• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.2
• /
• pp.193-201
• /
• 2021

The welding has a high proportion of the production and drying of ships or offshore plants. Non-destructive testing is carried out to verify the quality of welds in Korea, radiography test (RT) is mainly used. Currently, most shipyards adopt analog-type techniques to print the films through the shoot of welding parts. Therefore, the time required from radiography test to pass or fail judgment is long and complex, and is being manually carried out by qualified inspectors. To improve this problem, this paper covers a platform for scanning and digitalizing RT films occurring in shipyards with high resolution, accumulating them in management servers, and applying artificial intelligence (AI) technology to detect welding defects. To do this, we describe the process of designing and developing RT film scanning equipment, welding inspection information integrated management platform, fault reading algorithms, visualization software, and testing and verification of each developed element in conjunction.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.41 no.6
• /
• pp.751-760
• /
• 2021

Most bridges in Korea are over 20 years old, and many problems linked to their deterioration are being reported. The current practice for bridge inspection mainly depends on expert evaluation, which can be subjective. Recent studies have introduced data-driven methods using building information modeling, which can be more efficient and objective, but these methods require manual procedures that consume time and money. To overcome this, this study developed an image-based automaticbridge component classification network to reduce the time and cost required for converting the visual information of bridges to a digital model. The proposed method comprises two convolutional neural networks. The first network estimates the type of the bridge based on the superstructure, and the second network classifies the bridge components. In avalidation test, the proposed system automatically classified the components of 461 bridge images with 96.6 % of accuracy. The proposed approach is expected to contribute toward current bridge maintenance practice.

• Journal of Broadcast Engineering
• /
• v.27 no.2
• /
• pp.228-239
• /
• 2022

In this paper, we propose an unmanned vehicle scratch detection deep learning model for car sharing services. Conventional scratch detection models consist of two steps: 1) a deep learning module for scratch detection of images before and after rental, 2) a manual matching process for finding newly generated scratches. In order to build a fully automatic scratch detection model, we propose a one-step unmanned scratch detection deep learning model. The proposed model is implemented by applying transfer learning and fine-tuning to the deep learning model that detects changes in satellite images. In the proposed car sharing service, specular reflection greatly affects the scratch detection performance since the brightness of the gloss-treated automobile surface is anisotropic and a non-expert user takes a picture with a general camera. In order to reduce detection errors caused by specular reflected light, we propose a preprocessing process for removing specular reflection components. For data taken by mobile phone cameras, the proposed system can provide high matching performance subjectively and objectively. The scores for change detection metrics such as precision, recall, F1, and kappa are 67.90%, 74.56%, 71.08%, and 70.18%, respectively.

• Journal of Broadcast Engineering
• /
• v.27 no.3
• /
• pp.391-401
• /
• 2022

As deep learning-based object detection and recognition research have been developed recently, the scope of application to industry and real life is expanding. But deep learning-based systems in the construction system are still much less studied. Calculating materials in the construction system is still manual, so it is a reality that transactions of wrong volumn calculation are generated due to a lot of time required and difficulty in accurate accumulation. A fast and accurate automatic drawing recognition system is required to solve this problem. Therefore, we propose an AI-based automatic drawing recognition accumulation system that detects and recognizes steel materials in construction drawings. To accurately detect steel materials in construction drawings, we propose data augmentation techniques and spatial attention modules for improving small object detection performance based on YOLOv4. The detected steel material area is recognized by text, and the number of steel materials is integrated based on the predicted characters. Experimental results show that the proposed method increases the accuracy and precision by 1.8% and 16%, respectively, compared with the conventional YOLOv4. As for the proposed method, Precision performance was 0.938. The recall was 1. Average Precision AP_0.5 was 99.4% and AP_0.5:0.95 was 67%. Accuracy for character recognition obtained 99.9.% by configuring and learning a suitable dataset that contains fonts used in construction drawings compared to the 75.6% using the existing dataset. The average time required per image was 0.013 seconds in the detection, 0.65 seconds in character recognition, and 0.16 seconds in the accumulation, resulting in 0.84 seconds.

• Journal of the Korea Institute of Building Construction
• /
• v.21 no.6
• /
• pp.665-676
• /
• 2021

As the number of deteriorated buildings increases, the importance of safety diagnosis and maintenance of buildings has been rising. Existing visual investigations and building safety diagnosis objectivity and reliability are poor due to their reliance on the subjective judgment of the examiner. Therefore, this study presented the limitations of the previously conducted appearance investigation and proposed 3D Point Cloud data to increase the accuracy of existing detailed inspection data. In addition, this study conducted a calculation of an objective building safety grade using a Deep-Neural Network(DNN) structure. The DNN structure is generated using the existing detailed inspection data and precise safety diagnosis data, and the safety grade is calculated after applying the state evaluation data obtained using a 3D Point Cloud model. This proposed process was applied to 10 deteriorated buildings through the case study, and achieved a time reduction of about 50% compared to a conventional manual safety diagnosis based on the same building area. Subsequently, in this study, the accuracy of the safety grade calculation process was verified by comparing the safety grade result value with the existing value, and a DNN with a high accuracy of about 90% was constructed. This is expected to improve economic feasibility in the future by increasing the reliability of calculated safety ratings of old buildings, saving money and time compared to existing technologies.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.6
• /
• pp.363-369
• /
• 2021

