Search

This study proposes a novel framework for fish species classification that combines FRCNN (Faster Region-based Convolutional Neural Network), VGG16 (Visual Geometry Group 16), and SPPNet (Spatial Pyramid Pooling network). The proposed FRCNN-VGG16-SPPNet framework combines the strengths of FRCNN's fast object detection and localization, VGG16's convenient transfer learning and fast classification performance, and SPPNet's image processing flexibility and robustness in handling input images of any size. First, FRCNN is used to detect and extract target objects from images containing multiple objects. Subsequently, photos of various fish species at different scales are fed into VGG16-SPPNet, which performs basic feature extraction using transfer learning theory. SPPNet further processes the input images by performing pooling operations of different scales. Finally, VGG16 identifies important features to perform object classification. The proposed framework achieves higher accuracy compared to traditional single VGG16 models, particularly in classifying objects of different sizes, with an accuracy rate of 0.9318, which is 26% higher than traditional single VGG16 models. The proposed framework is efficient, convenient, reliable, and robust for object classification and has potential for various applications in image recognition and classification.

Polychlorinated biphenyl (PCBs) are persistent organic compounds used between 1930 and 1970 as thermal insulators in electrical equipment such as transformers, electrical condensers, heat exchangers, and hydraulic systems. Currently, this xenobiotic is a potential source of pollution to soil and water resources. This research aimed to isolate and characterize bacteria strains possessing a high potential to degrade PCBs contained in used dielectric oils. The bacteria strains were isolated from a soil sample taken from “Alao” Hydroelectric Power Center. Four culture media were used for the soil sample culturing: Agar for Standard Methods Medium (STD), Phosphorus, Ammonium, and Salts Medium (PAS), Minimum Medium (Mm), and Minimum Medium (Mm) supplemented with 25 µL of oil contaminated with 43,6 ppm of PCBs. The isolation, screening, and stabilization process were carried out using three culture media: STD, PAS, and Mm, each supplemented with 25 µL of oil contaminated with 43,6 ppm of PCBs. Nineteen bacterial colonies were isolated and exposed to a higher concentration of PCBs. Eighteen colonies could tolerate 250 µL of oil contaminated with 500 ppm of PCBs. Once stabilized and purified these bacteria strains, were characterized according to their microscopic and macroscopic characteristics and tolerance to heavy metals. Through biochemical tests, 18 strains were identified belonging to Providencia alcalifaciens, Proteus mirabillis, Providencia rettgeri, Providencia stuartii, Pseudomona alcaligenes, Morganella morganii, and Shigella disenteriae

Although many machine learning methods have been widely used to predict PM2.5 concentrations, these single or hybrid methods still have some shortcomings. This study integrated the advantages of convolutional neural network (CNN) feature extraction and the regression ability of random forest (RF) to propose a novel CNN-RF ensemble framework for PM2.5 concentration modeling. The observational data from 13 monitoring stations in Kaohsiung in 2021 were selected for model training and testing. First, CNN was implemented to extract key meteorological and pollution data. Subsequently, the RF algorithm was employed to train the model with five input factors, namely the extracted features from the CNN and spatiotemporal factors, including the day of the year, the hour of the day, latitude, and longitude. Independent observations from two stations were used to evaluate the models. The findings demonstrated that the proposed CNN–RF model had better modeling capability compared with the independent CNN and RF models: the average improvements in root mean square error (RMSE) and mean absolute error (MAE) ranged from 8.10% to 11.11%, respectively. In addition, the proposed CNN–RF hybrid model has fewer excess residuals at thresholds of 10 μg/m3, 20 μg/m3, and 30 μg/m3. The results revealed that the proposed CNN–RF ensemble framework is a stable, reliable, and accurate method that can generate superior results compared with the single CNN and RF methods. The proposed method could be a valuable reference for readers and may inspire researchers to develop even more effective methods for air pollution modeling. This research has important implications for air pollution research, data analysis, model estimation, and machine learning.

The Global Positioning System (GPS) is satellite-based, with receiving equipment worldwide utilizing geographic positioning satellites in Earth orbit. The system is unaffected by the radio positioning system, so it provides highly accurate three-dimensional positioning, velocity, and time data to users. In this paper, Alishan Township, Chiayi County, central Taiwan, is selected to test an automatic real-time monitoring system comprising of one machine with multiple GPS antennas. To this end, the Alishan Public Works Section installed advanced measuring instruments and a landslide-monitoring system composed of a high-efficiency transmission system. A pre- and post-rainfall data survey was conducted on this slope section. Together with the rainfall records, real-time ground-slip monitoring data was collected, and subsequently analyzed to understand the disaster situation and ground slip characteristics of the Alishan Highway following an earthquake (1998). In the future, more effective management values will be set to reduce the loss of slope disasters to conserve land and public safety. For the first time in Taiwan, this paper presents displacement data indicating that after the typhoon rains, a maximum surface movement velocity of 2.5 cm every six days is attained, while the total displacement per month is as high as 10 cm. These data can be used as for the remediation of this section of the Alishan Highway.

