• International Journal of Computer Science & Network Security
• /
• v.22 no.2
• /
• pp.214-222
• /
• 2022

Convolutional Neural networks (CNNs) are a category of deep learning networks that have proven very effective in computer vision tasks such as image classification. Notwithstanding, not much has been seen in its use for remote sensing image classification in developing countries. This is majorly due to the scarcity of training data. Recently, transfer learning technique has successfully been used to develop state-of-the art models for remote sensing (RS) image classification tasks using training and testing data from well-known RS data repositories. However, the ability of such model to classify RS test data from a different dataset has not been sufficiently investigated. In this paper, we propose a deep CNN model that can classify RS test data from a dataset different from the training dataset. To achieve our objective, we first, re-trained a ResNet-50 model using EuroSAT, a large-scale RS dataset to develop a base model then we integrated Augmentation and Ensemble learning to improve its generalization ability. We further experimented on the ability of this model to classify a novel dataset (Nig_Images). The final classification results shows that our model achieves a 96% and 80% accuracy on EuroSAT and Nig_Images test data respectively. Adequate knowledge and usage of this framework is expected to encourage research and the usage of deep CNNs for land cover mapping in cases of lack of training data as obtainable in developing countries.

• Journal of Sensor Science and Technology
• /
• v.21 no.2
• /
• pp.83-89
• /
• 2012

A simple array of metal-insulator-metal capacitive elements was proposed for a potential application in humidity sensing platforms. We fabricated meso-scale sensors with different sizes(large-size: $2.7{\times}2.7mm^2$ ; mid-size: $1.5{\times}1.5mm^2$ ; small-size: $0.7{\times}0.7mm^2$) and characterized the performance of each design. Polyimide films were utilized as a humidity-sensitive layer. Capacitance changes of the polyimide layer were measured with respect to water absorption. The device showed sensitivity in the full range of relative humidity (RH) with excellent linearity(correlation coefficient > 0.994). This array structure exhibits unique advantages including easy fabrication process, high batch productivity, and high structural compatibility with various substrate materials. It is anticipated that this device structure will be potentially useful in unique applications including mapping spatial humidity variations over a meso-scale area and implementing flexible humidity sensing element arrays.

• Earthquakes and Structures
• /
• v.22 no.3
• /
• pp.277-287
• /
• 2022

The ancient underground cities are a collection of self-supporting spaces that have been manually excavated in the soil or rock in the past. Because these structures have a very high cultural value due to their age, the study of their stability under the influence of natural hazards, such as earthquakes, is very important. In this research, while introducing the underground city of Ouyi Nushabad located in the center of Iran as one of the largest man-made underground cities of the old world, the analysis of dynamic stability is performed. For this purpose, the dynamic stress-displacement analysis has been performed through numerical modeling using the finite element software PLAXIS. At this stage, by simulating the Khorgo earthquake as one of the large-scale earthquakes that occurred in Iran, with a magnitude of 6.9 on the Richter scale, dynamic analysis by time history method has been performed on three selected sections of underground spaces. This study shows that the maximum amount of horizontal and vertical dynamic displacement is 12.9 cm and 17.7 cm, respectively, which was obtained in section 2. The comparison of the results shows that by increasing the cross-sectional area of the excavation, especially the distance between the roof and the floor, in addition to increasing the amount of horizontal and vertical dynamic displacement, the obtained maximum acceleration is intensified compared to the mapping acceleration applied to the model floor. Therefore, preventive actions should be taken to stabilize the excavations in order to prevent damage caused by a possible earthquake.

• Genomics & Informatics
• /
• v.18 no.1
• /
• pp.10.1-10.9
• /
• 2020

Advancements in next generation sequencing (NGS) technologies have significantly increased the translational use of genomics data in the medical field as well as the demand for computational infrastructure capable processing that data. To enhance the current understanding of software and hardware used to compute large scale human genomic datasets (NGS), the performance and accuracy of optimized versions of GATK algorithms, including Parabricks and Sentieon, were compared to the results of the original application (GATK V4.1.0, Intel x86 CPUs). Parabricks was able to process a 50× whole-genome sequencing library in under 3 h and Sentieon finished in under 8 h, whereas GATK v4.1.0 needed nearly 24 h. These results were achieved while maintaining greater than 99% accuracy and precision compared to stock GATK. Sentieon's somatic pipeline achieved similar results greater than 99%. Additionally, the IBM POWER9 CPU performed well on bioinformatic workloads when tested with 10 different tools for alignment/mapping.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2016.06a
• /
• pp.260-263
• /
• 2016

