• Journal of Korean Tunnelling and Underground Space Association
• /
• v.24 no.4
• /
• pp.341-353
• /
• 2022

Pipelines are buried in urban area, and the position (depth and orientation) of buried pipeline should be clearly identified before ground excavation. Although various geophysical methods can be used to detect the buried pipeline, it is not easy to identify the exact information of pipeline due to heterogeneous ground condition. Among various non-destructive geo-exploration methods, ground penetration radar (GPR) can explore the ground subsurface rapidly with relatively low cost compared to other exploration methods. However, the exploration data obtained from GPR requires considerable experiences because interpretation is not intuitive. Recently, researches on automated detection technology for GPR data using deep learning have been conducted. However, the lack of GPR data which is essential for training makes it difficult to build up the reliable detection model. To overcome this problem, we conducted a preliminary study to improve the performance of the detection model using finite difference time domain (FDTD)-based numerical analysis. Firstly, numerical analysis was performed with homogeneous soil media having single permittivity. In case of heterogeneous ground, numerical analysis was performed considering the ground heterogeneity using fractal technique. Secondly, deep learning was carried out using convolutional neural network. Detection Model-A is trained with data set obtained from homogeneous ground. And, detection Model-B is trained with data set obtained from homogeneous ground and heterogeneous ground. As a result, it is found that the detection Model-B which is trained including heterogeneous ground shows better performance than detection Model-A. It indicates the ground heterogeneity should be considered to increase the performance of automated detection model for GPR exploration.

• Korean Journal of Remote Sensing
• /
• v.38 no.5_1
• /
• pp.511-521
• /
• 2022

Cloud detection and removal from satellite images is an essential process for topographic observation and analysis. Threshold-based cloud detection techniques show stable performance because they detect using the physical characteristics of clouds, but they have the disadvantage of requiring all channels' images and long computational time. Cloud detection techniques using deep learning, which have been studied recently, show short computational time and excellent performance even using only four or less channel (RGB, NIR) images. In this paper, we confirm the performance dependence of the deep learning network according to the heterogeneous learning dataset with different resolutions. The DeepLabV3+ network was improved so that channel features of cloud detection were extracted and learned with two published heterogeneous datasets and mixed data respectively. As a result of the experiment, clouds' Jaccard index was low in a network that learned with different kind of images from test images. However, clouds' Jaccard index was high in a network learned with mixed data that added some of the same kind of test data. Clouds are not structured in a shape, so reflecting channel features in learning is more effective in cloud detection than spatial features. It is necessary to learn channel features of each satellite sensors for cloud detection. Therefore, cloud detection of heterogeneous sensors with different resolutions is very dependent on the learning dataset.

• Journal of the Korean Society for Marine Environment & Energy
• /
• v.19 no.4
• /
• pp.341-348
• /
• 2016

Micro-algae, one of the biological resources for alternative energy, has been heavily studied. Among various methods to analyze the status of the micro-algae including counting, screening, and flocculation, the flocculation approach has been widely accepted in many critical applications such as red tide removal study or microalgae resource study. To characterize the flocculation status of the micro-alga. A traditional optical modality, i.e., photospectrometry, measuring the optical density of the flocs has been frequently employed. While this traditional optical method needs shorter time than the counting method in flocculation status analysis, it has relatively lower detection accuracy. To address this issue, a novel real-time micro-algae flocculation analysis method based on the lens-free shadow imaging technique (LSIT) is introduced. Both single cell detection and floc detection are simultaneously available with a proposed lens-free shadow image, confirmed by comparing the results with optical microscope images. And three shadow parameters, e.g., number of flocs, effective area of flocs, and maximum size of floc, enabling quantification of the flocculation phenomenon of micro-alga, are firstly demonstrated in this article. The efficacy of each shadow parameter is verified with the real-time flocculation monitoring experiments using custom developed cohesive agents.

