Sung-sik Park

Sung-sik Park


Research Focus & Metrics

Academic Career Length

17Years (2010-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 November 2021 in Applied Sciences
Verified
Journal Article
Open access
PDF file
SPSung-sik Park
Sung-sik Park
TTThan Van Tran
Than Van Tran
Dong-Eun LeeDong-Eun Lee
Dong-Eun Lee
Pothole repair is one of the paramount tasks in road maintenance. Effective road surface monitoring is an ongoing challenge to the management agency. The current pothole detection, which is conducted image processing with a manual operation, is labour-intensive and time-consuming. Computer vision offers a mean to automate its visual inspection process using digital imaging, hence, identifying potholes from a series of images. The goal of this study is to apply different YOLO models for pothole detection. Three state-of-the-art object detection frameworks (i.e., YOLOv4, YOLOv4-tiny, and YOLOv5s) are experimented to measure their performance involved in real-time responsiveness and detection accuracy using the image set. The image set is identified by running the deep convolutional neural network (CNN) on several deep learning pothole detectors. After collecting a set of 665 images in 720 × 720 pixels resolution that captures various types of potholes on different road surface conditions, the set is divided into training, testing, and validation subsets. A mean average precision at 50% Intersection-over-Union threshold (mAP_0.5) is used to measure the performance of models. The study result shows that the mAP_0.5 of YOLOv4, YOLOv4-tiny, and YOLOv5s are 77.7%, 78.7%, and 74.8%, respectively. It confirms that the YOLOv4-tiny is the best fit model for pothole detection.

1 September 2015 in KSCE Journal of Civil Engineering
Verified
Journal Article
Open access
PDF file
INIn-Hyun Nam
In-Hyun Nam
CCChul-Min Chon
Chul-Min Chon
KJKa-Young Jung
Ka-Young Jung
SCSun-Gyu Choi
Sun-Gyu Choi
HCHyeoksun Choi
Hyeoksun Choi
SPSung-sik Park
Sung-sik Park
An extract of the indigenous plant jack bean (Canavalia ensiformis) was used to produce calcite, an effective biomaterial. The formation of calcite using this extract was compared, under stable conditions, to that using commercially available purified urease. Xray diffraction and scanning electron microscopy were employed to elucidate the mechanism of calcite formation from the crude plant extracts. The results revealed that urease in the jack bean crude extracts catalyzed the hydrolysis of urea in liquid-state cultures. The procedure described herein is a simple and useful method of calcite mineral precipitation that does not require cultivation of microorganisms or further purification of crude extracts. This study suggests that crude extracts of Canavalia ensiformis have the potential to be used in place of purified forms of urease during remediation of cracks and to increase the strength of materials.

3 May 2022 in Sensors
Verified
Journal Article
Open access
PDF file
POPeter Ogunjinmi
Peter Ogunjinmi
SPSung-sik Park
Sung-sik Park
BKBubRyur Kim
BubRyur Kim
Dong-Eun LeeDong-Eun Lee
Dong-Eun Lee
The adoption of artificial intelligence in post-earthquake inspections and reconnaissance has received considerable attention in recent years, owing to its exponential increase in computation capabilities and inherent potential in addressing disadvantages associated with manual inspections. Herein, we present the effectiveness of automated deep learning in enhancing the assessment of damage caused by the 2017 Pohang earthquake. Six classical pre-trained convolutional neural network (CNN) models are implemented through transfer learning (TL) on a small dataset, comprising 1780 manually labeled images of structural damage. Feature extraction and fine-tuning TL methods are trained on the image datasets. The performances of various CNN models are compared on a testing image dataset. Results confirm that the MobileNet fine-tuned model offers the best performance. Therefore, the model is further developed as a web-based application for classifying earthquake damage. The severity of damage is quantified by assigning damage assessment values, derived using the CNN model and gradient-weighted class activation mapping. The web-based application can effectively and automatically classify structural damage resulting from earthquakes, rendering it suitable for decision making, such as in resource allocation, policy development, and emergency response.

