGG16 CNN model, but amended such that it can handle MRI DICOM images. A dataset comprised of scans acquired using two different scanners was used for the purposes of verification of the algorithm. A group of 49 patients had volumetric MRI scans taken at onset of the disease and then again one year later using one of the two scanners. In total, this yielded 7360 images which were then used for training, validation, and testing of the algorithm. Initially, these raw images were taken through 4 steps of preprocessing. In order to boost the efficiency of the process, we pretrained our algorithm using the publicly available ADNI dataset used to classify Alzheimer's disease. Finally, we used our preprocessed dataset to train and test the algorithm. Clinical evaluation conducted a year after the first time point revealed that 26 of the 49 patients had converted to CDMS, while the remaining 23 had not. Results of testing showed that our algorithm was able to predict the clinical results with an accuracy of 88.8% and with an area under the curve (AUC) of 91%. A highly accurate algorithm was developed using CNN approach to reliably predict conversion of patients with CIS to CDMS using MRI data from two different scanners. [ABSTRACT FROM AUTHOR]