Search results for: temporal convolutional net

Authors: Masood Varshosaz, Kamyar Hasanpour

Abstract:

In recent years, we have seen growing interest in recognizing humans in drone images for post-disaster search and rescue operations. Deep learning algorithms have shown great promise in this area, but they often require large amounts of labeled data to train the models. To keep the data acquisition cost low, augmentation techniques can be used to create additional data from existing images. There are many techniques of such that can help generate variations of an original image to improve the performance of deep learning algorithms. While data augmentation is potentially assumed to improve the accuracy and robustness of the models, it is important to ensure that the performance gains are not outweighed by the additional computational cost or complexity of implementing the techniques. To this end, it is important to evaluate the impact of data augmentation on the performance of the deep learning models. In this paper, we evaluated the most currently available 2D data augmentation techniques on a standard convolutional network which was trained for recognizing humans in drone images. The techniques include rotation, scaling, random cropping, flipping, shifting, and their combination. The results showed that the augmented models perform 1-3% better compared to a base network. However, as the augmented images only contain the human parts already visible in the original images, a new data augmentation approach is needed to include the invisible parts of the human body. Thus, we suggest a new method that employs simulated 3D human models to generate new data for training the network.

Keywords: human recognition, deep learning, drones, disaster mitigation

Procedia PDF Downloads 76

Authors: Muhammad Arslan Usman, Muhammad Rehan Usman, Soo Young Shin

Abstract:

Live video streaming is one of the most widely used service among end users, yet it is a big challenge for the network operators in terms of quality. The only way to provide excellent Quality of Experience (QoE) to the end users is continuous monitoring of live video streaming. For this purpose, there are several objective algorithms available that monitor the quality of the video in a live stream. Subjective tests play a very important role in fine tuning the results of objective algorithms. As human perception is considered to be the most reliable source for assessing the quality of a video stream, subjective tests are conducted in order to develop more reliable objective algorithms. Temporal impairments in a live video stream can have a negative impact on the end users. In this paper we have conducted subjective evaluation tests on a set of video sequences containing temporal impairment known as frame freezing. Frame Freezing is considered as a transmission error as well as a hardware error which can result in loss of video frames on the reception side of a transmission system. In our subjective tests, we have performed tests on videos that contain a single freezing event and also for videos that contain multiple freezing events. We have recorded our subjective test results for all the videos in order to give a comparison on the available No Reference (NR) objective algorithms. Finally, we have shown the performance of no reference algorithms used for objective evaluation of videos and suggested the algorithm that works better. The outcome of this study shows the importance of QoE and its effect on human perception. The results for the subjective evaluation can serve the purpose for validating objective algorithms.

Keywords: objective evaluation, subjective evaluation, quality of experience (QoE), video quality assessment (VQA)

Procedia PDF Downloads 588

Authors: Kazım Kaba, Erdem Erdi, M. Akif Erdoğan, H. Mustafa Kandırmaz

Abstract:

Rainfall is crucial data source for very different discipline such as agriculture, hydrology and climate. Therefore rain rate should be known well both spatial and temporal for any area. Rainfall is measured by using rain-gauge at meteorological ground stations traditionally for many years. At the present time, rainfall products are acquired from radar and satellite images with a temporal and spatial continuity. In this study, we investigated the accuracy of these rainfall data according to rain-gauge data. For this purpose, we used Adana-Hatay radar hourly total precipitation product (RN1) and Meteosat convective rainfall rate (CRR) product over Seyhan plain. We calculated daily rainfall values from RN1 and CRR hourly precipitation products. We used the data of rainy days of four stations located within range of the radar from October 2013 to November 2015. In the study, we examined two rainfall data over Seyhan plain and the correlation between the rain-gauge data and two raster rainfall data was observed lowly.

Keywords: meteosat, radar, rainfall, rain-gauge, Turkey

Procedia PDF Downloads 307

Authors: A. Cilek, S. Berberoglu, A. Akin Tanriover, C. Donmez

Abstract:

Wetlands are the areas which have important effects and functions on protecting human life, adjust to nature, and biological variety, besides being potential exploitation sources. Observing the changes in these sensitive areas is important to study for data collecting and correct planning for the future. Remote sensing and Geographic Information System are being increasingly used for environmental studies such as biotope mapping and habitat monitoring. Akyatan Lagoon, one of the most important wetlands in Turkey, has been facing serious threats from agricultural applications in recent years. In this study, seasonal and temporal monitoring in wetlands system are determined by using remotely sensed data and Geographic Information Systems (GIS) between 1985 and 2015. The research method is based on classifying and mapping biotopes in the study area. The natural biotope types were determined as coastal sand dunes, salt marshes, river beds, coastal woods, lakes, lagoons.

