• Journal of Korea Multimedia Society
• /
• v.25 no.2
• /
• pp.145-156
• /
• 2022

Nowadays, the complexity of object tracking models among hardware applications has become a more in-demand duty to complete in various indeterminable environment tracking situations with multifunctional algorithm skills. In this paper, we propose a virtual city environment using AirSim (Aerial Informatics and Robotics Simulation - AirSim, CityEnvironment) and use the DQN (Deep Q-Learning) model of deep reinforcement learning model in the virtual environment. The proposed object tracking DQN network observes the environment using a deep reinforcement learning model that receives continuous images taken by a virtual environment simulation system as input to control the operation of a virtual drone. The deep reinforcement learning model is pre-trained using various existing continuous image sets. Since the existing various continuous image sets are image data of real environments and objects, it is implemented in 3D to track virtual environments and moving objects in them.

• Ocean and Polar Research
• /
• v.24 no.4
• /
• pp.391-398
• /
• 2002

Trace organic contaminants in deep-sea sediments near Dokdo were analyzed. Total PAMs concentration ranged 14.8-314 ng/g dry weight and high molecular weight PAHs were dominant. The highest PAHs concentration was detected at A19 which located at Ulleung Basin. Most of organochlorines were under detection limit. Among the detected organochlorines, DDT compounds were dominant and followed by HCHs and HCB. Butyltin compounds and most of organophosphorus pesticides were not detected. Vertical distribution of PAHs showed typical sub-surface maximum and decreasing trends depending on depth. The highest PAHs concentration reached 454ng/g. Some organochlorines, DDT, HCH was detected and also showed decreasing trends. Other target organic pollutants were not detected in core sediments. Abnormally high level of PAHs concentration in A19 was discussed and the input sources were inferred to be the transport of sludge derived pollutant dumped at dumping site 'Byung' by deep current.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.24 no.6
• /
• pp.89-96
• /
• 2021

In this study, a deep-learning image analysis model was established and validated for AI-based monitoring of the tidal flat ecosystem for marine protected creatures Ocypode stimpsoni and their habitat. The data in the study was constructed using an unmanned aerial vehicle, and the U-net model was applied for the deep learning model. The accuracy of deep learning model learning results was about 0.76 and about 0.8 each for the Ocypode stimpsoni and their burrow whose accuracy was higher. Analyzing the distribution of crabs and burrows by putting orthomosaic images of the entire study area to the learned deep learning model, it was confirmed that 1,943 Ocypode stimpsoni and 2,807 burrow were distributed in the study area. Through this study, the possibility of using the deep learning image analysis technology for monitoring the tidal ecosystem was confirmed. And it is expected that it can be used in the tidal ecosystem monitoring field by expanding the monitoring sites and target species in the future.

• Ocean and Polar Research
• /
• v.25 no.3
• /
• pp.257-267
• /
• 2003

Deep-sea bottom photographs acquired in the Clarion-Clipperton fracture zone of the northeast equatorial Pacific were analyzed to reveal the controlling processes for the spatial variation of manganese nodule. The results show that regional-scale occurrence variations of manganese nodule are mainly controlled by primary productivity of surface water, sedimentation rate, and water depth (or carbonate compensation depth). As a result, the diagenetic accretion on nodules increases toward southwest while hydrogenetic accretion increases toward northeast. Considering the northwestward movement of Pacific Plate, this regional-scale variation of manganese nodule occurrence seems to be affected by oceanic environment during the active growth period (Oligocene-Miocene) of Pacific Plate.

