• Journal of Software Assessment and Valuation
• /
• v.15 no.1
• /
• pp.63-70
• /
• 2019

These days, as Interest in Image recognition with deep learning is increasing, there has been a lot of research in image recognition using deep learning. In this study, we propose a system for classifying rocks through rock images of 18 types of rock(6 types of igneous, 6 types of metamorphic, 6 types of sedimentary rock) which are addressed in the high school curriculum, using CNN model based on Tensorflow, deep learning open source framework. As a result, we developed a classifier to distinguish rocks by learning the images of rocks and confirmed the classification performance of rock classifier. Finally, through the mobile application implemented, students can use the application as a learning tool in classroom or on-site experience.

• Explosives and Blasting
• /
• v.8 no.4
• /
• pp.30-43
• /
• 1990

본고는 현존하는 암석과 암반의 풍화분류방법에 대한 고찰 및 이러한 방법들을 실제로 우리나라의 토목현장에 적용할 경우 발생되는 여러 가지 문제점을 서술하고, 이에 따른 문제점 극복을 위해 우리나라의 여건에 맞도록 암석 및 암반의 풍화상태를 조사 기술하므로서 각종 토목공사에서 보다 유용하게 사용할 수 있는 풍화분류방법을 제시한다.

• Journal of the Korean earth science society
• /
• v.35 no.6
• /
• pp.467-483
• /
• 2014

This study aimed to investigate the classification ability with naked eyes according to the understanding level about rocks of pre-service science teachers. We developed a questionnaire concerning misconception about minerals and rocks. The participants were 132 pre-service science teachers. Data were analyzed using Rasch model. Participants were divided into a master group and a novice group according to their understanding level. Seventeen rocks samples (6 igneous, 5 sedimentary, and 6 metamorphic rocks) were presented to pre-service science teachers to examine their classification ability, and they classified the rocks according to the criteria we provided. The study revealed three major findings. First, the pre-service science teachers mainly classified rocks according to textures, color, and grain size. Second, while they relatively easily classified igneous rocks, participants were confused when distinguishing sedimentary and metamorphic rocks from one another by using the same classification criteria. On the other hand, the understanding level of rocks has shown a statistically significant correlation with the classification ability in terms of the formation mechanism of rocks, whereas there was no statistically significant relationship found with determination of correct name of rocks. However, this study found that there was a statistically significant relationship between the classification ability with regard to formation mechanism of rocks and the determination of correct name of rocks.

• The Journal of the Petrological Society of Korea
• /
• v.28 no.4
• /
• pp.307-317
• /
• 2019

Petrological studies were performed on the Sanbangsan lava dome, located in the southwest of Jeju Island Volcanic Field. According to the lava ejection method, it is 'an internal primitive form' that is gradually pushed up and expanded by continuous magma injection from the bottom to the top of the vent and it corresponds to the 'low lava dome'. The rocks are partly plotted in the field of benmoreite, but mostly plotted in the field of trachyte of the Cox et al.(1979) classification diagram, and also mainly plotted in the field of trachyte of Le Maitre et al.(2002) and Zr/TiO₂-Nb/Y classification diagram. Therefore, the expression that described the rock of Sanbangsan lava dome as 'trachy-andesite' should be corrected to 'trachyte'. The volcanic rocks that consists in the Sanbangsan lava dome are trachyte containing normative quartz and shows differentiation trend in the range of 59.75-63.46 wt.% SiO₂.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2003.06b
• /
• pp.51-64
• /
• 2003

Bieniawski에 의해 개발된 RMR은 암석강도 및 불연속면, 지하수 등의 6개 인자에 따라 분류되어, 이들을 합산하여 결정된다. RMR 분류법은 각 요소들에 대한 평가가 비교적 쉽고, 다양한 응용을 거쳐 여러 분야에 적용되어 국내에서도 가장 널리 사용하고 있는 암반분류 방법 중의 하나이다. RMR 분류결과는 터널의 유지시간, 최대 무지보폭의 예측, 지보량 산정, 암반의 물리적 특성값 예측 등에 적용될 수 있다. 또한 RMR 분류법을 사면안정, 댐 기초, 심부 광산 등에 적용하거나, RMR 분류법의 미비한 부분을 보완하기 위한 여러 가지 수정방법이 제시되었다.

