• Journal of Korean Tunnelling and Underground Space Association
• /
• v.25 no.6
• /
• pp.555-567
• /
• 2023

The cracks in the tunnel are currently determined through visual inspections conducted by inspectors based on images acquired using tunnel imaging acquisition systems. This labor-intensive approach, relying on inspectors, has inherent limitations as it is subject to their subjective judgments. Recently research efforts have actively explored the use of deep learning to automatically detect tunnel cracks. However, most studies utilize public datasets or lack sufficient objectivity in the analysis process, making it challenging to apply them effectively in practical operations. In this study, we selected test datasets consisting of images in the same format as those obtained from the actual inspection system to perform an objective evaluation of deep learning models. Additionally, we introduced ensemble techniques to complement the strengths and weaknesses of the deep learning models, thereby improving the accuracy of crack detection. As a result, we achieved high recall rates of 80%, 88%, and 89% for cracks with sizes of 0.2 mm, 0.3 mm, and 0.5 mm, respectively, in the test images. In addition, the crack detection result of deep learning included numerous cracks that the inspector could not find. if cracks are detected with sufficient accuracy in a more objective evaluation by selecting images from other tunnels that were not used in this study, it is judged that deep learning will be able to be introduced to facility safety inspection.

• Korean Journal of Heritage: History & Science
• /
• v.53 no.3
• /
• pp.142-163
• /
• 2020

In the early Joseon dynasty, the meritorious vassals' portrait of Jeokgae (1467, 1476), Jeongguk (1506)·Jeongnan (1507), and Wisa (1545) were defined as the 2^nd period of the official uniform, in which the acheongsaek HeukDanryung appeared and settled. A total of seven portraits were analyzed, including three portraits of Jeokgae (1467, portrait production 1476) Son-so, Oh Jachi, and Jang Malson, and four portraits of Jeongguk (1506)·Jeongnan (1507) Yoo Sunjung, Hong Gyeongju, Lee Woo, and Yoo Hong. In the portrait of the Jeokgae wearing a Yebok HeukDanryung, in the case of the Samo, the height of the Samo, which had been raised at the beginning of King Seongjong, was lowered again, and the parietal part was rounded. The pattern was not expressed on the wide oval side wings of the Samo. In addition, the pattern was not expressed on the acheongsaek Danryung, Dapho, and Cheolrik. The colors of Dapho and Cheolrik were unified in green and red, respectively. The Mu of Danryung and Dapho was expressed as 'in and out wrinkled Mu', and the shoes were expressed in Baekhwa. Another feature is that the Jeokgae's portrait is expressed not as the Pumgye at the time of appointment, but as the Pumgye of the portrait production time. In the portrait of the Jeongguk·Jeongnan wearing a Sibok HeukDanryung, the height of the Samo lowered from the Jeokgae was maintained, while the parietal part was slightly narrowed and bent forward. The side wings of the Samo were expressed as an oval with a uniform width, and still the pattern was not expressed on the side wings. The pattern was not expressed in the costume of the Danghagwan's portrait, but the cloud pattern began to be expressed in the costume of the Dangsanggwan's portrait. And the Danryung was expressed as single Danryung without lining. In the portrait of Dangsang, the red Dapho and green Cheolrik were expressed, and in the portrait of Dangha, the green Dapho and blue Cheolrik were expressed. Therefore, it was found that the color of the undergarment was distinguished according to the Dangsang and Dangha. In Danryung and Dapho, the 'in and out wrinkled Mu' were maintained. The shoes were expressed in Heukhwa, and the rank badge and rank belt reflect the Pumgye at the time of the appointment. On the other hand, the portrait of Wisa could not be confirmed, but through the recording painting of the reign of King Meongjong, it was found that it was similar to the Danryung system of the reign of King Jungjong. Through the 2^nd period of the meritorious vassal's portrait of Jeokgae, it was confirmed that the HeukDanryung outfit began to be used as the official uniform of the meritorious vassal's portrait, and that the Pumgye of the portrait production time was reflected in the portrait. After the meritorious vassal's portrait of Jeongguk, the pattern began to be expressed in the costumes, and in particular, it was possible to distinguish Dasang and Dangha by the presence or absence of the HeukDanryung's pattern. In addition, it was confirmed that the expressed aspect of the official uniform was consistent with the practices and regulations at the time.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.11 no.1
• /
• pp.1534-1548
• /
• 1969