Keywords: biotope mapping, GIS, remote sensing, wetlands

Procedia PDF Downloads 374

Authors: Sophia Shi

Abstract:

Acute kidney injury (AKI) is the sudden onset of kidney damage in which the kidneys cannot filter waste from the blood, requiring emergency hospitalization. AKI patient mortality rate is high in the ICU and is virtually impossible for doctors to predict because it is so unexpected. Currently, there is no hybrid model predicting AKI that takes advantage of two types of data. De-identified patient data from the MIMIC-III database and de-identified kidney images and corresponding patient records from the Beijing Hospital of the Ministry of Health were collected. Using data features including serum creatinine among others, two numeric models using MIMIC and Beijing Hospital data were built, and with the hospital ultrasounds, an image-only model was built. Convolutional neural networks (CNN) were used, VGG and Resnet for numeric data and Resnet for image data, and they were combined into a hybrid model by concatenating feature maps of both types of models to create a new input. This input enters another CNN block and then two fully connected layers, ending in a binary output after running through Softmax and additional code. The hybrid model successfully predicted AKI and the highest AUROC of the model was 0.953, achieving an accuracy of 90% and F1-score of 0.91. This model can be implemented into urgent clinical settings such as the ICU and aid doctors by assessing the risk of AKI shortly after the patient’s admission to the ICU, so that doctors can take preventative measures and diminish mortality risks and severe kidney damage.

Keywords: Acute kidney injury, Convolutional neural network, Hybrid deep learning, Patient record data, ResNet, Ultrasound kidney images, VGG

Procedia PDF Downloads 116

Authors: Zheng Wang, Zhenhong Li, Jon Mills

Abstract:

Ground-based synthetic aperture radar (GBSAR) represents a powerful remote sensing tool for deformation monitoring towards various geohazards, e.g. landslides, mudflows, avalanches, infrastructure failures, and the subsidence of residential areas. Unlike spaceborne SAR with a fixed revisit period, GBSAR data can be acquired with an adjustable temporal resolution through either continuous or discontinuous operation. However, challenges arise from processing high temporal-resolution continuous GBSAR data, including the extreme cost of computational random-access-memory (RAM), the delay of displacement maps, and the loss of temporal evolution. Moreover, repositioning errors between discontinuous campaigns impede the accurate measurement of surface displacements. Therefore, a versatile package with two complete chains is developed in this study in order to process both continuous and discontinuous GBSAR data and address the aforementioned issues. The first chain is based on a small-baseline subset concept and it processes continuous GBSAR images unit by unit. Images within a window form a basic unit. By taking this strategy, the RAM requirement is reduced to only one unit of images and the chain can theoretically process an infinite number of images. The evolution of surface displacements can be detected as it keeps temporarily-coherent pixels which are present only in some certain units but not in the whole observation period. The chain supports real-time processing of the continuous data and the delay of creating displacement maps can be shortened without waiting for the entire dataset. The other chain aims to measure deformation between discontinuous campaigns. Temporal averaging is carried out on a stack of images in a single campaign in order to improve the signal-to-noise ratio of discontinuous data and minimise the loss of coherence. The temporal-averaged images are then processed by a particular interferometry procedure integrated with advanced interferometric SAR algorithms such as robust coherence estimation, non-local filtering, and selection of partially-coherent pixels. Experiments are conducted using both synthetic and real-world GBSAR data. Displacement time series at the level of a few sub-millimetres are achieved in several applications (e.g. a coastal cliff, a sand dune, a bridge, and a residential area), indicating the feasibility of the developed GBSAR data processing package for deformation monitoring of a wide range of scientific and practical applications.

Keywords: ground-based synthetic aperture radar, interferometry, small baseline subset algorithm, deformation monitoring

Procedia PDF Downloads 145

Authors: Melissa C. Ingram

Abstract:

The use of new literacies within science classrooms needs to be balanced by teachers to both teach different forms of communication while assessing content area proficiency. Using new literacies such as Twitter and Facebook needs to be incorporated into science content area literacy studies in addition to continuing to use generally-accepted forms of scientific content area presentation, which include scientific papers and textbooks. The research question this literature review seeks to answer is “What are some ways in which new forms of literacy are better suited to teach scientific content area literacy to 21st Century learners?” The research question is addressed through a literature review that highlights methods currently being used to educate the next wave of learners in the world of science content area literacy. Both temporal discourse analysis (TDA) and critical discourse analysis (CDA) were used to determine the need to use new literacies to teach science content area literacy. Increased use of digital technologies and a change in science content area pedagogy were explored.

Keywords: science content area literacy, new literacies, critical discourse analysis, temporal discourse analysis

Procedia PDF Downloads 210

Authors: Javad Abdi, Azam Famil Khalili

Abstract:

Modeling emotions have attracted much attention in recent years, both in cognitive psychology and design of artificial systems. However, it is a negative factor in decision-making; emotions have shown to be a strong faculty for making fast satisfying decisions. In this paper, we have adapted a computational model based on the limbic system in the mammalian brain for control engineering applications. Learning in this model based on Temporal Difference (TD) Learning, we applied the proposed controller (termed BELBIC) for a simple model of a submarine. The model was supposed to reach the desired depth underwater. Our results demonstrate excellent control action, disturbance handling, and system parameter robustness for TDBELBIC. The proposal method, regarding the present conditions, the system action in the part and the controlling aims, can control the system in a way that these objectives are attained in the least amount of time and the best way.

Keywords: artificial neural networks, temporal difference, brain emotional learning based intelligent controller, heating- ventilating and air conditioning

Procedia PDF Downloads 422

Authors: Mannan Malik Muhammad Naeem, Jeong Myung Yung

Abstract:

Functional near infrared spectroscopy (fNIRS) has a favorable position in non-invasive brain imaging techniques. The concentration change of oxygenated hemoglobin and de-oxygenated hemoglobin during particular cognitive activity is the basis for this neuro-imaging modality. Two wavelengths of near-infrared light can be used with modified Beer-Lambert law to explain the indirect status of neuronal activity inside brain. The temporal resolution of fNIRS is very good for real-time brain computer-interface applications. The portability, low cost and an acceptable temporal resolution of fNIRS put it on a better position in neuro-imaging modalities. In this study, an optimization model for impulse response function has been used to estimate/predict initial dip using fNIRS data. In addition, the activity strength parameter related to motor based cognitive task has been analyzed. We found an initial dip that remains around 200-300 millisecond and better localize neural activity.

