• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.39 no.5
• /
• pp.631-636
• /
• 2019

Over the last decade, many researchers have investigated a number of vision-based construction object detection algorithms for the purpose of construction site monitoring. However, previous methods require the ground truth labeling, which is a process of manually marking types and locations of target objects from training image data, and thus a large amount of time and effort is being wasted. To address this drawback, this paper proposes a vision-based construction object detection framework that employs an active learning technique while reducing manual labeling efforts. For the validation, the research team performed experiments using an open construction benchmark dataset. The results showed that the method was able to successfully detect construction objects that have various visual characteristics, and also indicated that it is possible to develop the high performance of an object detection model using smaller amount of training data and less iterative training steps compared to the previous approaches. The findings of this study can be used to reduce the manual labeling processes and minimize the time and costs required to build a training database.

• Journal of Internet Computing and Services
• /
• v.21 no.6
• /
• pp.23-31
• /
• 2020

This study was carried out to generate various images of railroad surfaces with random defects as training data to be better at the detection of defects. Defects on the surface of railroads are caused by various factors such as friction between track binding devices and adjacent tracks and can cause accidents such as broken rails, so railroad maintenance for defects is necessary. Therefore, various researches on defect detection and inspection using image processing or machine learning on railway surface images have been conducted to automate railroad inspection and to reduce railroad maintenance costs. In general, the performance of the image processing analysis method and machine learning technology is affected by the quantity and quality of data. For this reason, some researches require specific devices or vehicles to acquire images of the track surface at regular intervals to obtain a database of various railway surface images. On the contrary, in this study, in order to reduce and improve the operating cost of image acquisition, we constructed the 'Defective Railroad Surface Regeneration Model' by applying the methods presented in the related studies of the Generative Adversarial Network (GAN). Thus, we aimed to detect defects on railroad surface even without a dedicated database. This constructed model is designed to learn to generate the railroad surface combining the different railroad surface textures and the original surface, considering the ground truth of the railroad defects. The generated images of the railroad surface were used as training data in defect detection network, which is based on Fully Convolutional Network (FCN). To validate its performance, we clustered and divided the railroad data into three subsets, one subset as original railroad texture images and the remaining two subsets as another railroad surface texture images. In the first experiment, we used only original texture images for training sets in the defect detection model. And in the second experiment, we trained the generated images that were generated by combining the original images with a few railroad textures of the other images. Each defect detection model was evaluated in terms of 'intersection of union(IoU)' and F1-score measures with ground truths. As a result, the scores increased by about 10~15% when the generated images were used, compared to the case that only the original images were used. This proves that it is possible to detect defects by using the existing data and a few different texture images, even for the railroad surface images in which dedicated training database is not constructed.

• Journal of the Korean Institute of Gas
• /
• v.15 no.5
• /
• pp.7-12
• /
• 2011

LPG(Liquefied Petroluem Gas) Vehicles in metropolitan area being applied to improve air quality and have been proven effective for the reduction of air pollution. These gas stations are required to safe the storage tank because of possibility of causing huge loss of life and property. While storage tanks above ground have potential risk of explosion if fire breaks out and those under-ground are difficult to inspect due to poor accessibility neither above nor under-ground tank can serve us well. This study used the RPS-TRIZ (Rapidly Problem Solving-Teoriya Resheniya Izobretatelskikh Zadatch) technique and suggested the use of under-ground containment storage tank as a solution for safety issues and safety inspection.

• Journal of the Korean Applied Science and Technology
• /
• v.38 no.3
• /
• pp.727-734
• /
• 2021

The purpose of this study was to analyze correlation between kinematic variables and ground slope perception ability during golf putting. 16 collegiate golfers were participated. To collect and analyze the kinematic data of putter head, SAMPutt lab was used(70 Hz). It was performed Pearson's correlation analysis using SPSS. The level of significance was at .05. As a results, right was correlated with backswing in flat. Toe-up was correlated with follow-through and left was correlated with aim, backswing, impact, follow-through, and loft in slice 1°. Toe-down was correlated with aim, backswing, impact in slice 2°. Toe-up was correlated with follow-through in hook 2°. In conclusion, it is important to improve the putting stroke through repetitive ground slope perception training.

• Journal of Fisheries and Marine Sciences Education
• /
• v.11 no.1
• /
• pp.69-87
• /
• 1999

This study had been analysed the establishment, the closing and the actual conditions of the fisheries school according to the four times revision and promulgation of the law of education of korea which were devided by the early term (the first Chosun educational decree), the middle term (the second Chosun educational decree) and the last term (the third and fourth Chosun educational decree), and also been investigated how the fisheries education of the school had been acted to the mobocracy and the assimilation policy and together with it's back ground through this analysis. The aim of this study is to contribute to the study of the history of the fisheries school education, analysis of change of the fisheries school education according to practical application of the Chosun educational decree. The summary of the characteristic of the each term are as under. First, in the early term of the fisheries education under the first Chosun educational decree, Experts were despatched to the each province with donated money from emperor and opened seasonal the fishing training centre, a kind of social fisheries education, and trained directly fishing technic and were going to train fisheries trainees regularly under name of elementary fisheries school. Japanese imperialism attached great importance to the vocational education in order to snack economical products from colonized Korea but actually had a purpose to train low quality technician who follow blindly their colony policy of Japanese imperialism. The fisheries schools in the circle of system in early time of Japanese imperialism, Kunsan public elementary fisheries school was established in April 1915, Yosu public elementary fisheries school was established ill May 1917 and Kyungnam Tongyoung training school was established in March 1917. Secondly, After 3.1 movement, the Japanese imperialism established an appeasement policy so called cultural politics and continued assimilation policy with skilful methods. After revision of the second Chosun educational decree, the Elementary vocational school was changed as the vocational training school. The school of fisheries education in middle of Japanese imperialism trained low quality technicians to snack fisheries resources from colonized Korea. After the middle of Japanese imperialism they paid attention on training fisheries technician through fisheries school rather than training school. With high interest and crowded volunteers, Kunsan public fisheries school was promoted in 1922, Tongyoung public fisheries school was promoted in 1923, Yongampo fisheries training school established in 1922 was promoted as Yongampo public fisheries school in 1926. Thirdly, in the time of the third and fourth Chosun educational decree, the end of Japanese imperialism, they met Pacific war after Japan vs China war. During the war time they considered the vocational school as the source of supply for materials and manpower and consequently had to expanded vocational education and systematically despatched students to war field and practiced military training. In 1938, Namhae public fisheries school was established and Chungjin fisheries school was permitted. But in order to supply manpower to Pacific war, the study period of Yosu public fisheries school was shorten from 5 years to 4 year in 1943 and also that of Tongyoung public fisheries school shorten in 1944.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.22 no.1
• /
• pp.140-153
• /
• 2019

