1,721,346 research outputs found
EcRD:Edge-Cloud Computing Framework for Smart Road Damage Detection and Warning
No full textRoad damages have caused numerous fatalities, thus the study of road damage detection, especially hazardous road damage detection and warning is critical for traffic safety. Existing road damage detection systems mainly process data at cloud, which suffers from a high latency caused by long-distance. Meanwhile, supervised machine learning algorithms are usually used in these systems requiring large precisely labeled data sets to achieve a good performance. In this article, we propose EcRD: an edge-cloud-based road damage detection and warning framework, that leverages the fast-responding advantage of edge and the large storage and computation resources advantages of cloud. There are three main contributions in this article: we first propose a simple yet efficient road segmentation algorithm to enable fast and accurate road area detection. Then, a light-weighted road damage detector is developed based on gray level co-occurrence matrix features at edge for rapid hazardous road damage detection and warning. Furthermore, a multitypes road damage detection model is introduced for long-term road management at cloud, embedded with a novel image generator based on cycle-consistent adversarial networks which automatically generates images with labels to further improve road damage detection accuracy. By comparing with the state-of-the-art, we demonstrate that the proposed EcRD can accurately detect both hazardous road damages at edge and multitypes road damages at cloud. Besides, it is around 579 times faster than cloud-based approaches without affecting users' experience and requiring very low storage and labeling cost
Investigation of trace metals in riverine waterways of Bangladesh using multivariate analyses: spatial toxicity variation and potential health risk assessment
No full textMinute quantities of trace metals have delirious effects on the human body causing acute and chronic toxicities. These trace metals have the ability to bind with enzymes and proteins causing an alteration in their activity, and, consequently, their damage. In this study, water samples were collected from five sites in Rupsa River (Bangladesh) during dry and wet seasons aiming to assess the trace metal concentration and the correlated health risk for people leaving in the area. Six trace metals, namely arsenic (Ar), cadmium (Cd), chromium (Cr), copper (Cu), lead (Pb), and nickel (Ni) were measured for further analyzing their spatial and seasonal variations. The measured trace metal concentrations followed this decreasing order: Cr > Pb > As > Cu > Ni > Cd for the dry season, and Cr > Pb > As > Cu > Ni > Cd for the wet season. Among the trace metals, As, Ni, Cr, and Pb exhibited a statistically significative variation throughout the study period. The PCA analysis accounted for 64.5% and 64.4% total variations of the trace metals in dry and wet seasons, respectively. The Euclidean distance of trace metals in water samples across five sites showed significantly different distribution patterns, which were further confirmed by PERMANOVA. Furthermore, CAP model disclosed that trace metals are source-specific: brickfields and sewage effluents were potential sources for Cd, whereas different industries were potential sources for As, Cr, Cu, Ni, and Pb. Correlation analysis showed that Ni and Cr significantly correlated with pH and electrical conductivity. Correlation among the trace metals unveiled that they depended on each other as for their origin, magnitude, and existence in the riverine waterways. As for the health risk assessment, a non-carcinogenic health hazard due to ingestion during regular activities and dermal contact during fishing activity to all kind of people (adult males, adult females, and children) in the studied area was retrieved based on the hazard index (HI) of trace metals, which was higher than recommended value (HI > 1). Moreover, also the carcinogenic risks of Ni and As due to regular activities via ingestion and dermal contact pathways were higher than the standard value (CR > 1.0E-04), suggesting the occurrence of cancer risk to humans in the study area
A decision support model for assessing and prioritization of industry 5.0 cybersecurity challenges