Keywords: fNIRS, brain-computer interface, optimization algorithm, adaptive signal processing

Procedia PDF Downloads 208

Authors: Patrizia Bonaventura

Abstract:

The aim of this study is to analyze the influence of prosody on the acquisition of temporal aspects of V-V anticipatory lingual coarticulation in productions by Italian-speaking children. Two twin 7-years old male children, native Italian speakers, interacted with the same adult, repeating nonsense disyllables containing VtV sequences where V1 = {i, a} and V2 = {a,e, i, o,u}, with different stress patterns (e.g. pi’ta, pi’ta). The duration of the VC F2 transitions and the CV/VC F2 transitions durations ratios in different V2 contexts and stress conditions were measured by spectrographic analysis and compared between pronunciations by each child vs. the adult to test whether the child was able to imitate the duration of the transitions as produced by the adult in different stress conditions. Consequences highlighted a significant difference in durations of VC transitions between children and adult: longer VC transitions durations, indicating a greater amount of coarticulation, were found for one child in every context, and for the other, only in stressed [it] sequences. The data support the hypothesis of the presence of different temporal patterns of anticipatory coarticulation in adults and children, and of a greater amount of coarticulation in children, with different strategies of implementation across different prosodic conditions.

Keywords: speech acquisition, coarticulation, Italian language, prosody

Procedia PDF Downloads 72

Authors: Gholba Niranjan Dilip, Anil Kumar

Abstract:

Conduct of accurate reliable mapping of oil palm plantations and census of individual palm trees is a huge challenge. This study addresses this challenge and developed an optimized solution implemented deep learning techniques on remote sensing data. The oil palm is a very important tropical crop. To improve its productivity and land management, it is imperative to have accurate census over large areas. Since, manual census is costly and prone to approximations, a methodology for automated census using panchromatic images from Cartosat-2, SkySat and World View-3 satellites is demonstrated. It is selected two different study sites in Indonesia. The customized set of training data and ground-truth data are created for this study from Cartosat-2 images. The pre-trained model of Single Shot MultiBox Detector (SSD) Lite MobileNet V2 Convolutional Neural Network (CNN) from the TensorFlow Object Detection API is subjected to transfer learning on this customized dataset. The SSD model is able to generate the bounding boxes for each oil palm and also do the counting of palms with good accuracy on the panchromatic images. The detection yielded an F-Score of 83.16 % on seven different images. The detections are buffered and dissolved to generate polygons demarcating the boundaries of the oil palm plantations. This provided the area under the plantations and also gave maps of their location, thereby completing the automated census, with a fairly high accuracy (≈100%). The trained CNN was found competent enough to detect oil palm crowns from images obtained from multiple satellite sensors and of varying temporal vintage. It helped to estimate the increase in oil palm plantations from 2014 to 2021 in the study area. The study proved that high-resolution panchromatic satellite image can successfully be used to undertake census of oil palm plantations using CNNs.

Keywords: object detection, oil palm tree census, panchromatic images, single shot multibox detector

Procedia PDF Downloads 149

Authors: Shweta Singh, Sudaman Katti

Abstract:

The most well-known sequence models make use of complex recurrent neural networks in an encoder-decoder configuration. The model used in this research makes use of a transformer, which is based purely on a self-attention mechanism, without relying on recurrence at all. More specifically, encoders and decoders which make use of self-attention and operate based on a memory, are used. In this research work, results for various 3D visual and non-visual reinforcement learning tasks designed in Unity software were obtained. Convolutional neural networks, more specifically, nature CNN architecture, are used for input processing in visual tasks, and comparison with standard long short-term memory (LSTM) architecture is performed for both visual tasks based on CNNs and non-visual tasks based on coordinate inputs. This research work combines the transformer architecture with the proximal policy optimization technique used popularly in reinforcement learning for stability and better policy updates while training, especially for continuous action spaces, which are used in this research work. Certain tasks in this paper are long horizon tasks that carry on for a longer duration and require extensive use of memory-based functionalities like storage of experiences and choosing appropriate actions based on recall. The transformer, which makes use of memory and self-attention mechanism in an encoder-decoder configuration proved to have better performance when compared to LSTM in terms of exploration and rewards achieved. Such memory based architectures can be used extensively in the field of cognitive robotics and reinforcement learning.

