Abstract: With the development of intelligent vehicle systems, a high-precision road map is increasingly needed in many aspects. The automatic lane lines extraction and modeling are the most essential steps for the generation of a precise lane-level road map. In this paper, an automatic lane-level road map generation system is proposed. To extract the road markings on the ground, the multi-region Otsu thresholding method is applied, which calculates the intensity value of laser data that maximizes the variance between background and road markings. The extracted road marking points are then projected to the raster image and clustered using a two-stage clustering algorithm. Lane lines are subsequently recognized from these clusters by the shape features of their minimum bounding rectangle. To ensure the storage efficiency of the map, the lane lines are approximated to cubic polynomial curves using a Bayesian estimation approach. The proposed lane-level road map generation system has been tested on urban and expressway conditions in Hefei, China. The experimental results on the datasets show that our method can achieve excellent extraction and clustering effect, and the fitted lines can reach a high position accuracy with an error of less than 10 cm.
Abstract: Autonomous driving systems require high reliability to provide people with a safe and comfortable driving experience. However, despite the development of a number of vehicle sensors, it is difficult to always provide high perceived performance in driving environments that vary from time to season. The image segmentation method using deep learning, which has recently evolved rapidly, provides high recognition performance in various road environments stably. However, since the system controls a vehicle in real time, a highly complex deep learning network cannot be used due to time and memory constraints. Moreover, efficient networks are optimized for GPU environments, which degrade performance in embedded processor environments equipped simple hardware accelerators. In this paper, a semantic segmentation network, matrix multiplication accelerator network (MMANet), optimized for matrix multiplication accelerator (MMA) on Texas instrument digital signal processors (TI DSP) is proposed to improve the recognition performance of autonomous driving system. The proposed method is designed to maximize the number of layers that can be performed in a limited time to provide reliable driving environment information in real time. First, the number of channels in the activation map is fixed to fit the structure of MMA. By increasing the number of parallel branches, the lack of information caused by fixing the number of channels is resolved. Second, an efficient convolution is selected depending on the size of the activation. Since MMA is a fixed, it may be more efficient for normal convolution than depthwise separable convolution depending on memory access overhead. Thus, a convolution type is decided according to output stride to increase network depth. In addition, memory access time is minimized by processing operations only in L3 cache. Lastly, reliable contexts are extracted using the extended atrous spatial pyramid pooling (ASPP). The suggested method gets stable features from an extended path by increasing the kernel size and accessing consecutive data. In addition, it consists of two ASPPs to obtain high quality contexts using the restored shape without global average pooling paths since the layer uses MMA as a simple adder. To verify the proposed method, an experiment is conducted using perfsim, a timing simulator, and the Cityscapes validation sets. The proposed network can process an image with 640 x 480 resolution for 6.67 ms, so six cameras can be used to identify the surroundings of the vehicle as 20 frame per second (FPS). In addition, it achieves 73.1% mean intersection over union (mIoU) which is the highest recognition rate among embedded networks on the Cityscapes validation set.
Abstract: This study helps Public Water Bureaus in giving reliable answers to water concession requests. Rapidly increasing water requests can be supported provided that further uses of a river course are not totally compromised, and environmental features are protected as well. Strictly speaking, a water concession can be considered a continuous drawing from the source and causes a mean annual streamflow reduction. Therefore, deciding if a water concession is appropriate or inappropriate seems to be easily solved by comparing the generic demand to the mean annual streamflow value at disposal. Still, the immediate shortcoming for such a comparison is that streamflow data are information available only for few catchments and, most often, limited to specific sites. Subsequently, comparing the generic water demand to mean daily discharge is indeed far from being completely satisfactory since the mean daily streamflow is greater than the water withdrawal for a long period of a year. Consequently, such a comparison appears to be of little significance in order to preserve the quality and the quantity of the river. In order to overcome such a limit, this study aims to complete the information provided by flow duration curves introducing a link between Flow Duration Curves (FDCs) and recession curves and aims to show the chronological sequence of flows with a particular focus on low flow data. The analysis is carried out on 25 catchments located in North-Eastern Italy for which daily data are provided. The results identify groups of catchments as hydrologically homogeneous, having the lower part of the FDCs (corresponding streamflow interval is streamflow Q between 300 and 335, namely: Q(300), Q(335)) smoothly reproduced by a common recession curve. In conclusion, the results are useful to provide more reliable answers to water request, especially for those catchments which show similar hydrological response and can be used for a focused regionalization approach on low flow data. A mathematical link between streamflow duration curves and recession curves is herein provided, thus furnishing streamflow duration curves information upon a temporal sequence of data. In such a way, by introducing assumptions on recession curves, the chronological sequence upon low flow data can also be attributed to FDCs, which are known to lack this information by nature.