• Proceedings of the Korean Society of Soil and Groundwater Environment Conference
• /
• 2003.09a
• /
• pp.511-514
• /
• 2003

Nowadays, wells tapping the deep aquifers become general because water quality of the shallow groundwater has been gradually degraded over the last 30 years as a result of rapid industrialization and intensive agricultural activities. However, many of the deep wells also suffer problems of water-quality degradation in several years after the well installation, nevertheless those were properly completed and managed. It is believed that the heavy pumping from deep wells causes the doward movement of the contaminated, shallow groundwaters and introduces them into the deep aquifers. In this study, we introduces a shallow capture well system, which could effectively prevent the shallow groundwaters of poor water duality from moving into the deep aquifers by pumping of deep production wells. Even though additional costs are required to apply this system, we believe that this method could be appropriate for the deep wells that are important for the public health.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.3
• /
• pp.911-931
• /
• 2021

Recently, most cloud services use Docker container environment to provide their services. However, there are no researches to evaluate the performance of communication libraries for multi-GPU based distributed deep learning in a Docker container environment. In this paper, we propose an efficient communication architecture for multi-GPU based deep learning in a Docker container environment by evaluating the performances of various communication libraries. We compare the performances of the parameter server architecture and the All-reduce architecture, which are typical distributed deep learning architectures. Further, we analyze the performances of two separate multi-GPU resource allocation policies - allocating a single GPU to each Docker container and allocating multiple GPUs to each Docker container. We also experiment with the scalability of collective communication by increasing the number of GPUs from one to four. Through experiments, we compare OpenMPI and MPICH, which are representative open source MPI libraries, and NCCL, which is NVIDIA's collective communication library for the multi-GPU setting. In the parameter server architecture, we show that using CUDA-aware OpenMPI with multi-GPU per Docker container environment reduces communication latency by up to 75%. Also, we show that using NCCL in All-reduce architecture reduces communication latency by up to 93% compared to other libraries.

• Smart Media Journal
• /
• v.11 no.11
• /
• pp.25-31
• /
• 2022

Deep learning-based image processing is essential for autonomous vehicles. To process road images in real-time in a System-on-Chip (SoC) environment, we need to execute deep learning models on a NPU (Neural Procesing Units) specialized for deep learning operations. In this study, we imported seven open-source image processing deep learning models, that were developed on GPU servers, to Texas Instrument Deep Learning (TIDL) NPU environment. We confirmed that the models imported in this study operate normally in the SoC virtual environment through performance evaluation and visualization. This paper introduces the problems that occurred during the migration process due to the limitations of NPU environment and how to solve them, and thereby, presents a reference case worth referring to for developers and researchers who want to port deep learning models to SoC environments.

• Journal of Korean Society of Water and Wastewater
• /
• v.34 no.6
• /
• pp.481-493
• /
• 2020

Deep learning models, which imitate the function of human brain, have drawn attention from many engineering fields (mechanical, agricultural, and computer engineering etc). The major advantages of deep learning in engineering fields can be summarized by objects detection, classification, and time-series prediction. As well, it has been applied into environmental science and engineering fields. Here, we compiled our previous attempts to apply deep learning models in water-environment field and presented the future opportunities.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2005.10a
• /
• pp.15-26
• /
• 2005

This paper concerns the assessment of impact of deep excavation on surrounding environment with emphasis on the groundwater lowering. Fundamentals of ground excavation and groundwater interaction were reviewed and the stress-pore pressure coupled analysis approach as a tool for assessment was introduced. A case study concerning the use of coupled analysis for deep excavation design was presented. Implications of the finding from from this study were discussed.

• Journal of Welding and Joining
• /
• v.15 no.3
• /
• pp.99-108
• /
• 1997

For the conventional welding method, the high heat transfer makes the crystallization of the work material unavoidable. Whereas the laser is able to weld the amorphous metal without a crystallized zone, because heat transfer is limited withn a very small restricted volume. In this paper, the possibilities and the limits of the laser welding in a deep frozen environment by liquid nitrogen were studied to utilize the advantageous properties of amorphous metal foils. The author investigated, after laser welding in a deep frozen environment with a solid state laser (Nd:YAG-laser), the achievable strengths and the influences of the laser beam parameters on the strengths.