• Journal of Korea Water Resources Association
• /
• v.53 no.4
• /
• pp.313-322
• /
• 2020

There is always a risk of water disasters due to sudden storms in mountainous regions in Korea, which is more than 70% of the country's land. In this study, a radar-based risk prediction technique for sudden downpour is applied in the mountainous region and is evaluated for its applicability using Mt. Biseul rain radar. Eight local heavy rain events in mountain regions are selected and the information was calculated such as early detection of cumulonimbus convective cells, automatic detection of convective cells, and risk index of detected convective cells using the three-dimensional radar reflectivity, rainfall intensity, and doppler wind speed. As a result, it was possible to confirm the initial detection timing and location of convective cells that may develop as a localized heavy rain, and the magnitude and location of the risk determined according to whether or not vortices were generated. In particular, it was confirmed that the ground rain gauge network has limitations in detecting heavy rains that develop locally in a narrow area. Besides, it is possible to secure a time of at least 10 minutes to a maximum of 65 minutes until the maximum rainfall intensity occurs at the time of obtaining the risk information. Therefore, it would be useful as information to prevent flash flooding disaster and marooned accidents caused by heavy rain in the mountainous area using this technique.

• Geophysics and Geophysical Exploration
• /
• v.6 no.2
• /
• pp.57-63
• /
• 2003

For the dectection of small cavity in the hard rock, we investigated the feasibility of crosswell travel-time tomography and Kirchhoff migration technique. In travel-time tomography, first arrival anomaly caused by small cavity was investigated by numerical modeling based on the knowledge of actual field information. First arrival delay was very small (<0.125 msec) and detectable receiver offset range was limited to 4m with respect to $1\%$ normalized first arrival anomaly. As a consequence, it was turned out that carefully designed survey array with both sufficient narrow spatial spacing and temporal (<0.03125 msec) sampling were required for small cavity detection. Also, crosswell Kirchhoff migration technique was investigated with both numerical and real data. Stack section obtained by numerical data shows the good cavity image. In crosswell seismic data, various unwanted seismic events such as direct wave and various mode converted waves were alto recorded. To remove these noises und to enhance the diffraction signal, combination of median and bandpass filtering was applied and prestack and stacked migration images were created. From this, we viewed the crosswell migration technique as one of the adoptable method for small cavity detection.

• Economic and Environmental Geology
• /
• v.40 no.5
• /
• pp.563-574
• /
• 2007

Laser induced breakdown spectroscopy (LIBS) is a recently developed analytical technique that is based upon the measurement of emission lines generated by atomic species close to the surface of the sample, thus allowing their chemical detection, identification and quantification. With powerful advantages of LIBS compared to the conventional analytical methodology, this technique can be applied in the detection of heavy metals in the field. LIBS allows the rapid analysis by avoiding laborious chemical steps. LES have already been applied for the determination of element concentration in a wide range of materials in the solid, liquid and gaseous phase with simplicity of the instrument and diversity of the analytical application. These feasibility of rapid multi elemental analysis are appealing proprieties for the in-situ analytical technique in geochemical investigation, exploration and environmental analysis. There remain still some limitations to be solved for LIBS to be applied in soil environment as an in-situ analytical technology. We would like to provide the basic principle related to the plasma formation and laser-induced breakdown of sample materials. In addition, the matrix effect, laser properties and the various factors affecting on the analytical signal of LIBS was dealt with to enhance understanding of LIBS through literature review. Ultimately, it was investigated the feasibility of LIBS application in soil environment monitoring by considering the basic idea to enhance the data quality of LIBS including the calibration method for the various effects on the analytical signal of LIBS.

