• Tunnel and Underground Space
• /
• v.29 no.4
• /
• pp.230-242
• /
• 2019

While various automatic rock fracture survey methods have been researched, the evaluation of the accuracy of these methods raises issues due to the absence of a metric which fully expresses the similarity between automatic and manual fracture maps. Therefore, this paper proposes a geometry similarity metric which is especially designed to determine the overall similarity of fracture maps and to evaluate the accuracy of rock fracture survey methods by a single number. The proposed metric, Scanline Intersection Similarity (SIS), is derived by conducting a large number of scanline surveys upon two fracture maps using Python code. By comparing the frequency of intersections over a large number of scanlines, SIS is able to express the overall similarity between two fracture maps. The proposed metric was compared with Intersection Over Union (IoU) which is a widely used evaluation metric in computer vision. Results showed that IoU is inappropriate for evaluating the geometry similarity of fracture maps because it is overly sensitive to minor geometry differences of thin elongated objects. The proposed metric, on the other hand, reflected macro-geometry differences rather than micro-geometry differences, showing good agreement with human perception. The metric was further applied to evaluate the accuracy of a deep learning-based automatic fracture surveying method which resulted as 0.674 (SIS). However, the proposed metric is currently limited to 2D fracture maps and requires comparison with rock joint parameters such as RQD.

• Journal of Korea Multimedia Society
• /
• v.25 no.8
• /
• pp.1059-1068
• /
• 2022

Recently, research using image-based deep learning is being conducted to determine similar fonts or recommend fonts. In order to increase the accuracy in judging the similarity of Hangul fonts, a previous study was conducted to calculate the similarity according to the combination of stroke elements. In this study, we tried to solve this problem by designing an integrated model that reflects the weights for each stroke element. By comparing the results of the user's font similarity calculation conducted in the previous study and the weighted model, it was confirmed that there was no difference in the ranking of the influence of the stroke elements. However, as a result of comparison by letter sizes, it was confirmed that there was a difference in the ranking of the influence of stroke elements. Accordingly, we proposed a weighted model set separately for each font size.

• Journal of Biomedical Engineering Research
• /
• v.41 no.2
• /
• pp.94-100
• /
• 2020

The purpose of this study was to present the models for classifying the wrist X-ray images by types and for segmenting the radius automatically in each image using deep learning and to verify the learned models. The data were a total of 904 wrist X-rays with the distal radius fracture, consisting of 472 anteroposterior (AP) and 432 lateral images. The learning model was the ResNet50 model for AP/lateral image classification, and the U-Net model for segmentation of the radius. In the model for AP/lateral image classification, 100.0% was showed in precision, recall, and F1 score and area under curve (AUC) was 1.0. The model for segmentation of the radius showed an accuracy of 99.46%, a sensitivity of 89.68%, a specificity of 99.72%, and a Dice similarity coefficient of 90.05% in AP images and an accuracy of 99.37%, a sensitivity of 88.65%, a specificity of 99.69%, and a Dice similarity coefficient of 86.05% in lateral images. The model for AP/lateral classification and the segmentation model of the radius learned through deep learning showed favorable performances to expect clinical application.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2009.03a
• /
• pp.308-319
• /
• 2009

In the case of relatively good ground and construction condition in the deep excavation for the construction of subway, railway, building etc., flexible earth retaining systems are often used in an economical point of view. It is generally known that the mechanism of behavior in the flexible earth retaining system is relatively more complicated than the rigid earth retaining system. Moreover in the case of long span strut supporting system the analysis of strut axial force change becomes more difficult when the differences of ground condition and excavation work progress on both sides of excavation section are added. When deeper excavation than the specification or installation delay of supporting system is done or change of ground condition is faced due to the construction conditions during construction process, lots of axial force can be induced in some struts and that can threaten the safety of construction. This paper introduces two examples of long span deep excavation where struts and rock bolts were used as a supporting system with flexible wall structure. And the sections of two examples are 50 meters apart in one construction site, they have almost similar design and construction conditions. The characteristics of ground deformation and strut axial force change were analysed, the similarity and difference between measurement results of tow examples were compared and investigated. The effort of this article aims to improve and develop the technique of design and construction in the coming projects having similar ground condition and supporting method.