This study evaluates droughts in the Mekong River Basin (MKB) based on a multidisciplinary method, mainly using the Standardized Precipitation Index (SPI) and Mann–Kendall (MK) test. Precipitation data corresponding to the seasonality of the regional climate were retrieved from Integrated Multi-satellitE Retrievals for Global Precipitation Measurement from 2001 to 2020, at a monthly temporal scale and 0.1 degree spatial resolution. Drought events and their average interval, duration, and severity were determined based on Run theory. Our results revealed the most extreme drought period was in January 2014, at the time the lowest precipitation occurred. Spatial extreme drought results indicated that Zone 2 in the upstream MKB has the highest frequency of drought, with 44 events observed during 19 years, and experiences the most severe droughts, whereas Zone 24 in the downstream MKB has the most prolonged drought duration of seven months. The periods and locations of extreme drought were identified using the SPI, corresponding to historic droughts of the MKB. Furthermore, the MK test shows an increasing trend of droughts in the lower MKB and the cluster analysis identified six clusters of times series. Overall, our study provides essential findings for international and national water resource stakeholders in identifying trends of extreme drought in the MKB.

Ban Khoang is a mountainous commune in Sa Pa district located in the central part of Lao Cai province, Vietnam. Landslides occur frequently in this area and seriously affect the local living conditions. To help the local authority in developing a landslide disaster action plan, the statistical index method for landslide susceptibility mapping is applied. As the result, the landslide susceptibility zonation (LSZ) map was created. The LSZ map indicates that areas of low, moderate, high and very high landslide susceptibility zones are, respectively, 20.3 km2, 12.4 km2, 15.4 km2, and 5.2 km2; most of the observed landslide areas that are well predicted belong to high or very high landslide susceptibility classes. In detail, 80% observed landslide areas and 78.57% number of observed landslides were well predicted, and the area (AUC) under the receiver operating characteristic (ROC) curve obtained 80.3%. Hence, the high and very high landslide susceptibility classes in the LSZ map can be considered highly believable, and the LSZ map will be reliable to use in the practice.

This paper investigates a novel hybrid model that employs Henry’s gas solubility optimization algorithm to search for adaptive weights of a deep neural network for a landslide susceptibility application. The model is trained using 13 features from a case study site in Viet Nam, including topographically derived data, satellite-derived indexes, physical and climatic information. The training and validating process terminated with the Root mean square error = 0.3898, Mean absolute error = 0.3115, Area under Receiving characteristic curves = 0.862, and Overall accuracy = 0.7831. The results show that the proposed model outperforms other methods, and it was used to map the susceptible level of the entire study area. Therefore, this model can be used as an alternative method for similar landslide studies and other natural hazard analyses, given input data availability.

The Linbeiken area is located in the village of Pingding, Taiwan. Since the Mindulle and Aere Typhoons in 2004, and as a result of the landslide triggered by the continuous heavy rainfall on 9 June 2006, there has been a persistent collapse of side slopes in the area. This paper describes the equipment that was installed to collect on-site topographic and hydrological information in the Linbeiken area upstream of the Pingding River and to monitor changes in the landslide area, as well as the measurements that were collected during the 2008 Typhoon Sinlaku. A case study of a landslide in Pingding, Taiwan was used to monitor the accurate coordinate changes in the potential landslide areas during typhoons. The goal of this study was to establish warning indexes, and to strengthen the software and hardware at the local disaster response center in the hope of gaining a full idea of the surface movement in landslide areas in future flood seasons. This is important for boosting the preparedness to adapt to landslide hazards, for improving disaster warnings, and for reporting efficiently to better protect the lives and property of local residents. The results show that the landslide disaster monitoring and warning system in Taiwan, as applied during Typhoon Sinlaku in 2008, is both effective and comprehensive.

Forest fires constitute a major environmental problem in tropical countries, especially in the context of climate change and increasing human populations. This paper aims to identify the causes of frequent forest fires in Son La Province, a fire-prone and forested mountainous region in northwest Vietnam, with a view to constructing a forest fire-related database with multiple layers of natural, social and economic information, extracted largely on the basis of Landsat 7 images. The assessment followed an expert systems approach, applying multi-criteria analysis (MCA) with an analytical hierarchy process (AHP) to determine the weights of the individual parameters related to forest fires. A multi-indicator function with nine parameters was constructed to establish a forest fire risk map at a scale of 1:100,000 for use at the provincial level. The results were verified through regression analysis, yielding R2 = 0.86. A real-time early warning system for forest fire areas has been developed for practical use by the relevant government authorities to provide more effective forest fire prevention planning for Son La Province.