This experiment was conducted, making use of the 'NONG-RIM6' arecommended variety of rice for the year of 1968. Main purposes of the experiment are to explore possibilities of; a) ways and means of saving irringation water and, b) overcoming drought at the same time so that an increased yield in rice could be resulted in. Specifically, it was tried to determine the effects of the Rotation irrigation method combined with differentiated thickness of lining upon the growth and yield of rice. Some of the major findings are summarized in the following. 1) The different thicknesses show a significant relationship with the weight of 1,000 grains. In the case of 9cm lined plot, the grain weight is 23.5grams, the heaviest. Next in order is 3cm lined plot, 6cm lined plot, control plot, and wheat straw lined-plot. 2) In rice yield, it is found that there is a considerably moderate significant relationship with both the different thickness of lining and the number of irrigation, as shown in the table. 3) There is little or no difference among different plots in terms of a) physical and chemical properties of soil, b) quality of irrigation water, c) climatic conditions, and rainfalls. 4) It is found that there is a significant relationship between differences in the method of rotation irrigation and the number of ears per hill. The plot irrigated at an interval of 7 days shows 17.4 ears and plot irrigated at an interval of 6 days, 16.3 5) In vinyl-treated plots, it is shown that both yield and component elements are greatest in the case of the plot ith whole of $3cm/m^2$ Next in order are the plot with a hole of $2cm/m^2$ the plot with a hole of $1cm/m^2$ In the case of the plot with no hole it is found that both yield and component elements are decreased as compared to the control plot. 6) The irrigation water reqirement is measured for the actual irrigation days of 72 which are the number subtracted the days of rainfall of 30 from the total irrigation days of 102. It is found that the irrigation water requirement for the uncontrol plot is 1,590mm as compared to 876mm(44.9% saved) for the 9cm-lined plot, 959mm(39.7% saved) for the 6cm-lined plot 1,010mm(36% saved) for the 3cm-lined plot and 1,082mm(32% saved) for the wheat straw lined plot. In the case of the Rotation irrigation method it is found that the water requirement for the plot irrigated at an interval of 8 days is 538mm(65% saved), as compared to 617mm(61.6% saved) for plot irrigated at an interval of 7 day 672mm(57.7% saved) for plot irrigated at an interval of 6day, 746mm(53.0% saved) for the plot irrigated at an interval of 5 days, 890mm 44.0% saved) for the plot irrigated at an interval of 4 days, and 975mm(38.6% saved) for the plot irrigated at an interval of 3 days. 7) The rate of evapotranspiration is found 2.8 around the end of month of July, as compared to 2.6 at the begining of August 3.4 around the end of August and 2.6 at the begining of August 3.4 around the end of August and 2.6 at the begining of September. 8) It is found that the saturation quantity of 30mm per day is decreased to 20mm per day though the use of vinyl covering. 9) The husking rate shows 75 per cent which is considered better.

• Journal of the Korean Geotechnical Society
• /
• v.24 no.4
• /
• pp.101-114
• /
• 2008

Pressurized grouting is a common technique in geotechnical engineering applications to increase the stiffness and strength of the ground mass and to fill boreholes or void space in a tunnel lining and so on. Recently, the pressurized grouting has been applied to a soil-nailing system which is widely used to improve slope stability. Because interaction between pressurized grouting paste and adjacent ground mass is complicated and difficult to analyze, the soil-nailing design has been empirically performed in most geotechnical applications. The purpose of this study is to analyze the ground behavior induced by pressurized grouting paste with the aid of laboratory model tests. The laboratory tests are carried out for four kinds of granitic residual soils. When injecting pressure is applied to grout, the pressure measured in the adjacent ground initially increases for a while, which behaves in the way of the membrane model. With the lapse of time, the pressure in the adjacent ground decreases down to a value of residual stress because a portion of water in the grouting paste seeps into the adjacent ground. The seepage can be indicated by the fact that the ratio of water/cement in the grouting paste has decreased from a initial value of 50% to around 30% during the test. The reduction of the W/C ratio should cause to harden the grouting paste and increase the stiffness of it, which restricts the rebound of out-moved ground into the original position, and thus increase the in-situ stress by approximately 20% of the injecting pressures. The measured radial deformation of the ground under pressure is in good agreement with the expansion of a cylindrical cavity estimated by the cavity expansion theory. In-situ test revealed that the pullout resistance of a soil nailing with pressurized grouting is about 36% larger than that with regular grouting, caused by grout radius increase, residual stress effect, and/or roughness increase.