• Economic and Environmental Geology
• /
• v.55 no.3
• /
• pp.309-316
• /
• 2022

In order to minimize the human and time consumption required for rock classification, research on rock classification using artificial intelligence (AI) has recently developed. In this study, basic volcanic rocks were subdivided by using polarizing microscope thin section images. A convolutional neural network (CNN) model based on Tensorflow and Keras libraries was self-producted for rock classification. A total of 720 images of olivine basalt, basaltic andesite, olivine tholeiite, trachytic olivine basalt reference specimens were mounted with open nicol, cross nicol, and adding gypsum plates, and trained at the training : test = 7 : 3 ratio. As a result of machine learning, the classification accuracy was over 80-90%. When we confirmed the classification accuracy of each AI model, it is expected that the rock classification method of this model will not be much different from the rock classification process of a geologist. Furthermore, if not only this model but also models that subdivide more diverse rock types are produced and integrated, the AI model that satisfies both the speed of data classification and the accessibility of non-experts can be developed, thereby providing a new framework for basic petrology research.

• Tunnel and Underground Space
• /
• v.30 no.5
• /
• pp.462-472
• /
• 2020

Rock classification is fundamental discipline of exploring geological and geotechnical features in a site, which, however, may not be easy works because of high diversity of rock shape and color according to its origin, geological history and so on. With the great success of convolutional neural networks (CNN) in many different image-based classification tasks, there has been increasing interest in taking advantage of CNN to classify geological material. In this study, a feasibility of the deep CNN is investigated for automatically and accurately identifying rock types, focusing on the condition of various shapes and colors even in the same rock type. It can be further developed to a mobile application for assisting geologist in classifying rocks in fieldwork. The structure of CNN model used in this study is based on a deep residual neural network (ResNet), which is an ultra-deep CNN using in object detection and classification. The proposed CNN was trained on 10 typical rock types with an overall accuracy of 84% on the test set. The result demonstrates that the proposed approach is not only able to classify rock type using images, but also represents an improvement as taking highly diverse rock image dataset as input.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.6 no.3
• /
• pp.237-245
• /
• 2004

In designing a tunnel, RMR has been widely used to classify rock mass and to decide the support pattern according to the class of rock mass. However, this RMS system can't help relying on the empirical judgment of engineers who use variables which can be obtained only through consideration of the site conditions. In actuality, it is impossible to consider all the rating factors of RMS when using RMR system at the stage of designing. Therefore, in order to confirm possibility of RMR by use of only the quantitative factors for designing, this paper has done discriminant analysis. Rock strength or RQD has high coefficient of correlation with RMR value, and in consideration of the existing standards for rock mass classification, rock intensity and RQD are important factors for classification of rock mass. Through rock mass classification by the existing RMR system and rock mass classification by the discriminant analysis which has considered two variables only, the discriminant analysis using the rock intensity as an independent variable has shown 74.8% accuracy while the discriminant analysis using RQD as an independent variable has shown 74.3% accuracy. In case of the discriminant analysis which has considered both rock intensity and RQD, it has shown 82.5% accuracy. The existing cases have shown 40.3% accuracy at the stage of designing in which all the RMR factors are considered. It means that at the stage of designing, RMR system can work only with the rock intensity and RQD.

• Proceedings of the Petrological Society of Korea Conference
• /
• 2001.06a
• /
• pp.100-102
• /
• 2001

• Economic and Environmental Geology
• /
• v.37 no.3
• /
• pp.355-364
• /
• 2004

A study on the weathering grade classification has been performed for granite and granite gneiss in Korea. The qualitative classification criteria of weathering were reviewed and then modified with field studies for the weathered rock masses. The thin section observations and XRD analyses for the different weathering grades rock samples showed the petrographical and petrophysical difference with respect to the weathering : the proportion of weathering-resistant minerals suck at quartz and orthoclase has a tendency to increase with the development of weathering, but that of weathering-sensible minerals such as anorthite and biotite is decreased. The ranges of physical and mechanical rock properties for different weathering grades were obtained from the laboratory rock tests and field tests for the studied rocks. And then, along with $RDI_{sq}$(Fookes et al., 1988), the weathering index $I_{a}$, (Woo, 2003) has been developed in this study to demarcate the weathering grade. Those two indices rely mainly on the water absorption ratio of rock and on the different rock strength. The range of these weathering indices have been determined with the physical and mechanical rock properties that can be obtained from simple field or laboratory tests in 4 grades $I_{a}$＞ 7 for F, 3.5 ＜ $I_{a}$ ＜ 10 for SW, 1.0 $I_{a}$＜ 6.0 for MW and $I_{a}$＜ 2.5 for HW. Consequently, the weathering index could be utilized to classify quantitatively the rock weathering grade, especially for the studied granites and the granite gneiss in Korea.