High-definition road maps are used as the basic infrastructure for autonomous vehicles, so the latest road information must be quickly reflected. However, the current drawing and structural editing process of high-definition road maps are manually performed. In addition, it takes the longest time to generate road lanes, which are the main construction targets. In this study, the point cloud of the road lane markings, in which color types(white, blue, and yellow) were predicted through the PointNet model pre-trained in previous studies, were used as input data. Based on the point cloud, this study proposed a methodology for automatically drawing and structural editing of the layer of road lane markings. To verify the usability of the 3D vector data constructed through the proposed methodology, the accuracy was analyzed according to the quality inspection criteria of high-definition road maps. In the positional accuracy test of the vector data, the RMSE (Root Mean Square Error) for horizontal and vertical errors were within 0.1m to verify suitability. In the structural editing accuracy test of the vector data, the structural editing accuracy of the road lane markings type and kind were 88.235%, respectively, and the usability was verified. Therefore, it was found that the methodology proposed in this study can efficiently construct vector data of road lanes for high-definition road maps.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.40 no.5
• /
• pp.413-419
• /
• 2022

As the development of the underground space becomes active, safety accidents related to the underground are frequently occurring in recent years. In this regard, the Ministry of Land, Infrastructure and Transport is enforcing the 『Special Act on Underground Safety Management』 (enforced on January 1, 2018, hereafter referred to as the Underground Safety Act). Among the core contents of the Underground Safety Act, underground facilities(water supply, sewage, gas, power, communication, heating) buried underground, underground structures(subway, underpass, underpass, underground parking lot, underground shopping mall, common area), ground (Drilling, wells, geology) of 15 types of underground information can be checked at a glance on a three-dimensional basis by constructing an integrated underground spatial map and using it. The purpose of this study is to develop a program that can quickly inspect the three-dimensional model after creating a three-dimensional underground structure data among the underground spatial integration maps. To this end, we first investigated and reviewed the domestic and foreign status of technology that generates and automatically inspects 3D underground structure data. A quality inspection program was developed. Through this study, it is judged that it will be meaningful as a basic research for improving the quality of underground structures on the integrated map of underground space by automating more than 98% of the 3D model inspection process, which is currently being conducted manually.

• KOREAN JOURNAL OF PACKAGING SCIENCE & TECHNOLOGY
• /
• v.28 no.3
• /
• pp.183-192
• /
• 2022

With the growth of the packaging industry, demand on the packaging printing comes in various forms. Customers' orders are diversifying and the standards for quality are increasing. Offset printing is mainly used in the packaging printing since it is easy to print in large quantities. However, productivity of the offset printing decreases when printing various order. This is because it takes time to change colors for each printing unit. Therefore, scheduling that minimizes the color replacement time and shortens the overall makespan is required. By the existing manual method based on workers' experience or intuition, scheduling results may vary for workers and this uncertainty increase the production cost. In this study, we propose an automated scheduling method of parallel offset printing process for packaging printing. We decompose the original problem into assigning and sequencing orders, and ink arrangement for printing problems. Vehicle routing problem and assignment problem are applied to each part. Mixed integer programming is used to model the problem mathematically. But it needs a lot of computational time to solve as the size of the problem grows. So guided local search algorithm is used to solve the problem. Through actual data experiments, we reviewed our method's applicability and role in the field.

• The Korean Journal of Applied Statistics
• /
• v.35 no.3
• /
• pp.407-419
• /
• 2022

The semiconductor fabrication process is complex and time-consuming. There are sometimes errors in the process, which results in defective die on the wafer bin map (WBM). We can detect the faulty WBM by finding some patterns caused by dies. When one manually seeks the failure on WBM, it takes a long time due to the enormous number of WBMs. We suggest a two-step approach to discover the probable pattern on the WBMs in this paper. The first step is to separate the normal WBMs from the defective WBMs. We adapt a hierarchical clustering for de-noising, which nicely performs this work by wisely tuning the number of minimum points and the cutting height. Once declared as a faulty WBM, then it moves to the next step. In the second step, we classify the patterns among the defective WBMs. For this purpose, we extract features from the WBM. Then machine learning algorithm classifies the pattern. We use a real WBM data set (WM-811K) released by Taiwan semiconductor manufacturing company.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.24 no.4
• /
• pp.1-25
• /
• 2021

In order to respond and prevent underground safety accidents, the Korean government has been preparing a submitting completion drawing web system project for converting the current manual-based 3D Underground Geospatial Map construction and its update system to an automation-based 3D underground information construction. However, research on standard drawings required for the automatic update of 3D underground structures is insufficient, so detailed research is needed. In this research paper, a standard map-based 3D underground structure construction plan was presented for the six types of underground structures constituting the 3D Underground Geospatial Map, enabling rapid and accurate drawing data creation and systematically 3D underground structure drawing data could be managed. In addition, we developed a 3D construction drawing tool that can be used in underground information practice so that ordinary CAD program users can easily produce processing drawings. The results derived from this paper are expected to be major reference materials for the establishment of standard frameworks and practical application guidelines for the construction of 3D underground structures in the future.