29 June 2020 in Applied Sciences
Verified
Journal Article
Open access
PDF file
ZNZHENZHEN NONG
ZHENZHEN NONG
SPSung-sik Park
Sung-sik Park
SJSueng Won Jeong
Sueng Won Jeong
Dong-Eun LeeDong-Eun Lee
Dong-Eun Lee
The frequency of ground motions during earthquakes is typically in the order of a few hertz. As the earthquake-induced liquefaction of soils is widely assessed by performing laboratory tests, it is necessary to consider various loading frequencies generated by real earthquakes. The effect of loading frequency has been studied by cyclic triaxial tests; however, it has rarely been investigated by cyclic direct simple shear tests, which are more similar to the cyclic loading conditions associated with earthquakes. In this study, a series of cyclic direct simple shear tests were performed on clean sand with a relative density (Dr) of 40% (loose sand) and 80% (dense sand), obtained from Nakdong River. The parameters considered are the initial vertical effective stresses (σv0′ = 50, 100, and 200 kPa) and the loading frequencies (f = 0.05, 0.1, 0.5, and 1 Hz) to evaluate the effect of the loading frequency on the liquefaction prediction of clean sand. The results showed that the liquefaction resistance of the sand increases with the increase in the loading frequency, regardless of the initial vertical effective stress and relative density. When the loading frequency increased from 0.1 to 0.5 or 1 Hz, the maximum increase in the cyclic resistances were 15%, and 19% for loose and dense sand, respectively. For a given loading frequency, the liquefaction resistance of the sand decreased when the initial vertical effective stress increased.

1 December 2010 in Journal of Geotechnical and Geoenvironmental Engineering
Verified
Journal Article
Closed access
SPSung-sik Park
Sung-sik Park
A weakly cemented sand and gravel has been partly or entirely used in the construction of earth structures such as dams and retaining walls. Such cemented soils that are usually highly permeable can undergo repetitive wetting and drying during curing due to temporary rainfall or a change in the groundwater table. In this study, weakly cemented sand specimens with four different cement ratios were compacted at optimum water content and cured for 28 days. When the cemented sand specimens were exposed to repetitive wetting and drying during curing, their 28-day unconfined compressive strength was evaluated. Wetting for one day on the last day was found to decrease the unconfined compressive strength of cemented sand, whereas wetting for one day in the middle of curing resulted in an increase in strength. The strength reduction due to wetting on the last day decreases as the cement ratio increases. For a specimen under repetitive wetting and drying over 28-day curing, the strength increases as the number of w...

13 May 2020 in Sustainability
Verified
Journal Article
Open access
PDF file
SPSung-sik Park
Sung-sik Park
POPeter Ogunjinmi
Peter Ogunjinmi
SWSeung-wook Woo
Seung-wook Woo
Dong-Eun LeeDong-Eun Lee
Dong-Eun Lee
Conventionally, liquefaction-induced settlements have been predicted through numerical or analytical methods. In this study, a machine learning approach for predicting the liquefaction-induced settlement at Pohang was investigated. In particular, we examined the potential of an artificial neural network (ANN) algorithm to predict the earthquake-induced settlement at Pohang on the basis of standard penetration test (SPT) data. The performance of two ANN models for settlement prediction was studied and compared in terms of the R2 correlation. Model 1 (input parameters: unit weight, corrected SPT blow count, and cyclic stress ratio (CSR)) showed higher prediction accuracy than model 2 (input parameters: depth of the soil layer, corrected SPT blow count, and the CSR), and the difference in the R2 correlation between the models was about 0.12. Subsequently, an optimal ANN model was used to develop a simple predictive model equation, which was implemented using a matrix formulation. Finally, the liquefaction-induced settlement chart based on the predictive model equation was proposed, and the applicability of the chart was verified by comparing it with the interferometric synthetic aperture radar (InSAR) image.