Full text linkThe world is adopting the Industry 5.0 paradigm to increase human centricity, sustainability, and resilience in efficient, optimized, and profitable manufacturing systems. With benefits, however, come increased risks of economic and physical loss, driving the need for continuous improvement of Industry 5.0 cybersecurity. Implementation and advancement of adequate cybersecurity have created challenges that have been identified in the literature. In this study, key Industry 5.0 cybersecurity challenges and related sub-challenges are highlighted based on a literature review. Graph Theory and Matrix Approach (GTMA) is employed to analyze the challenges and determine relative importance based on permanent values of the variable permanent matrix (VPM). The results identify the most important Industry 5.0 cybersecurity challenges and reveal Industry 5.0 firms should primarily concentrate on supply chain vulnerabilities to decrease data loss and hacking in the organization’s supply chain network. This study also recommends that executives and lawmakers acquire knowledge regarding cybersecurity challenges and prepare to deal with them. Addressing these and other subsequently prioritized challenges—the top five rounded out with emergent cybersecurity trends, non-availability of cybersecurity curriculum in education, embedded technical constraints, and absence of skilled employees and training—will lead the methodical development of holistic, robust cybersecurity programs. Firms accepting of this reality may implement such programs to mitigate evolving cyber-risk towards harnessing and sustaining the benefits of novel Industry 5.0 technologies
Potential toxic elements in sediment of some rivers at Giresun, Northeast Turkey: A preliminary assessment for ecotoxicological status and health risk
No full textIslam, Md Saiful/0000-0002-3598-0315WOS: 000523335900072The concentration of globally alarming potential toxic elements (PTEs) like Aluminum (Al), chrome (Cr), manganese (Mn), iron (Fe), cobalt (Co), nickel (Ni), copper (Cu), zinc (Zn), arsenic (As), cadmium (Cd), lead (Pb), and uranium (U) were measured in surface sediment of seven major rivers residing in Giresun (one of the most important Hazelnut production areas of Turkey). The mean concentrations of PTEs in all river sediments showed the descending order of Al > Fe > Mn > Zn > Cu > Pb > Cr > Ni > Co > As > U > Cd. The level of studied metals in most of the rivers exceeded the threshold effect level (TEL), indicating a potential risk to the environment. Certain indices, including the sediment quality guidelines (SQGs), contamination factor (CF), pollution load index (PLI), enrichment factor (EF), potential ecological risk index (E-r(i)), geoaccumulation index (I-geo), toxic risk index (TRI), modified hazard quotient (mHQ) and ecological contamination index (ECI) were used to assess the ecological risk posed by PTEs in sediment. Contamination factor (CF) and geoaccumulation index (I-geo) demonstrated that most of the sediment samples were moderately to considerably contaminated by Cu, As, Cd and Pb. In view of the potential ecological risk index, sediments from Pazarsuyu Stream (PS), Batlama Stream (BS) and Gelevera Stream (GLS) showed considerable potential ecological risk. The sources of PTEs and the relations between them were determined by using principal component analysis/factor analysis (PCA/FA), Hierarchical clustering analysis (HCA) and Pearson correlation index (PCI). Three factors explaining 83.94% of the total variance was found by PCA/FA. 43.34% of the total variance explained by the first factor (F1) was correlated with Ni, Cr, Pb and Co elements. 28.35% of the total variance explained by the second factor (F2) was correlated with U, Al, the third factor (F3) explains 12.24% of the total variance and correlated with Zn, Cd, Cu and As elements. These factors revealed that the quality of the sediment was mainly influenced by anthropogenic effects. The extent of pollution by heavy metals in the studied streams implies that the condition is much frightening to the biota and inhabitants in the vicinity of these rivers as well.Scientific Project Office of Giresun University [FEN-BAP-A-230218-30]This research was financially supported by the Scientific Project Office of Giresun University (FEN-BAP-A-230218-30). We also thank Furkan Saltoglu for the map drawing
Deep Learning for Causal Discovery in Texts
No full textCausality detection in text data is a challenging natural language processing task. This is a trivial task for human beings as they acquire vast background knowledge throughout their lifetime. For example, a human knows from their experience that heavy rain may cause flood or plane accidents may cause death. However, it is challenging to automatically detect such causal relationships in texts due to the availability of limited contextual information and the unstructured nature of texts. The task is even more challenging for social media short texts such as Tweets as often they are informal, short, and grammatically incorrect. Generating hand-crafted linguistic rules is an option but is not always effective to detect causal relationships in text because they are rigid and require grammatically correct sentences. Also, the rules are often domain-specific and not always portable to another domain. Therefore, supervised learning techniques are more appropriate in the above scenario. Traditional machine learning-based model also suffers from the high dimensional features of texts. This is why deep learning-based approaches are becoming increasingly popular for natural language processing tasks such