Cloase-range image information from drones and ground-based camera has been frequently used in the field of disaster mitigation with 3D modeling and mapping. In addition, the utilization of virtual reality(VR) is being increased by implementing realistic 3D models with the VR technology simulating disaster circumstances in large scale. In this paper, we created a VR training program by extracting realistic 3D models from close-range images from unmanned aircraft and digital camera on hand and observed several issues occurring during the implementation and the effectiveness in the case of a VR application in training for disaster mitigation. First of all, we built up a scenario of disaster and created 3D models after image processing with the close-range imagery. The 3D models were imported into Unity, a software for creation of augmented/virtual reality, as a background for android-based mobile phones and VR environment was created with C#-based script language. The generated virtual reality includes a scenario in which the trainer moves to a safe place along the evacuation route in the event of a disaster, and it was considered that the successful training can be obtained with virtual reality. In addition, the training through the virtual reality has advantages relative to actual evacuation training in terms of cost, space and time efficiencies.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.55 no.4
• /
• pp.363-371
• /
• 2019

Thousands of demersal fishes inhabit in the waters around Korea and most of them are overexploited. One of reasons is technological development, which increases the efficiency of the vessels continuously. The analysis was conducted to identify the change of fishing power index to develop the vessel and gear technology that may have improved the fishing efficiency of the Danish seine fishery from 1960s to 2010s. Gross tonnage was decreased stably, but the horse power was increased annually. The length of ground rope, warp and hand rope was somewhat longer, but changed a little. Color fish finder was utilized from the mid-1960s and positioning system was used five years later. A hydraulic line hauler were introduced in the mid-1980s, and supply rate was gradually increased. Surveys on the supply and upgrading of fishing equipment utilized visiting researchers. Therefore, the relative fishing power index in the Danish seine fishery increased stably from 1.0 in 1970 to 1.0 in 1980, to 1.2 in 1990, to 1.3 in 2000 and to 1.3 in 2010. The results are expected to contribute to reasonable fisheries stock management.

• Korean Journal of Geomatics
• /
• v.5 no.1
• /
• pp.27-34
• /
• 2005

Unsupervised classification is an important area of research in image processing because supervised classification has the disadvantages such as long task-training time and high cost and low objectivity in training information. This paper focuses on unsupervised classification, which can extract ground object information with the minimum 'Spectral Angle Distance' operation on be behalf of 'Spectral Euclidian Distance' in the clustering process. Unlike previous studies, our algorithm uses the unit vector, not the spectral distance, to compute the cluster mean, and the Single-Pass algorithm automatically determines the seed points. Atmospheric correction for more accurate results was adapted on the Hyperion data and the results were analyzed. We applied the algorithm to the Hyperion and ETM+ data and compared the results with K-Means and the former USAM algorithm. From the result, USAM classified the water and dark forest area well and gave more accurate results than K-Means, so we believe that the 'Spectral Angle' can be one of the most accurate classifiers of not only multispectral images but hyperspectral images. And also the unit vector can be an efficient technique for characterizing the Remote Sensing data.

• Journal of Korean Society for Geospatial Information Science
• /
• v.8 no.2 s.16
• /
• pp.23-33
• /
• 2000

On this study, Among the classification methods of land cover using satellite imagery, we compared the classification accuracy of Neural Network Classifier and that of Maximum Likelihood Classifier which has the characteristics of parametric and non-parametric classification method. In the assessment of classification accuracy, we analyzed the classification accuracy about testing area as well as training area that many analysts use generally when assess the classification accuracy. As a result, Neural Network Classifier is superior to Maximum Likelihood Classifier as much as 3% in the classification of training data. When ground reference data is used, we could get poor result from both of classification methods, but we could reach conclusion that the classification result of Neural Network Classifier is superior to the classification result of Maximum Likelihood Classifier as much as 10%.

• Journal of Korean Society for Geospatial Information Science
• /
• v.8 no.2 s.16
• /
• pp.47-60
• /
• 2000

On this study, Among the classification methods of land cover using satellite imagery, we compared the classification accuracy of Neural Network Classifier and that of Maximum Likelihood Classifier which has the characteristics of parametric and non-parametric classification method. In the assessment of classification accuracy, we analyzed the classification accuracy about testing area as well as training area that many analysts use generally when assess the classification accuracy. As a result, Neural Network Classifier is superior to Maximum Likelihood Classifier as much as 3% in the classification of training data. When ground reference data is used, we could get poor result from both of classification methods, but we could reach conclusion that the classification result of Neural Network Classifier is superior to the classification result of Maximum Likelihood Classifier as much as 10%.