Recently high dynamic range imaging technique is hot issue in computer graphic area. We present a progressive tone mapping algorithm, which is based on weighted least squares optimization framework. Our approach combines weighted leastsquaresfiltering with iCAM06, for showing more perceptual high dynamic range images in conventional display, while avoiding visual halo artifacts. We decompose high dynamic range image into base layer and detail layer. The base layer has large scale variation, it is obtained by using weighted least squares filtering, and then the base layer incorporates iCAM06 model. Then, adaptive compression on the base layer according to human visual system. Only the base layer reduces contrast, and preserving detail. The resultshows more perceptual color appearance and preserve fine detail, while avoiding common artifacts.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2010.09a
• /
• pp.782-789
• /
• 2010

Landslides cause enormous economic losses and casualties. Korea has mountainous regions and heavy slopes in most parts of the land and has consistently built new roads and large-scale housing complexes according to its industrial and urban growth. As a result, the damage from landslides becomes greater every year. In this study, performed a GIS-based landslide hazard analysis by SINMAP(Stability Index MAPping) model in Gyeonggi Icheon area coupling with geomorphological and geological data. SINMAP model has its theoretical basis in the infinite plane slope stability model with wetness obtained from a topographically based steady state model of hydrology. To Gyeonggi Icheon area landslides hazards evaluated, these SINMAP model were analysed results while simultaneously referring to the stability index map, where lines distinguish the zones categorized into the different stability classes and a table giving summary statistics.

• International Journal of Aeronautical and Space Sciences
• /
• v.18 no.4
• /
• pp.767-779
• /
• 2017

Automated mission planning for unmanned aerial vehicles (UAVs) is difficult because of the propagation of several sources of error into the solution, as for any large scale autonomous system. To ensure reliable system performance, we quantify all sources of error and their propagation through a mission planner for operation of UAVs in an obstacle rich environment we developed in prior work. In this sequel to that work, we show that the mission planner developed before can be made robust to errors arising from the mapping, sensing, actuation, and environmental disturbances through creating systematic buffers around obstacles using the calculations of uncertainty propagation. This robustness makes the mission planner truly autonomous and scalable to many UAVs without human intervention. We illustrate with simulation results for trajectory generation of multiple UAVs in a surveillance problem in an urban environment while optimizing for either maximal flight time or minimal fuel consumption. Our solution methods are suitable for any well-mapped region, and the final collision free paths are obtained through offline sub-optimal solution of an mTSP (multiple traveling salesman problem).

• Wind and Structures
• /
• v.20 no.4
• /
• pp.565-578
• /
• 2015

A wind energy assessment is an integrated analysis of the potential of wind energy resources of a particular area. In this work, the wind energy potentials for Mauritius have been assessed using a Computational Fluid Dynamics (CFD) model. The approach employed in this work aims to enhance the assessment of wind energy potentials for the siting of large-scale wind farms in the island. Validation of the model is done by comparing simulated wind speed data to experimental ones measured at specific locations over the island. The local wind velocity resulting from the CFD simulations are used to compute the weighted-sum power density including annual directional inflow variations determined by wind roses. The model is used to generate contour maps of velocity and power, for Mauritius at a resolution of 500 m.

• Applied Microscopy
• /
• v.46 no.2
• /
• pp.100-104
• /
• 2016

Electron microscopy is currently the only available technique with a spatial resolution sufficient to identify fine neuronal processes and synaptic structures in densely packed neuropil. For large-scale volume reconstruction of neuronal connectivity, serial block-face scanning electron microscopy allows us to acquire thousands of serial images in an automated fashion and reconstruct neural circuits faster by reducing the alignment task. Here we introduce the whole reconstruction procedure of synaptic network in the rat hippocampal CA1 area and discuss technical issues to be resolved for improving image quality and segmentation. Compared to the serial section transmission electron microscopy, serial block-face scanning electron microscopy produced much reliable three-dimensional data sets and accelerated reconstruction by reducing the need of alignment and distortion adjustment. This approach will generate invaluable information on organizational features of our connectomes as well as diverse neurological disorders caused by synaptic impairments.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.16 no.2
• /
• pp.327-335
• /
• 1998

Many kinds of map projection have been used, since this cannot be done without a distortion and seemless representation of the globe. To overcome the continues representation of large digital terrain information by the global coordinate system , to find out the sources of projection error and optimizing method, and to increase the map accuracy, specific map projection researches are needed. In this study, improvements of map projections are suggested for small scale mapping which covering entire Korean peninsula based on theoretical evaluation and limits of current map projection. Furthermore, distortion factors, positional errors, and accuracy increasing method are evaluated.