Traffic speed forecasting in the short term is one of the most critical parts of any intelligent transportation system (ITS). Accurate speed forecasting can support travelers’ route choices, traffic guidance, and traffic control. This study proposes a deep learning approach using long short-term memory (LSTM) network with tuning hyper-parameters to forecast short-term traffic speed on an arterial parallel multi-lane road in a developing country such as Vietnam. The challenge of mishandling the location data of vehicles on small and adjacent multi-lane roads will be addressed in this study. To test the accuracy of the proposed forecasting model, its application is illustrated using historical voyage GPS-monitored data on the Le Hong Phong urban arterial road in Haiphong city of Vietnam. The results indicate that in comparison with other models (e.g., traditional models and convolutional neural network), the best performance in terms of root mean square error (RMSE), mean absolute error (MAE), and median absolute error (MDAE) is obtained by using the proposed model.

Climate change has been impacting strongly on economic sectors, facilities, people and wealth. Developing sustainable disaster prevention and mitigation technologies in the context of climate change is a long-term and urgent task of all levels, sectors and localities. Based on the analysis of scientific findings on the evolution of some major climatic factors (temperature and rainfall), sea level rise and the impact of these factors on the sectors of the sector Quang Ninh Tourism. At the same time, assessing the impact of climate change on all sectors, form the basis for new proposals for high-tech applications. Keywords Climate change; Disaster; Disaster prevention; Disaster mitigation; Quang Ninh; Quang Ninh flooding DOI: https://doi.org/10.23953/cloud.ijarsg.373

The Bridge Health Monitoring (BHM) system has become crucial for long-span bridge structures. This study aims to assess the bridge deck movement based-on GPS time series data and environmental data obtained from BHM system. The correlation and regression analysis are utilized to find out the pairwise relationship between the bridge deck movement in all directions and environmental factors as well as traffic load. The result shows the highest correlations in bridge deck vertical movements, the symmetrical oscillation in different parts of bridge deck and it also reveals the impact of air temperature change, wind velocity and traffic vehicle on bridge deck movement in all directions.

With the development of mobile mapping systems, point cloud is becoming an important data type to build the three-dimensional geospatial information. In particular, some recent applications such as robot navigation, self-driving drones/vehicles, and construction robotics need three-dimensional high-definition maps within cm-level accuracy to make a decision of action by themselves. These maps are based on point cloud data from three-dimensional surveys of geographic features on the Earth by laser scanning and photogrammetry. Point clouds provide a flexible and accurate geometric information for reconstructing 3D shapes of not only indoor but also outdoor features. However, it expends time on creating a map due to many manual operations of processing. In this paper, we design an interoperable framework of semantic annotation for point cloud data. Semantic annotation of point cloud is the process of attaching conceptual label information such as cars, buildings, desks, or people to each point in a set of data. This process takes first priority to create training data set for machine learning and deep learning tasks. An efficient annotation tool for 3D point cloud data is crucial to improve the quality of deep learning models and getting better results in deep learning would be helpful to reduce the manual processing. Even though there are already a number of tools, they are usually standalone and tightly coupled systems. This is likely to lead to a problem of limited interoperability of data and process between systems. This paper introduces a resource-oriented design for the annotation framework of point cloud data on the Web with the ability of quick adaptation of various machine learning techniques.

In recent years, as information technologies have been widely adopted, more countries have focused attention on the intelligentization of cities. In 2012, the United Nations listed “Smart City” as one of the progress indicators of a country. The original development of smart cities emphasized technology-based demands, but gradually promoted people-oriented needs, especially on issues related to living and environment, such as public safety, transportation, environmental sustainability and health care. The most important concern is to provide effective solutions to local problems. Spatial information has become wider and more diversified with the rapid development of high technologies, such as ICT, IoT and big data applications. It plays an important role in the planning of smart cities and smart homeland. Municipal governance such as traffic control, environment monitoring, disaster prevention information is based on geographic information system (GIS). The resources, information and data of cities can be spatially visualized. The spatially visualized information are not only texts and data, but integrated, coherent and spatially effective information which can identify and explore correlation and development trends in urban information. In view of this, the government uses smart homeland as the core to explain development and planning of smart city and smart village, and in consideration of the unique geographical characteristics of Taiwan and the importance of spatial resources management and strategy, promotes development of spatial information technologies and application of the technologies to smart homeland. On the basis of smart homeland concept, this study discusses key points of promotional strategies and action plans of the cities in Taiwan, and from the perspective of spatial information, explores and analyzes how to integrate spatial information technologies into urban governance, disaster prevention traffic and tourism under Taiwan's strategies for smart homeland information development. It can help the county and city governments to propose the smart city development strategies, improve government governance, promote regional economic growth, satisfy people's demands and promote the sustainable development of smart cities.