Keywords: convolutional neural networks, reinforcement learning, self-attention, transformers, unity

Procedia PDF Downloads 108

Authors: Farina Riaz, Shahab Abdulla, Srinjoy Ganguly, Hajime Suzuki, Ravinesh C. Deo, Susan Hopkins

Abstract:

Recently, quantum machine learning has received significant attention. For various types of data, including text and images, numerous quantum machine learning (QML) models have been created and are being tested. Images are exceedingly complex data components that demand more processing power. Despite being mature, classical machine learning still has difficulties with big data applications. Furthermore, quantum technology has revolutionized how machine learning is thought of, by employing quantum features to address optimization issues. Since quantum hardware is currently extremely noisy, it is not practicable to run machine learning algorithms on it without risking the production of inaccurate results. To discover the advantages of quantum versus classical approaches, this research has concentrated on colored image data. Deep learning classification models are currently being created on Quantum platforms, but they are still in a very early stage. Black and white benchmark image datasets like MNIST and Fashion MINIST have been used in recent research. MNIST and CIFAR-10 were compared for binary classification, but the comparison showed that MNIST performed more accurately than colored CIFAR-10. This research will evaluate the performance of the QML algorithm on the colored benchmark dataset CIFAR-10 to advance QML's real-time applicability. However, deep learning classification models have not been developed to compare colored images like Quantum Convolutional Neural Network (QCNN) to determine how much it is better to classical. Only a few models, such as quantum variational circuits, take colored images. The methodology adopted in this research is a hybrid approach by using penny lane as a simulator. To process the 10 classes of CIFAR-10, the image data has been translated into grey scale and the 28 × 28-pixel image containing 10,000 test and 50,000 training images were used. The objective of this work is to determine how much the quantum approach can outperform a classical approach for a comprehensive dataset of color images. After pre-processing 50,000 images from a classical computer, the QCNN model adopted a hybrid method and encoded the images into a quantum simulator for feature extraction using quantum gate rotations. The measurements were carried out on the classical computer after the rotations were applied. According to the results, we note that the QCNN approach is ~12% more effective than the traditional classical CNN approaches and it is possible that applying data augmentation may increase the accuracy. This study has demonstrated that quantum machine and deep learning models can be relatively superior to the classical machine learning approaches in terms of their processing speed and accuracy when used to perform classification on colored classes.

Keywords: CIFAR-10, quantum convolutional neural networks, quantum deep learning, quantum machine learning

Procedia PDF Downloads 106

Authors: Gizem Kodak, Birsen Koldemir

Abstract:

The Turkish Straits are at the crossroads of Europe and Asia continents and are unique waterways connecting the Black Sea countries to the rest of the world. Because of the geostrategic value of the location, passage of trade and war ships through the Turkish Straits has become a vital attraction and importance for the great powers and the riparian states throughout the history. This study contains a temporal analysis of the legal measures implemented in the Turkish Straits System. In this context, the historical alternation of the Turkish Straits has been examined, taking into account the relevant national and international regulations. In other words, relevant national and international regulations have been examined in this study according to historical time schedules. Parallel to the main concept mentioned above, the first chapter focuses on international regulations. These arrangements are organized according to date order and in three subheadings: Sèvres Treaty (1920), Lausanne Treaty (1923) and Montreux Convention (1936). Another topic, the national regulations, has been examined under five subheadings. These; (1982), Port Regulations of Canakkale (1982), Marine Traffic Regulations of the Turkish Straits and Marmara Region (1994) and Maritime Traffic Regulations for the Turkish Straits (1998). In doing so, the aim was to identify the differences in legal arrangements throughout the time regarding the navigation through the Turkish Straits. The current situation of the Turkish Straits has been presented in detail in the last part of the work, taking Montreux Convention into consideration. In this context, the articles of the Convention which regulate the passage of trade vessels have been examined from two perspectives; Peace time and war time. As for the measures that can be implemented in time of war, three options put forward depending on Turkey's stance: ‘Turkey not being belligerent’, ‘Turkey being belligerent’ and ‘situation in which Turkey considers herself threatened with imminent danger of war’.

Keywords: temporal analysis, maritime law, Turkish straits, maritime accidents

Procedia PDF Downloads 140

Authors: Hicham Maklache, Rajae Delhi, Fatiha Benzha, Mohamed Tahiri

Abstract:

In Morocco, where semi-arid climate is dominant, the supply of industrial and drink water is provided primarily by surface water. Morocco has currently 118 multi-purpose dams. If the construction of these works was a necessity to ensure in all seasons, the water essential to our country, it is impartial to control and protect the quality of running water. -Most dam reservoir used are threatened by eutrophication due to increased terrigenous and anthropogenic pollutants, coming from an over-fertilization of water by phosphorus and nitrogen nutrients and accelerated by uncontrolled development of microalgae aging. It should also be noted that the daily practices of citizens with respect to the resource, an essential component involved in almost all human activities (agriculture, agro-industries, hydropower, ...), has contributed significantly to the deterioration of water quality despite its treatment in several plants. Therefore, the treated water, provides a legacy of bad tastes and odors unacceptable to the consumer. -The present work exhibits results of water quality watershed Oum Erbia used to supply drinking water to the whole terraced area connecting the city of Khenifra to the city of Azemmour. The area south west of Great Casablanca (metropolis of the kingdom with about 4 million inhabitants) supplied 50% of its water needs by sourcing Dam Sidi Said Maachou located, last anchor point of the watershed before the spill in the Atlantic Ocean. The results were performed in a spatio-temporal scale and helped to establish a history of monitoring water quality during the 2009-2011 cycles, the study also presents the development of quality according to the seasonal rhythmicity and rainfall. It gives also an overview on the concept of watershed stewardship.