Abstract: Fake news detection research is still in the early stage as this is a relatively new phenomenon in the interest raised by society. Machine learning helps to solve complex problems and to build AI systems nowadays and especially in those cases where we have tacit knowledge or the knowledge that is not known. We used machine learning algorithms and for identification of fake news; we applied three classifiers; Passive Aggressive, Naïve Bayes, and Support Vector Machine. Simple classification is not completely correct in fake news detection because classification methods are not specialized for fake news. With the integration of machine learning and text-based processing, we can detect fake news and build classifiers that can classify the news data. Text classification mainly focuses on extracting various features of text and after that incorporating those features into classification. The big challenge in this area is the lack of an efficient way to differentiate between fake and non-fake due to the unavailability of corpora. We applied three different machine learning classifiers on two publicly available datasets. Experimental analysis based on the existing dataset indicates a very encouraging and improved performance.
Abstract: Poor system use, including inappropriate design of health information systems, causes difficulties in communication with patients and increased time spent by healthcare professionals in recording the necessary health information for medical records. System features like pop-up reminders, complex menus, and poor user interfaces can make medical records far more time consuming than paper cards as well as affect decision-making processes. Although errors associated with health information and their real and likely effect on the quality of care and patient safety have been documented for many years, more research is needed to measure the occurrence of these errors and determine the causes to implement solutions. Therefore, the purpose of this paper is to identify data integrity challenges in hospital information systems through a scoping review and based on the results provide recommendations on how to manage these. Only 34 papers were found to be most suitable out of 297 publications initially identified in the field. The results indicated that human and computerized systems are the most common challenges associated with data integrity and factors such as policy, environment, health workforce, and lack of awareness attribute to these challenges but if measures are taken the data integrity challenges can be managed.
Abstract: Telecommunication service providers demand accurate
and precise prediction of customer churn probabilities to increase the
effectiveness of their customer relation services. The large amount of
customer data owned by the service providers is suitable for analysis
by machine learning methods. In this study, expenditure data of
customers are analyzed by using an artificial neural network (ANN).
The ANN model is applied to the data of customers with different
billing duration. The proposed model successfully predicts the churn
probabilities at 83% accuracy for only three months expenditure data
and the prediction accuracy increases up to 89% when the nine month
data is used. The experiments also show that the accuracy of ANN
model increases on an extended feature set with information of the
changes on the bill amounts.
Abstract: The management of COVID-19 patients based on chest imaging is emerging as an essential tool for evaluating the spread of the pandemic which has gripped the global community. It has already been used to monitor the situation of COVID-19 patients who have issues in respiratory status. There has been increase to use chest imaging for medical triage of patients who are showing moderate-severe clinical COVID-19 features, this is due to the fast dispersal of the pandemic to all continents and communities. This article demonstrates the development of machine learning techniques for the test of COVID-19 patients using Chest X-Ray (CXR) images in nearly real-time, to distinguish the COVID-19 infection with a significantly high level of accuracy. The testing performance has covered a combination of different datasets of CXR images of positive COVID-19 patients, patients with viral and bacterial infections, also, people with a clear chest. The proposed AI scheme successfully distinguishes CXR scans of COVID-19 infected patients from CXR scans of viral and bacterial based pneumonia as well as normal cases with an average accuracy of 94.43%, sensitivity 95%, and specificity 93.86%. Predicted decisions would be supported by visual evidence to help clinicians speed up the initial assessment process of new suspected cases, especially in a resource-constrained environment.
Abstract: Current collection quality is one of the limiting factors when increasing trains movement speed in the rail sector. With the movement speed growth, the impact forces on the current collector from the rolling stock and the aerodynamic influence increase, which leads to the spread in the contact pressure values, separation of the current collector head from the contact wire, contact arcing and excessive wear of the contact elements. The upcoming trend in resolving this issue is the use of the automatic control systems providing stabilization of the contact pressure value. The present paper considers the features of the contemporary automatic control systems of the current collector’s pressure; their major disadvantages have been stated. A scheme of current collector pressure automatic control has been proposed, distinguished by a proactive influence on undesirable effects. A mathematical model of contact strips wearing has been presented, obtained in accordance with the provisions of the central composition rotatable design program. The analysis of the obtained dependencies has been carried out. The procedures for determining the optimal current collector pressure on the contact wire and the pressure control principle in the pneumatic drive have been described.