The aim of this study was to develop an online collaborative map to enable researchers to locate, explore, and share cancer data.This 2-scale (global and country-level) cancer map adopts a database-driven model, which was implemented using the Google Map Application Programming Interface (API) and asynchronous JavaScript and XML (AJAX) technology. Seven visualization techniques were used to present data. Data on worldwide cancer mortality between 1950 and 2013 were taken from the International Agency for Research on Cancer (IARC) database. Incidence data were from the IARC CI5plus database. Survival data were from the IARC SURVCAN study. Prevalence data between 1990 and 2017 were from the Institute for Health Metrics and Evaluation's (IHME) catalog while demographic data were from the World Bank Data Catalog. Cancer data for Taiwan between 1991 and 2016 were obtained from the Department of Health and Welfare. This study used visualization techniques that included: a choropleth map to display the prevalence of cancer; a tornado diagram to show the age-standardized mortality rates of all cancers among men and women in 2013; a treemap to show a ranking of cancer mortality data; a sunburst chart to show mortality rates of all cancers by gender; a line chart to show mortality trends for all cancers; a bar chart to show mortality and incidence rates and a heatmap to show variations in cancer across different countries.The world cancer map generated by this study can be accessed at http://worldmap.csmu-liawyp.tw. Country-level mortality data are presented as crude and age-standardized rates.We used visualization methodologies and constructed an easily maintainable web-based user interface with cancer data from administrative regions in 150 countries. This serves as a platform that allows researchers to manage and disseminate cancer data.

In recent years, the rapid development of remote sensing technology has proliferated high-quality images that occupy larger and larger storage spaces. Video has become widespread for environmental observation. Hence, digital data is growing exponentially, and geographic information systems must determine how to manage and process images and video effectively. Researchers cannot limit themselves to desktop PCs due to computational and storage limits. The aim of this article was to propose and implement an architectural design for a novel cloud computing platform based on two Web Coverage Service and Web Map Service interfaces from the Open Geospatial Consortium (OGC), cloud storage from Hadoop Distributed File System (HDFS), and image processing from MapReduce. Results are presented on tablet computers (Asus transformer pad) and websites. Within this framework, we implemented image management as well as simple WebGIS and created an experiment in read/write performance with four kinds of data sets (normal d...

Using GIS in the classroom enhance students’ computer skills and explore the range of knowledge. The paper highlights GIS integration on e-learning platform and introduces a variety of abundant educational resources. This research project will demonstrate tools for e-learning environment and delivers some case studies for learning interaction from Central Taiwan Universities. Feng Chia University (FCU) obtained a remarkable academic project subsidized by Ministry of Education and developed e-learning platform for excellence in teaching/learning programs among Central Taiwan’s universities. The aim of the project is to integrate the educational resources of 13 universities in central Taiwan. FCU is serving as the hub of Center University. To overcome the problem of distance, e-platforms have been established to create experiences with collaboration enhanced learning. The e-platforms provide coordination of web service access among the educational community and deliver GIS educational resources. Most of GIS related courses cover the development of GIS, principles of cartography, spatial data analysis and overlaying, terrain analysis, buffer analysis, 3D GIS application, Remote Sensing, GPS technology, and WebGIS, MobileGIS, ArcGIS manipulation. In each GIS case study, students have been taught to know geographic meaning, collect spatial data and then use ArcGIS software to analyze spatial data. On one of e-Learning platforms provide lesson plans and presentation slides. Students can learn Arc GIS online. As they analyze spatial data, they can connect to GIS hub to get data they need including satellite images, aerial photos, and vector data. Moreover, e-learning platforms provide solutions and resources. Different levels of image scales have been integrated into the systems. Multi-scale spatial development and analyses in Central Taiwan integrate academic research resources among CTTLRC partners. Thus, establish decision-making support mechanism in teaching and learning. Accelerate communication, cooperation and sharing among academic units

During nuclear accidents, when radioactive materials spread into the environment, the people in the affected areas should evacuate immediately. However, few information systems are available regarding escape guidelines for nuclear accidents. Therefore, this study constructs escape guidelines on mobile phones. This application is called Mobile Escape Guidelines (MEG) and adopts two techniques. One technique is the geographical information that offers multiple representations; the other is the augmented reality that provides semi-realistic information services. When this study tested the mobile escape guidelines, the results showed that this application was capable of identifying the correct locations of users, showing the escape routes, filtering geographical layers, and rapidly generating the relief reports. Users could evacuate from nuclear accident sites easily, even without relief personnel, since using slim devices to access the mobile escape guidelines is convenient. Overall, this study is a useful reference for a nuclear accident emergency response.