• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.311-323
• /
• 2023

Ever since ancient times, earthquakes have been a major threat to the civil infrastructures and the safety of human beings. The majority of casualties in earthquake disasters are caused by the damaged civil infrastructures but not by the earthquake itself. Therefore, the efficient and accurate post-earthquake assessment of the conditions of structural damage has been an urgent need for human society. Traditional ways for post-earthquake structural assessment rely heavily on field investigation by experienced experts, yet, it is inevitably subjective and inefficient. Structural response data are also applied to assess the damage; however, it requires mounted sensor networks in advance and it is not intuitional. As many types of damaged states of structures are visible, computer vision-based post-earthquake structural assessment has attracted great attention among the engineers and scholars. With the development of image acquisition sensors, computing resources and deep learning algorithms, deep learning-based post-earthquake structural assessment has gradually shown potential in dealing with image acquisition and processing tasks. This paper comprehensively reviews the state-of-the-art studies of deep learning-based post-earthquake structural assessment in recent years. The conventional way of image processing and machine learning-based structural assessment are presented briefly. The workflow of the methodology for computer vision and deep learning-based post-earthquake structural assessment was introduced. Then, applications of assessment for multiple civil infrastructures are presented in detail. Finally, the challenges of current studies are summarized for reference in future works to improve the efficiency, robustness and accuracy in this field.

• Journal of Preventive Medicine and Public Health
• /
• v.29 no.3 s.54
• /
• pp.521-527
• /
• 1996

The Hahn double 15 hue test is used for social and vacational aptitude test to separate strongly and mildly affected subjects among the colour vision defective persons, detected using colour vision test. However, the assessment of colour vision defect type and severity is based on the hue confusions which are represented diagrammatically on Hahn double 15 hue score sheet, this qualitative assessment of the test results have not provide a numerical score suitable for methematical analysis. This paper presented a new proposal for quantitatively scoring the Hahn double 15 hue test based on those hue confusions made by the subject. With this program large numbers of double 15 hue test results can be processed easily and rapidly, and program helps to compare the severity of specific type colour vision defect and monitor acquired colour vision defect which has various disease process, continuously.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.6
• /
• pp.2624-2630
• /
• 2013

In LED chip packaging, die bonding is a very important process which fixes the LED chip on the lead flame to provide enough strength for the next process. During the process, inspection processes are very important to detect exact locations of dispensed epoxy dots and to determine bonding status of dies whether they are lies at exact positions with sufficient bonding strength. In this study, a useful machine vision based inspection system is proposed for the LED die bonder. In the proposed system, 2 cameras are used for epoxy dot position detection and 2 cameras are sued for die attaching status determination. New vision processing algorithm is proposed, and its efficiency is verified through required field experiments. Measured position error is less than $X:-29{\mu}m$, $Y:-32{\mu}m$ and rotation error:$3^{\circ}$ using proposed vision algorithm. It is concluded that the proposed machine vision based inspection system can be successfully implemented on the developed die bonding system.

• Journal of IKEEE
• /
• v.27 no.3
• /
• pp.313-318
• /
• 2023

This paper proposes a neural network applying fine-tuning as a way to improve the performance of Video Classification based on Vision Transformer. Recently, the need for real-time video image analysis based on deep learning has emerged. Due to the characteristics of the existing CNN model used in Image Classification, it is difficult to analyze the association of consecutive frames. We want to find and solve the optimal model by comparing and analyzing the Vision Transformer and Non-local neural network models with the Attention mechanism. In addition, we propose an optimal fine-tuning neural network model by applying various methods of fine-tuning as a transfer learning method. The experiment trained the model with the UCF101 dataset and then verified the performance of the model by applying a transfer learning method to the UTA-RLDD dataset.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.41 no.10
• /
• pp.949-957
• /
• 2017