Abstract: Augmented Reality and Augmented Intelligence are radically changing information technology. The path that starts from the keyboard and then, passing through milestones such as Siri, Alexa and other vocal avatars, reaches a more fluid and natural communication with computers, thus converting the dichotomy between man and machine into a harmonious interaction, now heads unequivocally towards a new IT paradigm, where holographic computing will play a key role. The OLOS® platform contributes substantially to this trend in that it infuses computers with human features, by transferring the gestures and expressions of persons of flesh and bones to anthropomorphic holographic interfaces which in turn will use them to interact with real-life humans. In fact, we could say, boldly but with a solid technological background to back the statement, that OLOS® gives reality to an altogether new entity, placed at the exact boundary between nature and technology, namely the holographic human being. Holographic humans qualify as the perfect carriers for the virtual reincarnation of characters handed down from history and tradition. Thus, they provide for an innovative and highly immersive way of experiencing our cultural heritage as something alive and pulsating in the present.
Abstract: This paper briefly reviews the range of curatorial interventions made to preserve and display the Timbuktu Manuscripts. The government of South Africa and Mali collaborated to preserve the manuscripts, and brief notes will be presented about the value of archives in those specific spaces. The research initiatives of the Tombouctou Manuscripts Project, based at the University of Cape Town, feature prominently in the text. A brief overview of the history of the archive will be presented and its preservation as a key turning point in curating the intellectual history of the continent. The strategies of preservation, curation, publication and digitization are presented as complimentary interventions. Each materialization of the manuscripts contributes something significant; the complexity of the contribution is dependent primarily on the format of presentation. This integrated reading of the manuscripts is presented as a means to gain a more nuanced understanding of the past, which greatly surpasses how much information would be gleaned from relying on a single media format.
Abstract: Traffic control has been one of the most common and irritating problems since the time automobiles have hit the roads. Problems like traffic congestion have led to a significant time burden around the world and one significant solution to these problems can be the proper implementation of the Intelligent Transport System (ITS). It involves the integration of various tools like smart sensors, artificial intelligence, position technologies and mobile data services to manage traffic flow, reduce congestion and enhance driver's ability to avoid accidents during adverse weather. Road and traffic signs’ recognition is an emerging field of research in ITS. Classification problem of traffic signs needs to be solved as it is a major step in our journey towards building semi-autonomous/autonomous driving systems. The purpose of this work focuses on implementing an approach to solve the problem of traffic sign classification by developing a Convolutional Neural Network (CNN) classifier using the GTSRB (German Traffic Sign Recognition Benchmark) dataset. Rather than using hand-crafted features, our model addresses the concern of exploding huge parameters and data method augmentations. Our model achieved an accuracy of around 97.6% which is comparable to various state-of-the-art architectures.
Abstract: Microblogging has become increasingly popular for commenting on current events, spreading gossip, and encouraging individualism--which favors its low-context communication channel. These social media (SM) platforms allow users to express opinions while interacting with a wide range of populations. Hashtags allow immediate identification of like-minded individuals worldwide on a vast array of topics. The output of the analytic tool, Linguistic Inquiry and Word Count (LIWC)--a program that associates psychological meaning with the frequency of use of specific words--may suggest the nature of individuals’ internal states and general sentiments. When applied to groupings of SM posts unified by a hashtag, such information can be helpful to community leaders during periods in which the forming of public opinion happens in parallel with the unfolding of political, economic, or social events. This is especially true when outcomes stand to impact the well-being of the group. Here, we applied the online tools, Google Translate and the University of Texas’s LIWC, to a 90-posting sample from a corpus of Colombian Spanish microblogs. On translated disjoint sets, identified by hashtag as being authored by advocates of voting “No,” advocates voting “Yes,” and entities refraining from hashtag use, we observed the value of LIWC’s Tone feature as distinguishing among the categories and the word “peace,” as carrying particular significance, due to its frequency of use in the data.