Keywords: crude surface water quality, Oum Er Rbia hydraulic basin, spatio-temporal monitoring, Great Casablanca drink water quality, Morocco

Procedia PDF Downloads 427

Authors: Modupe O. Jimoh

Abstract:

During manhole overflows, its contents pollute the immediate environment. Understanding the pollutant transfer characteristics between manhole’s incoming sewer and the overflow is therefore of great importance. A square manhole with sides 388 mm by 388 mm and height 700 mm with an overflow facility was used in the laboratory to carry out overflow concentration measurements. Two scenarios were investigated using three flow rates. The first scenario corresponded to when the exit of the pipe becomes blocked and the only exit for the flow is the manhole. The second scenario is when there is an overflow in combination with a pipe exit. The temporal concentration measurements showed that the peak concentration of pollutants in the flow was attenuated between the inlet and the overflow. A deconvolution software was used to predict the Residence time distribution (RTD) and consequently the Cumulative Residence time distribution (CRTD). The CRTDs suggest that complete mixing is occurring between the pipe inlet and the overflow, like what is obtained in a low surcharged manhole. The results also suggest that an instantaneous stirred tank reactor model can describe the mixing characteristics.

Keywords: CRTDs, instantaneous stirred tank reactor model, overflow, square manholes, surcharge, temporal concentration profiles

Procedia PDF Downloads 121

Authors: Mohammad Reza Boromand, Masoud Moradi, Amin Eskandari

Abstract:

The present study attempts to determine ranking of managerial parameters impacting upon performance of football referees in Iran. The population consisted of all referees in Leagues 1, 2 and 3 as well as super league of Iran (N=273), of which we selected 160 referees and assistant referees in 2013-2014. A research-designed questionnaire was used for data collection which was divided into two sections: (1) Demographic details (age range, Marital status, employment, refereeing experience, education level, refereeing level and proficiency) and (2) items related to parameters impacting upon performance of referees (structural parameters, operational parameters, environmental parameters, temporal parameters, economic parameters, facilities and tools, personal performance and performance evaluation). Internal consistency was calculated by Cronbach's alpha (r=0.85). For data analysis, we performed Freedman's Test and used SPSS software (α>0.05), along with descriptive statistics. The findings showed the following ranking for the above-mentioned managerial parameters: Facilities and tools, personal performance, economic parameters, structural parameters, operational parameters, environmental parameters, temporal parameters, and performance evaluation.

Keywords: Iran, football referees, managerial parameters, performance

Procedia PDF Downloads 552

Authors: Rubab Z. Kahlon, Ibtisam Butt, Isma Younis, Aamer G. Mufti

Abstract:

Worldwide lung cancer 1.61 million cases were seen in both genders. Lung carcinoma is the major cause of both morbidity and mortality in the world. Purpose of the present study was to describe the spatio- temporal trends of lung cancer in both genders. A retrospective study was conducted. Total 1498 patients of lung carcinoma were examined. Only lung cancer patients from all over the Punjab were included in the present study. MS Excel 2010 was used for data tabulation and calculation while the Arc GIS version 9.3 was used for geographical representation of the data. 1498 cases of Lung cancer were found from 2008-2012. The number of male patients was 1236 and female was 262. Majority of the patients were from Lahore districts with 807 patients. Lung cancer was more prevalent in male as compared to female in our region. Increase in the prevalence of lung cancer was prominently seen in the most populated and industrial areas of the Punjab province. Time trend of five years showed fluctuation in the lung cancer incidence during the study period.

Keywords: districts, gender, lung cancer trends, Punjab province of Pakistan

Procedia PDF Downloads 512

Authors: Yuanchao Sang, Zhihao Gong, Longsheng Chen, Long Chen

Abstract:

Clothing fashion represents a human’s aesthetic appreciation towards everyday outfits and appetite for fashion, and it reflects the development of status in society, humanity, and economics. However, modelling fashion by machine is extremely challenging because fashion is too abstract to be efficiently described by machines. Even human beings can hardly reach a consensus about fashion. In this paper, we are dedicated to answering a fundamental fashion-related problem: what image feature best describes clothing fashion? To address this issue, we have designed and evaluated various image features, ranging from traditional low-level hand-crafted features to mid-level style awareness features to various current popular deep neural network-based features, which have shown state-of-the-art performance in various vision tasks. In summary, we tested the following 9 feature representations: color, texture, shape, style, convolutional neural networks (CNNs), CNNs with distance metric learning (CNNs&DML), AutoEncoder, CNNs with multiple layer combination (CNNs&MLC) and CNNs with dynamic feature clustering (CNNs&DFC). Finally, we validated the performance of these features on two publicly available datasets. Quantitative and qualitative experimental results on both intra-domain and inter-domain fashion clothing image retrieval showed that deep learning based feature representations far outweigh traditional hand-crafted feature representation. Additionally, among all deep learning based methods, CNNs with explicit feature clustering performs best, which shows feature clustering is essential for discriminative fashion feature representation.

Keywords: convolutional neural network, feature representation, image processing, machine modelling

Procedia PDF Downloads 123

Authors: Muhammad Rehan Usman, Muhammad Arslan Usman, Soo Young Shin

Abstract:

The new era of digital communication has brought up many challenges that network operators need to overcome. The high demand of mobile data rates require improved networks, which is a challenge for the operators in terms of maintaining the quality of experience (QoE) for their consumers. In live video transmission, there is a sheer need for live surveillance of the videos in order to maintain the quality of the network. For this purpose objective algorithms are employed to monitor the quality of the videos that are transmitted over a network. In order to test these objective algorithms, subjective quality assessment of the streamed videos is required, as the human eye is the best source of perceptual assessment. In this paper we have conducted subjective evaluation of videos with varying spatial and temporal impairments. These videos were impaired with frame freezing distortions so that the impact of frame freezing on the quality of experience could be studied. We present subjective Mean Opinion Score (MOS) for these videos that can be used for fine tuning the objective algorithms for video quality assessment.