A laser-vision-based height measurement system is developed and implemented for the inspection of two-step variable valve lift module assemblies. The proposed laser-vision sensor module is designed based on the principle of laser triangulation. This paper summarizes the work on 3D point cloud data collection and height difference measurements. The configuration of the measurement system and the proposed height measurement algorithm are described and analyzed in detail. Additional measurement experiments on the height differences of valves and lash adjusters of a two-step variable valve lift module were implemented repeatedly to evaluate the accuracy and repeatability of the proposed measurement system. Experimental results show that the proposed laser-vision-based height measurement system achieves high accuracy, repeatability, and stabilization for the inspection of two-step variable valve lift module assemblies.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.35 no.5
• /
• pp.259-266
• /
• 2022

In this study, we developed a deep learning and vision sensor-based assembly performance evaluation method isfor prefabricated steel structures. The assembly parts were segmented using a modified version of the receptive field block convolution module inspired by the eccentric function of the human visual system. The quality of the assembly was evaluated by detecting the bolt holes in the segmented assembly part and calculating the bolt hole positions. To validate the performance of the evaluation, models of standard and defective assembly parts were produced using a 3D printer. The assembly part segmentation network was trained based on the 3D model images captured from a vision sensor. The sbolt hole positions in the segmented assembly image were calculated using image processing techniques, and the assembly performance evaluation using the k-nearest neighbor algorithm was verified. The experimental results show that the assembly parts were segmented with high precision, and the assembly performance based on the positions of the bolt holes in the detected assembly part was evaluated with a classification error of less than 5%.

• Smart Media Journal
• /
• v.12 no.9
• /
• pp.28-37
• /
• 2023

The main factor affecting the decline in rice yield is damage caused by localized heavy rains or typhoons. The method of analyzing the rice lodging area is difficult to obtain objective results based on visual inspection and judgment based on field surveys visiting the affected area. it requires a lot of time and money. In this paper, we propose the method of estimation and diagnosis for rice lodging areas using a Vision Transformer-based Segformer for RGB images, which are captured by unmanned aerial vehicles. The proposed method estimates the lodging, normal, and background area using the Segformer model, and the lodging rate is diagnosed through the rice field inspection criteria in the seed industry Act. The diagnosis result can be used to find the distribution of the rice lodging areas, to show the trend of lodging, and to use the quality management of certified seed in government. The proposed method of rice lodging area estimation shows 98.33% of mean accuracy and 96.79% of mIoU.

• Journal of Sensor Science and Technology
• /
• v.32 no.5
• /
• pp.320-327
• /
• 2023

In this study, an FPGA-based sealer inspection system was developed to inspect the sealer applied to install vehicle glass on a car body. The sealer is a liquid or paste-like material that promotes adhesion such as sealing and waterproofing for mounting and assembling vehicle parts to a car body. The system installed in the existing vehicle design parts line does not detect the sealer in the glass rotation section and takes a long time to process. This study developed a line laser camera sensor and an FPGA vision signal processing module to solve this problem. The line laser camera sensor was developed such that the resolution and speed of the camera for data acquisition could be modified according to the irradiation angle of the laser. Furthermore, it was developed considering the mountability of the entire system to prevent interference with the sealer ejection machine. In addition, a vision signal processing module was developed using the Zynq-7020 FPGA chip to improve the processing speed of the algorithm that converted the profile to the sealer shape image acquired from a 2D camera and calculated the width and height of the sealer using the converted profile. The performance of the developed sealer application inspection system was verified by establishing an experimental environment identical to that of an actual automobile production line. The experimental results confirmed the performance of the sealer application inspection at a level that satisfied the requirements of automotive field standards.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.56-59
• /
• 1997

Vision system is mainly consist of three parts of perception, action. One of these parts, perception system detects visual target in surrounding environment. Block-based motion estimation with compensation is one of the popular approaches without accuracy. The hierarchical method the optical flow with gradient is used to improve optical flow time delay.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2012.10a
• /
• pp.47-48
• /
• 2012