Abstract: This paper presents and benchmarks a number of
end-to-end Deep Learning based models for metaphor detection in
Greek. We combine Convolutional Neural Networks and Recurrent
Neural Networks with representation learning to bear on the metaphor
detection problem for the Greek language. The models presented
achieve exceptional accuracy scores, significantly improving the
previous state-of-the-art results, which had already achieved accuracy
0.82. Furthermore, no special preprocessing, feature engineering or
linguistic knowledge is used in this work. The methods presented
achieve accuracy of 0.92 and F-score 0.92 with Convolutional
Neural Networks (CNNs) and bidirectional Long Short Term Memory
networks (LSTMs). Comparable results of 0.91 accuracy and 0.91
F-score are also achieved with bidirectional Gated Recurrent Units
(GRUs) and Convolutional Recurrent Neural Nets (CRNNs). The
models are trained and evaluated only on the basis of training tuples,
the related sentences and their labels. The outcome is a state-of-the-art
collection of metaphor detection models, trained on limited labelled
resources, which can be extended to other languages and similar
tasks.
Abstract: Counterfeit goods and documents are a global problem, which needs more and more sophisticated methods of resolving it. Existing techniques using watermarking or embedding symbols on objects are not suitable for all use cases. To address those special needs, we created complete system allowing authentication of paper documents and physical objects with flat surface. Objects are marked using orientation independent and resistant to camera noise 2D graphic codes, named DotAuth. Based on the identifier stored in 2D code, the system is able to perform basic authentication and allows to conduct more sophisticated analysis methods, e.g., relying on augmented reality and physical properties of the object. In this paper, we present the complete architecture, algorithms and applications of the proposed system. Results of the features comparison of the proposed solution and other products are presented as well, pointing to the existence of many advantages that increase usability and efficiency in the means of protecting physical objects.
Abstract: Cluster Redevelopment is a new concept in the city of Mumbai. Its regulations were laid down by the government in 2009. The concept of cluster redevelopment encompasses a group of buildings defined by a boundary as specified by the municipal authority (in this case, Mumbai), which may be dilapidated or approved for redevelopment. The study analyses the effect of cluster redevelopment in the form of renewal of old group of buildings as compared to refurbishment or restoration - on energy consumption. The methodology includes methods of assessment to determine increase or decrease in energy consumption in cluster redevelopment based on different criteria such as carpet area of the units, building envelope and its architectural elements. Results show that as the area and number of units increase the Energy consumption increases and the EPI (energy performance index) decreases as compared to the base case. The energy consumption per unit area declines by 29% in the proposed cluster redevelopment as compared to the original settlement. It is recommended that although the development is spacious and provides more light and ventilation, aspects such as glass type, traditional architectural features and consumer behavior are critical in the reduction of energy consumption.
Abstract: Accurate segmentation of the optic disc is very
important for computer-aided diagnosis of several ocular diseases
such as glaucoma, diabetic retinopathy, and hypertensive retinopathy.
The paper presents an accurate and fast optic disc detection and
segmentation method using an attention based fully convolutional
network. The network is trained from scratch using the fundus images
of extended MESSIDOR database and the trained model is used for
segmentation of optic disc. The false positives are removed based on
morphological operation and shape features. The result is evaluated
using three-fold cross-validation on six public fundus image databases
such as DIARETDB0, DIARETDB1, DRIVE, AV-INSPIRE, CHASE
DB1 and MESSIDOR. The attention based fully convolutional
network is robust and effective for detection and segmentation of
optic disc in the images affected by diabetic retinopathy and it
outperforms existing techniques.
Abstract: Electrocardiographic (ECG) machine is an important equipment to diagnose heart problems. Besides, the ECG signals are used to detect many other features of human body and behavior. But it is not so cheap and simple in operation to be used in the countries like Bangladesh, where most of the people are very low income earners. Therefore, in this paper, we have tried to implement a simple and portable ECG machine. Since Arduino Uno microcontroller is very cheap, we have used it in our system to minimize the cost. Our designed system is powered by a 2-voltage level DC power supply. It provides wireless connectivity to have ECG data either in smartphone having android operating system or a PC/laptop having Windows operating system. To get the data, a graphic user interface has been designed. Android application has also been made using IDE for Android 2.3 and API 10. Since it requires no USB host API, almost 98% Android smartphones, available in the country, will be able to use it. We have calculated the heart rate from the measured ECG by our designed machine and by an ECG machine of a reputed diagnostic center in Dhaka city for the same people at the same time on same day. Then we calculated the percentage of errors between the readings of two machines and computed its average. From this computation, we have found out that the average percentage of error is within an acceptable limit.