Keywords: frame freezing, mean opinion score, objective assessment, subjective evaluation

Procedia PDF Downloads 474

Authors: Bahareh Golchin, Nooshin Riahi

Abstract:

One of the most significant issues as attended a lot in recent years is that of recognizing the sentiments and emotions in social media texts. The analysis of sentiments and emotions is intended to recognize the conceptual information such as the opinions, feelings, attitudes and emotions of people towards the products, services, organizations, people, topics, events and features in the written text. These indicate the greatness of the problem space. In the real world, businesses and organizations are always looking for tools to gather ideas, emotions, and directions of people about their products, services, or events related to their own. This article uses the Twitter social network, one of the most popular social networks with about 420 million active users, to extract data. Using this social network, users can share their information and opinions about personal issues, policies, products, events, etc. It can be used with appropriate classification of emotional states due to the availability of its data. In this study, supervised learning and deep neural network algorithms are used to classify the emotional states of Twitter users. The use of deep learning methods to increase the learning capacity of the model is an advantage due to the large amount of available data. Tweets collected on various topics are classified into four classes using a combination of two Bidirectional Long Short Term Memory network and a Convolutional network. The results obtained from this study with an average accuracy of 93%, show good results extracted from the proposed framework and improved accuracy compared to previous work.

Keywords: emotion classification, sentiment analysis, social networks, deep neural networks

Procedia PDF Downloads 123

Authors: F. Ghazalnaz Sharifonnasabi, Iman Makhdoom

Abstract:

Breast cancer is a serious health concern that affects many people around the world. According to a study published in the Breast journal, the global burden of breast cancer is expected to increase significantly over the next few decades. The number of deaths from breast cancer has been increasing over the years, but the age-standardized mortality rate has decreased in some countries. It’s important to be aware of the risk factors for breast cancer and to get regular check- ups to catch it early if it does occur. Machin learning techniques have been used to aid in the early detection and diagnosis of breast cancer. These techniques, that have been shown to be effective in predicting and diagnosing the disease, have become a research hotspot. In this study, we consider two deep learning approaches including: Multi-Layer Perceptron (MLP), and Convolutional Neural Network (CNN). We also considered the five-machine learning algorithm titled: Decision Tree (C4.5), Naïve Bayesian (NB), Support Vector Machine (SVM), K-Nearest Neighbors (KNN) Algorithm and XGBoost (eXtreme Gradient Boosting) on the Breast Cancer Wisconsin Diagnostic dataset. We have carried out the process of evaluating and comparing classifiers involving selecting appropriate metrics to evaluate classifier performance and selecting an appropriate tool to quantify this performance. The main purpose of the study is predicting and diagnosis breast cancer, applying the mentioned algorithms and also discovering of the most effective with respect to confusion matrix, accuracy and precision. It is realized that CNN outperformed all other classifiers and achieved the highest accuracy (0.982456). The work is implemented in the Anaconda environment based on Python programing language.

Keywords: breast cancer, multi-layer perceptron, Naïve Bayesian, SVM, decision tree, convolutional neural network, XGBoost, KNN

Procedia PDF Downloads 56

Authors: Anastasiya Paltarzhitskaya

Abstract:

Subjective time plays an important role in consciousness processes and self-awareness at the moment. The concept of intrinsic neural timescales (INT) explains the difference in perceiving various time intervals. The capacity to experience the present builds on the fundamental properties of temporal cognition. The challenge that both philosophy and neuroscience try to answer is how the brain differentiates the present from the past and future. In our work, we analyze papers which describe mechanisms involved in the perception of ‘present’ and ‘non-present’, i.e., future and past moments. Taking into account that we perceive time intervals even during rest or relaxation, we suppose that the default-mode network activity can code time features, including the present moment. We can compare some results of time perceptual studies, where brain activity was shown in states with different flows of time, including resting states and during “mental time travel”. According to the concept of mental traveling, we employ a range of scenarios which demand episodic memory. However, some papers show that the hippocampal region does not activate during time traveling. It is a controversial result that is further complicated by the phenomenological aspect that includes a holistic set of information about the individual’s past and future.

Keywords: temporal consciousness, time perception, memory, present

Procedia PDF Downloads 55

Authors: Stefan K. Behfar

Abstract:

The rapid growth of the Ethereum network has brought forth the urgent need for scalable analysis methods to handle the increasing volume of blockchain data. In this research, we propose efficient methodologies for making Ethereum network analysis scalable. Our approach leverages a combination of graph-based data representation, probabilistic sampling, and parallel processing techniques to achieve unprecedented scalability while preserving critical network insights. Data Representation: We develop a graph-based data representation that captures the underlying structure of the Ethereum network. Each block transaction is represented as a node in the graph, while the edges signify temporal relationships. This representation ensures efficient querying and traversal of the blockchain data. Probabilistic Sampling: To cope with the vastness of the Ethereum blockchain, we introduce a probabilistic sampling technique. This method strategically selects a representative subset of transactions and blocks, allowing for concise yet statistically significant analysis. The sampling approach maintains the integrity of the network properties while significantly reducing the computational burden. Graph Convolutional Networks (GCNs): We incorporate GCNs to process the graph-based data representation efficiently. The GCN architecture enables the extraction of complex spatial and temporal patterns from the sampled data. This combination of graph representation and GCNs facilitates parallel processing and scalable analysis. Distributed Computing: To further enhance scalability, we adopt distributed computing frameworks such as Apache Hadoop and Apache Spark. By distributing computation across multiple nodes, we achieve a significant reduction in processing time and enhanced memory utilization. Our methodology harnesses the power of parallelism, making it well-suited for large-scale Ethereum network analysis. Evaluation and Results: We extensively evaluate our methodology on real-world Ethereum datasets covering diverse time periods and transaction volumes. The results demonstrate its superior scalability, outperforming traditional analysis methods. Our approach successfully handles the ever-growing Ethereum data, empowering researchers and developers with actionable insights from the blockchain. Case Studies: We apply our methodology to real-world Ethereum use cases, including detecting transaction patterns, analyzing smart contract interactions, and predicting network congestion. The results showcase the accuracy and efficiency of our approach, emphasizing its practical applicability in real-world scenarios. Security and Robustness: To ensure the reliability of our methodology, we conduct thorough security and robustness evaluations. Our approach demonstrates high resilience against adversarial attacks and perturbations, reaffirming its suitability for security-critical blockchain applications. Conclusion: By integrating graph-based data representation, GCNs, probabilistic sampling, and distributed computing, we achieve network scalability without compromising analytical precision. This approach addresses the pressing challenges posed by the expanding Ethereum network, opening new avenues for research and enabling real-time insights into decentralized ecosystems. Our work contributes to the development of scalable blockchain analytics, laying the foundation for sustainable growth and advancement in the domain of blockchain research and application.

Keywords: Ethereum, scalable network, GCN, probabilistic sampling, distributed computing

Procedia PDF Downloads 56

Authors: Zahra Ramezanpanah, Joachim Carvallo, Aurelien Rodriguez

Abstract:

This paper aims to detect anomalies in streaming data using machine learning algorithms. In this regard, we designed two separate pipelines and evaluated the effectiveness of each separately. The first pipeline, based on supervised machine learning methods, consists of two phases. In the first phase, we trained several supervised models using the UNSW-NB15 data-set. We measured the efficiency of each using different performance metrics and selected the best model for the second phase. At the beginning of the second phase, we first, using Argus Server, sniffed a local area network. Several types of attacks were simulated and then sent the sniffed data to a running algorithm at short intervals. This algorithm can display the results of each packet of received data in real-time using the trained model. The second pipeline presented in this paper is based on unsupervised algorithms, in which a Temporal Graph Network (TGN) is used to monitor a local network. The TGN is trained to predict the probability of future states of the network based on its past behavior. Our contribution in this section is introducing an indicator to identify anomalies from these predicted probabilities.

Keywords: temporal graph network, anomaly detection, cyber security, IDS

Procedia PDF Downloads 87

Authors: Naif Albelwi, Alan Kwan, Yacine Rezgui

Abstract:

Cities with rapid growth face tremendous challenges not only to provide services to meet this growth but also to assure that this growth occurs in a sustainable way. The consumption of material, energy, and water resources is inextricably linked to population growth with a unique impact in urban areas, especially in light of significant investments in infrastructure to support urban development. Urban Metabolism (UM) is becoming popular as it provides a framework accounting the mass and energy flows through a city. The objective of this study is to determine the energy and material flows of Riyadh, Saudi Arabia using locally generated data from 1996 and 2012 and analyzing the temporal trends of energy and material flows. Preliminary results show that while the population of Riyadh grew 90% since 1996, the input and output flows have increased at higher rate. Results also show increasing in energy mobile consumption from 61k TJ in 1996 to 157k TJ in 2012 which points to Riyadh’s inefficient urban form. The study findings highlight the importance to develop effective policies for improving the use of resources.

Keywords: energy and water consumption, sustainability, urban development, urban metabolism

Procedia PDF Downloads 257

Authors: Oumaima Khlifati, Khadija Baba

Abstract:

Pavement distress is the main factor responsible for the deterioration of road structure durability, damage vehicles, and driver comfort. Transportation agencies spend a high proportion of their funds on pavement monitoring and maintenance. The auscultation of pavement distress was based on the manual survey, which was extremely time consuming, labor intensive, and required domain expertise. Therefore, the automatic distress detection is needed to reduce the cost of manual inspection and avoid more serious damage by implementing the appropriate remediation actions at the right time. Inspired by recent deep learning applications, this paper proposes an algorithm for automatic road distress detection and classification using on the Deep Convolutional Neural Network (DCNN). In this study, the types of pavement distress are classified as transverse or longitudinal cracking, alligator, pothole, and intact pavement. The dataset used in this work is composed of public asphalt pavement images. In order to learn the structure of the different type of distress, the DCNN models are trained and tested as a multi-label classification task. In addition, to get the highest accuracy for our model, we adjust the structural optimization hyper parameters such as the number of convolutions and max pooling, filers, size of filters, loss functions, activation functions, and optimizer and fine-tuning hyper parameters that conclude batch size and learning rate. The optimization of the model is executed by checking all feasible combinations and selecting the best performing one. The model, after being optimized, performance metrics is calculated, which describe the training and validation accuracies, precision, recall, and F1 score.