as causality detection. However, deep learning models often require large datasets with high-quality features to perform well. Extracting deeply-learnable causal features and applying them to a carefully designed deep learning model is important. Also, preparing a large human-labeled training dataset is expensive and time-consuming. Even if a large training dataset is available, it is computationally expensive to train a deep learning model due to the complex structure of neural networks. We focus on addressing the following challenges: (i) extracting highquality causal features, (ii) designing an effective deep learning model to learn from the causal features, and (iii) reducing the dependency on large training datasets. Our main goals in this thesis are as follows: (i) we aim to study the different aspects of causality and causal discovery in text in depth. (ii) We aim to develop strategies to model causality in text, (iii) and finally, we aim to develop frameworks to design effective and efficient deep neural network structures to discover causality in texts.Thesis (PhD Doctorate)Doctor of Philosophy (PhD)School of Info & Comm TechScience, Environment, Engineering and TechnologyFull Tex
Corrosion inhibition of mild steel by metal cations in high pH simulated fresh water at different temperatures
Full text linkCorrosion inhibition effect of metal cations on mild steel was investigated by immersion tests and electrochemical impedance spectroscopy in simulated fresh water with high pH at different temperatures. Immersion tests showed the different corrosion rates in the different solutions, and the Zn2+ containing solution showed the minimum corrosion rate at the experimental temperatures. The specimen immersed in the Zn2+ containing solution showed comparatively smooth surface which was observed by scanning electron microscopy and atomic force microscopy. EIS and XPS results suggested that Zn2+ attached to the steel surface and formed a layer, thereby improving the corrosion inhibition ability of steel
A Secure and Efficient Communication Framework for Software-Defined Wireless Body Area Network
No full textDue to the recent development and advancement of communication technologies, healthcare industries are becoming more attracted towards information and communication technology services. One of the interesting services is the remote monitoring of patients through the use of a wireless body area network (WBAN), which enables healthcare providers to monitor, diagnose and prescribe patients without being present physically. To develop reliable and
exible remote patient monitoring services, in this thesis, the current state-of-the art of WBAN and the limitations of current WBAN technologies are investigated in the healthcare domain. To this end, the relevant background, implementation challenges and limitations of WBAN are overviewed. The in-depth literature survey identifies the lack of a current WBAN architecture in terms of administrative control, static architecture, vendor dependency, traffic priority arrangements, resource utilization, secure data sharing etc. To find a solution to the limitations of WBAN, software-defined networking (SDN) is considered to be one of the promising solutions in this paradigm. However, the incorporation of SDN into WBAN has several challenges in terms of architectural framework, resource optimization and secure data sharing.
In this thesis, an SDN-based WBAN (SDWBAN) architecture is proposed to incorporate the functionalities and principles of SDN on top of the traditional WBAN architecture to overcome the existing barriers of WBAN. The proposed communication model of the SDWBAN framework utilizes the sector-based distance (SBD) routing protocol for data packet dissemination. Furthermore, an application classification algorithm is developed to prioritize emergency applications over normal applications.
The proposed architecture and communication model have been simulated and experiments are conducted in Castalia 3.2. The simulation outcome demonstrates enhanced performance in terms of the packet delivery rate (PDR) and the latency of the emergency applications in comparison to normal applications.
For resource optimization, a mathematical model is developed to optimize the design of the control plane in the proposed SDWBAN framework. The purpose of the model is to reduce the unnecessary wastage of resources and find an optimal relationship among the number of controllers, SDN-enabled switches (SDESWs) and body sensors (BSs) which can potentially maximize network performance. The key factors in the proposed mathematical model encompass the number of controllers,
ow resolution time and number of SDESWs and BSs. The specific number of controllers returned by the model is used in the proposed SDWBAN and experiments are conducted in Castalia 3.2. The simulation results reveal that the optimal number of controllers returned by our model produces an acceptable range of PDR and latency.