Abstract: Seismic analysis of bedforms has proven to be one of the best ways to study deepwater sedimentary features. Canyons are known to be sediment transportation conduit. Sediment wave are large-scale depositional bedforms in various parts of the world's oceans formed predominantly by suspended load transport. These undulating objects usually have tens of meters to a few kilometers in wavelength and a height of several meters. Cyclic steps have long long-wave upstream-migrating bedforms confined by internal hydraulic jumps. They usually occur in regions with high gradients and slope breaks. Cyclic steps and migrating sediment waves are the most common bedform on the seafloor. Cyclic steps and related sediment wave bedforms are significant to the morpho-dynamic evolution of deep-water depositional systems architectural elements, especially those located along tectonically active margins with high gradients and slope breaks that can promote internal hydraulic jumps in turbidity currents. This report examined sedimentary activities and sediment transportation in submarine canyons and provided distinctive insight into factors that created a complex seabed canyon system in the Ceara Fortaleza basin Brazilian Equatorial Margin (BEM). The growing importance of cyclic steps made it imperative to understand the parameters leading to their formation, migration, and architecture as well as their controls on sediment transport in canyon thalweg. We extracted the parameters of the observed bedforms and evaluated the aspect ratio and asymmetricity. We developed a relationship between the hydraulic jump magnitude, depth of the hydraulic fall and the length of the cyclic step therein. It was understood that an increase in the height of the cyclic step increases the magnitude of the hydraulic jump and thereby increases the rate of deposition on the preceding stoss side. An increase in the length of the cyclic steps reduces the magnitude of the hydraulic jump and reduces the rate of deposition at the stoss side. Therefore, flat stoss side was noticed at most preceding cyclic step and sediment wave.
Abstract: Modeling dam-break flows over non-flat beds requires
an accurate representation of the topography which is the main
source of uncertainty in the model. Therefore, developing robust
and accurate techniques for reconstructing topography in this class
of problems would reduce the uncertainty in the flow system. In
many hydraulic applications, experimental techniques have been
widely used to measure the bed topography. In practice, experimental
work in hydraulics may be very demanding in both time and cost.
Meanwhile, computational hydraulics have served as an alternative
for laboratory and field experiments. Unlike the forward problem,
the inverse problem is used to identify the bed parameters from the
given experimental data. In this case, the shallow water equations
used for modeling the hydraulics need to be rearranged in a way
that the model parameters can be evaluated from measured data.
However, this approach is not always possible and it suffers from
stability restrictions. In the present work, we propose an adaptive
optimal control technique to numerically identify the underlying bed
topography from a given set of free-surface observation data. In this
approach, a minimization function is defined to iteratively determine
the model parameters. The proposed technique can be interpreted
as a fractional-stage scheme. In the first stage, the forward problem
is solved to determine the measurable parameters from known data.
In the second stage, the adaptive control Ensemble Kalman Filter is
implemented to combine the optimality of observation data in order to
obtain the accurate estimation of the topography. The main features
of this method are on one hand, the ability to solve for different
complex geometries with no need for any rearrangements in the
original model to rewrite it in an explicit form. On the other hand, its
achievement of strong stability for simulations of flows in different
regimes containing shocks or discontinuities over any geometry.
Numerical results are presented for a dam-break flow problem over
non-flat bed using different solvers for the shallow water equations.
The robustness of the proposed method is investigated using different
numbers of loops, sensitivity parameters, initial samples and location
of observations. The obtained results demonstrate high reliability and
accuracy of the proposed techniques.
Abstract: Low gear noise is an important sound quality feature in modern passenger cars. Annoying gear noise from the gearbox is influenced by the gear design, gearbox shaft layout, manufacturing deviations in the components, assembly errors and the mounting arrangement of the complete gearbox. Geometrical deviations in the form of profile and lead errors are often present on the flanks of the inspected gears. Ghost frequencies of a gear are very challenging to identify in standard gear measurement and analysis process due to small wavelengths involved. In this paper, gear whine noise occurring at non-integral multiples of gear mesh frequency of passenger car gearbox is investigated and the root cause is identified using the displacement deviation analysis (DDA) method. DDA method is applied to identify ghost frequency excitations on the flanks of gears arising out of generation grinding. Frequency identified through DDA correlated with the frequency of vibration and noise on the end-of-line machine as well as vehicle level measurements. With the application of DDA method along with standard lead profile measurement, gears with ghost frequency geometry deviations were identified on the production line to eliminate defective parts and thereby eliminate ghost frequency noise from a vehicle. Further, displacement deviation analysis can be used in conjunction with the manufacturing process simulation to arrive at suitable countermeasures for arresting the ghost frequency.