• Geophysics and Geophysical Exploration
• /
• v.20 no.2
• /
• pp.61-71
• /
• 2017

Recently, ground-penetrating radar (GPR) surveys have been actively carried out for precise subsurface void investigation because of the rapid increase of subsidence in urban areas. However, since the interpretation of GPR data was conducted based on the interpreter's subjective decision after applying only the basic data processing, it can result in reliability problems. In this research, to solve these problems, we analyzed the difference between the events generated from subsurface voids and those of strong diffraction sources such as the buried pipeline by applying the edge detection technique, which is one of image processing technologies. For the analysis, we applied the image processing technology to the GRP field data containing events generated from the cavity or buried pipeline. As a result, the main events by the subsurface void or diffraction source were effectively separated using the edge detection technique. In addition, since subsurface voids associated with the subsidence has a relatively wide scale, it is recorded as a gentle slope event unlike the event caused by the strong diffraction source recorded with a sharp slope. Therefore, the directional analysis of amplitude variation in the image enabled us to effectively separate the events by the subsurface void from those by the diffraction source. Interpretation based on these kinds of objective analysis can improve the reliability. Moreover, if suggested techniques are verified to various GPR field data sets, these approaches can contribute to semiautomatic interpretation of large amount of GPR data.

• Journal of Broadcast Engineering
• /
• v.28 no.1
• /
• pp.31-41
• /
• 2023

Directly converting the focal depth and image size of computer-generated-hologram (CGH), which is obtained by calculating the interference pattern of light from the 3D image, is known to be quite difficult because of the less similarity between the CGH and the original image. This paper proposes a method for separately converting the each of focal length of the given CGH, which is composed of multi-depth images. Firstly, the proposed technique converts the 3D image reproduced from the CGH into a Light-Field (LF) image composed of a set of 2D images observed from various angles, and the positions of the moving objects for each observed views are checked using an object detection algorithm YOLOv5 (You-Only-Look-Once-version-5). After that, by adjusting the positions of objects, the depth-transformed LF image and CGH are generated. Numerical simulations and experimental results show that the proposed technique can change the focal length within a range of about 3 cm without significant loss of the image quality when applied to the image which have original depth of 10 cm, with a spatial light modulator which has a pixel size of 3.6 ㎛ and a resolution of 3840⨯2160.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.5D
• /
• pp.635-644
• /
• 2009

Image processing technique in the outdoor environment is very sensitive, and it tends to lose a lot of accuracy when it rapidly changes by outdoor environment. Therefore, in order to calculate accurate traffic information using the traffic monitoring system, we must resolve removing shadow in transition time, Distortion by the vehicle headlights at night, noise of rain, snow, and fog, and occlusion. In the research, we developed a system to calibrate the amount of traffic, speed, and time occupancy by using image processing technique in a variety of outdoor environments change. This system were tested under outdoor environments at the Gonjiam test site, which is managed by Korea Institute of Construction Technology (www.kict.re.kr) for testing performance. We evaluated the performance of traffic information, volume counts, speed, and occupancy time, with 4 lanes (2 lanes are upstream and the rests are downstream) from the 16th to 18th December, 2008. The evaluation method performed as based on the standard data is a radar detection compared to calculated data using image processing technique. The System evaluation results showed that the amount of traffic, speed, and time occupancy in period (day, night, sunrise, sunset) are approximately 92-97% accuracy when these data compared to the standard data.

• Journal of the Korea Concrete Institute
• /
• v.17 no.3 s.87
• /
• pp.369-374
• /
• 2005

This study proposes a technique for the recognition of crack patterns, which includes horizontal, vertical, diagonal($-45^{\circ}$), diagonal($+45^{\circ}$), and random cracks, based on image processing technique and artificial neural network. A MATLAB code was developed for the proposed image processing algorithm and artificial neural network. Features were determined using total projection technique, and the structure(no. of layers and hidden neurons) and weight of artificial neural network were determined by learning from artificial crack images. In this process, we adopted Bayesian regularization technique as a generalization method to eliminate overfitting Problem. Numerical tests were performed on thirty-eight crack images to examine validity of the algorithm. Within the limited tests in the present study, the proposed algorithm was revealed as accurately recognizing the crack patterns when compared to those classified by a human expert.