• Journal of IKEEE
• /
• v.23 no.1
• /
• pp.242-248
• /
• 2019

Fire flame and smoke detection algorithm studies are challenging task in computer vision due to the variety of shapes, rapid spread and colors. The performance of a typical sensor based fire detection system is largely limited by environmental factors (indoor and fire locations). To solve this problem, a deep learning method is applied. Because it extracts the feature of the object using several methods, so that if a similar shape exists in the frame, it can be detected as false postive. This study proposes a new algorithm to reduce false positives by using frame similarity before using deep learning to decrease the false detection rate. Experimental results show that the fire detection performance is maintained and the false positives are reduced by applying the proposed method. It is confirmed that the proposed method has excellent false detection performance.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.3
• /
• pp.591-598
• /
• 2021

It's possible to get the word-vector by the statistical SVD or deep-learning CBOW and LSTM methods and theses ones learn the contexts of forward/backward words or the sequence of following words. It's used to analyze the poems by Ki Hyung-do with similar words recommended by the word-vector showing the core images of the poetry. It seems at first sight that the words don't go well with the images but they express the similar style described by the reference words once you look close the contexts of the specific poems. The word-vector can analogize the words having the same relations with the ones between the representative words for the core images of the poems. Therefore you can analyze the poems in depth and in variety with the similarity and analogy operations by the word-vector estimated with the statistical SVD or deep-learning CBOW and LSTM methods.

• The Journal of the Korea Contents Association
• /
• v.21 no.8
• /
• pp.1-9
• /
• 2021

Deep learning in computer vision has made accelerated improvement over a short period but large-scale learning data and computing power are still essential that required time-consuming trial and error tasks are involved to derive an optimal network model. In this study, we propose a similar image classification performance improvement method based on CR (Confusion Rate) that considers only the characteristics of the data itself regardless of network optimization or data reinforcement. The proposed method is a technique that improves the performance of the deep learning model by calculating the CRs for images in a dataset with similar characteristics and reflecting it in the weight of the Loss Function. Also, the CR-based recognition method is advantageous for image identification with high similarity because it enables image recognition in consideration of similarity between classes. As a result of applying the proposed method to the Resnet18 model, it showed a performance improvement of 0.22% in HanDB and 3.38% in Animal-10N. The proposed method is expected to be the basis for artificial intelligence research using noisy labeled data accompanying large-scale learning data.

• International Journal of Internet, Broadcasting and Communication
• /
• v.9 no.4
• /
• pp.1-7
• /
• 2017

Clipart is artificial visual contents that are created using various tools such as Illustrator to highlight some information. Here, the style of the clipart plays a critical role in determining how it looks. However, previous studies on clipart are focused only on the object recognition [16], segmentation, and retrieval of clipart images using hand-craft image features. Recently, some clipart classification researches based on the style similarity using CNN have been proposed, however, they have used different CNN-models and experimented with different benchmark dataset so that it is very hard to compare their performances. This paper presents an experimental analysis of the clipart classification based on the style similarity with two well-known CNN-models (Inception Resnet V2 [13] and VGG-16 [14] and transfers learning with the same benchmark dataset (Microsoft Style Dataset 3.6K). From this experiment, we find out that the accuracy of Inception Resnet V2 is better than VGG for clipart style classification because of its deep nature and convolution map with various sizes in parallel. We also find out that the end-to-end training can improve the accuracy more than 20% in both CNN models.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.2
• /
• pp.121-129
• /
• 2021

This paper deals with a super-resolution that improves the resolution of side scan sonar images using learning-based compressive sensing. Learning-based compressive sensing combined with deep learning and compressive sensing takes a structure of a feed-forward network and parameters are set automatically through learning. In particular, we propose a method that can effectively extract additional information required in the super-resolution process through various initialization methods. Representative experimental results show that the proposed method provides improved performance in terms of Peak Signal-to-Noise Ratio (PSNR) and Structure Similarity Index Measure (SSIM) than conventional methods.

• Journal of Information Processing Systems
• /
• v.16 no.6
• /
• pp.1407-1423
• /
• 2020

A graph is a data structure consisting of nodes and edges between these nodes. Graph embedding is to generate a low dimensional vector for a given graph that best represents the characteristics of the graph. Recently, there have been studies on graph embedding, especially using deep learning techniques. However, until now, most deep learning-based graph embedding techniques have focused on unweighted graphs. Therefore, in this paper, we propose a graph embedding technique for weighted graphs based on long short-term memory (LSTM) autoencoders. Given weighted graphs, we traverse each graph to extract node-weight sequences from the graph. Each node-weight sequence represents a path in the graph consisting of nodes and the weights between these nodes. We then train an LSTM autoencoder on the extracted node-weight sequences and encode each nodeweight sequence into a fixed-length vector using the trained LSTM autoencoder. Finally, for each graph, we collect the encoding vectors obtained from the graph and combine them to generate the final embedding vector for the graph. These embedding vectors can be used to classify weighted graphs or to search for similar weighted graphs. The experiments on synthetic and real datasets show that the proposed method is effective in measuring the similarity between weighted graphs.