Keywords: distress pavement, hyperparameters, automatic classification, deep learning

Procedia PDF Downloads 68

Authors: Asmamaw Yehun, Tsegaye Gogie, Martin Vermeer, Addisu Hunegnaw

Abstract:

The Global Positioning System (GPS) is a space-based radio positioning system, which is capable of providing continuous position, velocity, and time information to users anywhere on or near the surface of the Earth. The main objective of this work was to estimate the integrated precipitable water vapour (IPWV) using ground GPS and Low Earth Orbit (LEO) Radio Occultation (RO) to study spatial-temporal variability. For LEO-GPS RO, we used Constellation Observing System for Meteorology, Ionosphere, and Climate (COSMIC) datasets. We estimated the daily and monthly mean of IPWV using six selected ground-based GPS stations over a period of range from 2012 to 2016 (i.e. five-years period). The main perspective for selecting the range period from 2012 to 2016 is that, continuous data were available during these periods at all Ethiopian GPS stations. We studied temporal, seasonal, diurnal, and vertical variations of precipitable water vapour using GPS observables extracted from the precise geodetic GAMIT-GLOBK software package. Finally, we determined the cross-correlation of our GPS-derived IPWV values with those of the European Centre for Medium-Range Weather Forecasts (ECMWF) ERA-40 Interim reanalysis and of the second generation National Oceanic and Atmospheric Administration (NOAA) model ensemble Forecast System Reforecast (GEFS/R) for validation and static comparison. There are higher values of the IPWV range from 30 to 37.5 millimetres (mm) in Gambela and Southern Regions of Ethiopia. Some parts of Tigray, Amhara, and Oromia regions had low IPWV ranges from 8.62 to 15.27 mm. The correlation coefficient between GPS-derived IPWV with ECMWF and GEFS/R exceeds 90%. We conclude that there are highly temporal, seasonal, diurnal, and vertical variations of precipitable water vapour in the study area.

Keywords: GNSS, radio occultation, atmosphere, precipitable water vapour

Procedia PDF Downloads 65

Authors: Nkechinyere Joy Olawuyi, Babajide Samuel Afolabi, Bola Ibitoye

Abstract:

This study collected a preprocessed dataset of chest radiographs and formulated a deep neural network model for detecting abnormalities. It also evaluated the performance of the formulated model and implemented a prototype of the formulated model. This was with the view to developing a deep neural network model to automatically classify abnormalities in chest radiographs. In order to achieve the overall purpose of this research, a large set of chest x-ray images were sourced for and collected from the CheXpert dataset, which is an online repository of annotated chest radiographs compiled by the Machine Learning Research Group, Stanford University. The chest radiographs were preprocessed into a format that can be fed into a deep neural network. The preprocessing techniques used were standardization and normalization. The classification problem was formulated as a multi-label binary classification model, which used convolutional neural network architecture to make a decision on whether an abnormality was present or not in the chest radiographs. The classification model was evaluated using specificity, sensitivity, and Area Under Curve (AUC) score as the parameter. A prototype of the classification model was implemented using Keras Open source deep learning framework in Python Programming Language. The AUC ROC curve of the model was able to classify Atelestasis, Support devices, Pleural effusion, Pneumonia, A normal CXR (no finding), Pneumothorax, and Consolidation. However, Lung opacity and Cardiomegaly had a probability of less than 0.5 and thus were classified as absent. Precision, recall, and F1 score values were 0.78; this implies that the number of False Positive and False Negative is the same, revealing some measure of label imbalance in the dataset. The study concluded that the developed model is sufficient to classify abnormalities present in chest radiographs into present or absent.

Keywords: transfer learning, convolutional neural network, radiograph, classification, multi-label

Procedia PDF Downloads 99

Authors: Jaiden Xuan Schraut, Leon Liu, Yiqiao Yin

Abstract:

Computer vision in medical diagnosis has achieved a high level of success in diagnosing diseases with high accuracy. However, conventional classifiers that produce an image to-label result provides insufficient information for medical professionals to judge and raise concerns over the trust and reliability of a model with results that cannot be explained. In order to gain local insight into cancerous regions, separate tasks such as imaging segmentation need to be implemented to aid the doctors in treating patients, which doubles the training time and costs which renders the diagnosis system inefficient and difficult to be accepted by the public. To tackle this issue and drive AI-first medical solutions further, this paper proposes a multi-output network that follows a U-Net architecture for image segmentation output and features an additional convolutional neural networks (CNN) module for auxiliary classification output. Class activation maps are a method of providing insight into a convolutional neural network’s feature maps that leads to its classification but in the case of lung diseases, the region of interest is enhanced by U-net-assisted Class Activation Map (CAM) visualization. Therefore, our proposed model combines image segmentation models and classifiers to crop out only the lung region of a chest X-ray’s class activation map to provide a visualization that improves the explainability and is able to generate classification results simultaneously which builds trust for AI-led diagnosis systems. The proposed U-Net model achieves 97.61% accuracy and a dice coefficient of 0.97 on testing data from the COVID-QU-Ex Dataset which includes both diseased and healthy lungs.

Keywords: multi-output network model, U-net, class activation map, image classification, medical imaging analysis

Procedia PDF Downloads 179