Finally, a secure data-sharing platform is proposed for our SDWBAN framework. The platform is developed based on the cutting-edge blockchain technology and considers multiple entities such as healthcare professionals from various clinics, medical researchers and health insurers etc. The platform is implemented with a proof-of-concept (PoC) smart contract in Ethereum private blockchain using the Solidity programming language. The platform is validated in terms of time to execute functions in a data-sharing contract (DS-Contract) and hash-contract, the time to receive data packets from the gateway and the transaction time to run the smart contract. A low overhead is observed in the experiment which justifies the suitability of the platform to be used as a secure datasharing platform for SDWBAN.Thesis (PhD Doctorate)Doctor of Philosophy (PhD)School of Info & Comm TechScience, Environment, Engineering and TechnologyFull Tex
Molecular Detection and Antibiotic Resistance of Vibrio cholerae, Vibrio parahaemolyticus, and Vibrio alginolyticus from Shrimp (Penaeus monodon) and Shrimp Environments in Bangladesh
No full textSome Vibrio species can cause food-borne diseases in humans, including cholera, vomiting, septicemia, and gastroenteritis, which are associated with the consumption of contaminated seafood products. The study was conducted to detect antimicrobial-resistant Vibrio species in shrimp and shrimp environments in Bangladesh. Samples of shrimp (n = 50), water (n = 50), and mud (n = 50) were collected aseptically from 50 different shrimp culture ponds in the Khulna region of Bangladesh. Identification of Vibrio species was based on cultural and staining characteristics, biochemical tests, and polymerase chain reaction (PCR). Antimicrobial resistance profiles were determined using a disk diffusion assay. By PCR, Vibrio isolates were found in 34% (95% CI: 26.9%–41.9%) of the samples, of which the detection rate was significantly higher in shrimp (54%), compared to mud (26%) and water (22%). Moreover, V. cholerae, V. parahaemolyticus, and V. alginolyticus were detected in 24.7%, 15.3%, and 4% of the samples, respectively. Among them, the detection rate of V. cholerae and V. alginolyticus was significantly higher in shrimp samples than in other samples. V. parahaemolyticus was also higher in the shrimp samples, but the difference was not statistically significant. Vibrio isolates showed high to moderate resistance (92.2%–15.7%) to ampicillin, amikacin, cefotaxime, tetracycline, ceftazidime, gentamicin, nalidixic acid, levofloxacin, and ciprofloxacin, and low resistance (3.9%) to imipenem, meropenem, chloramphenicol, and trimethoprim-sulfamethoxazole. Interestingly, 52.9% of the isolates were multidrug resistant, and the multiple antibiotic resistance index was up to 1.0. To our knowledge, this is the first study in Bangladesh detecting these three Vibrio species (V. parahaemolyticus, V. alginolyticus, and V. cholerae) from shrimp and shrimp environments by molecular approach in the same study. These findings reveal the alarmingly high occurrence of antimicrobial-resistant Vibrio species in shrimp and shrimp environments, which should be of concern to both the shrimp industry and public health management
A review of microplastics pollution in the soil and terrestrial ecosystems: A global and Bangladesh perspective
No full textCorrosion inhibition effects of metal cations on SUS304 in 0.5 M Cl- aqueous solution
Full text linkThe corrosion characteristics of SUS304 exposed to 0.5 M Cl- aqueous solution containing different metal cations were studied with immersion tests, surface analysis and electrochemical tests. The mechanism of corrosion with metal cations was clarified by the XPS analysis results together with the hard and soft acid and base (HSAB) concept and the passive films structure. It is supposed that metal cations with large hardness make a layer by chemical bonding with the passive films. The passive films are protected by the metal cation layer from Cl- attack, and consequently corrosion reactions are inhibited
- …
