• Journal of Industrial Convergence
• /
• v.20 no.12
• /
• pp.195-202
• /
• 2022

Underground Geospatial Information Map Management System(UGIMMS) integrates various underground facilities in the underground space into 3D mesh data, and supports to check the 3D image and location of the underground facilities in the mobile app. However, there is a problem that it takes a long time to run in the app because various underground facilities can exist in some areas executed by the app and can be seen layer by layer. In this paper, we propose a deep learning-based K-means vertex clustering algorithm as a method to reduce the execution time in the app by reducing the size of the data by reducing the number of vertices in the 3D mesh data within the range that does not cause a problem in visibility. First, our proposed method obtains refined vertex feature information through a deep learning encoder-decoder based model. And second, the method was simplified by grouping similar vertices through K-means vertex clustering using feature information. As a result of the experiment, when the vertices of various underground facilities were reduced by 30% with the proposed method, the 3D image model was slightly deformed, but there was no missing part, so there was no problem in checking it in the app.

• Smart Media Journal
• /
• v.9 no.1
• /
• pp.16-22
• /
• 2020

In this study, a neural network method performing both Detection and Classification of diseases and insects in paprika is proposed with Multi-Tasking U-net. Paprika on farms does not have a wide variety of diseases in this study, only two classes such as powdery mildew and mite, which occur relatively frequently are made as the targets. Aiming to this, a U-net is used as a backbone network, and the last layers of the encoder and the decoder of the U-net are utilized for classification and segmentation, respectively. As the result, the encoder of the U-net is shared for both of detection and classification. The training data are composed of 680 normal leaves, 450 mite-damaged leaves, and 370 powdery mildews. The test data are 130 normal leaves, 100 mite-damaged leaves, and 90 powdery mildews. Its test results shows 89% of recognition accuracy.

• Tunnel and Underground Space
• /
• v.29 no.3
• /
• pp.184-196
• /
• 2019

This study aims to extract a 3D image of micro-cracks generated by hydraulic fracturing tests, using the deep learning method and X-ray computed tomography images. The pixel-level cracks are difficult to be detected via conventional image processing methods, such as global thresholding, canny edge detection, and the region growing method. Thus, the convolutional neural network-based encoder-decoder network is adapted to extract and analyze the micro-crack quantitatively. The number of training data can be acquired by dividing, rotating, and flipping images and the optimum combination for the image augmentation method is verified. Application of the optimal image augmentation method shows enhanced performance for not only the validation dataset but also the test dataset. In addition, the influence of the original number of training data to the performance of the deep learning-based neural network is confirmed, and it leads to succeed the pixel-level crack detection.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.27 no.1
• /
• pp.115-127
• /
• 2024

This research aimed to construct models with various structures based on the Transformer module and to perform land cover classification, thereby examining the applicability of the Transformer module. For the classification of land cover, the Unet model, which has a CNN structure, was selected as the base model, and a total of four deep learning models were constructed by combining both the encoder and decoder parts with the Transformer module. During the training process of the deep learning models, the training was repeated 10 times under the same conditions to evaluate the generalization performance. The evaluation of the classification accuracy of the deep learning models showed that the Model D, which utilized the Transformer module in both the encoder and decoder structures, achieved the highest overall accuracy with an average of approximately 89.4% and a Kappa coefficient average of about 73.2%. In terms of training time, models based on CNN were the most efficient. however, the use of Transformer-based models resulted in an average improvement of 0.5% in classification accuracy based on the Kappa coefficient. It is considered necessary to refine the model by considering various variables such as adjusting hyperparameters and image patch sizes during the integration process with CNN models. A common issue identified in all models during the land cover classification process was the difficulty in detecting small-scale objects. To improve this misclassification phenomenon, it is deemed necessary to explore the use of high-resolution input data and integrate multidimensional data that includes terrain and texture information.

• The Journal of Korea Robotics Society
• /
• v.17 no.1
• /
• pp.58-67
• /
• 2022

In this paper, we propose an approach resolving inaccuracy of the low-cost redundant manipulator workspace with low encoder and low stiffness. When the manipulators are manufactured with low-cost encoders and low-cost links, the robots can run into workspace inaccuracy issues. Furthermore, trajectory generation based on conventional forward/inverse kinematics without taking into account inaccuracy issues will introduce the risk of end-effector fluctuations. Hence, we propose an optimization for the trajectory generation method based on the DDPG (Deep Deterministic Policy Gradient) algorithm for the low-cost redundant manipulators reaching the target position in Euclidean space. We designed the DDPG algorithm minimizing the distance along with the jacobian condition number. The training environment is selected with an error rate of randomly generated joint spaces in a simulator that implemented real-world physics, the test environment is a real robotic experiment and demonstrated our approach.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.9
• /
• pp.367-374
• /
• 2021

Stock price prediction is a subject of research in various fields such as economy, statistics, computer engineering, etc. In recent years, researches on predicting the movement of stock prices by learning artificial intelligence models from various indicators such as basic indicators and technical indicators have become active. This study proposes a deep learning model that predicts the ups and downs of KOSPI from overseas indices such as S&P500, past KOSPI indices, and trading trends by KOSPI investors. The proposed model extracts a latent variable using a stacked auto-encoder to predict stock price fluctuations, and predicts the fluctuation of the closing price compared to the market price of the day by learning an LSTM suitable for learning time series data from the extracted latent variable to decide to buy or sell based on the value. As a result of comparing the returns and prediction accuracy of the proposed model and the comparative models, the proposed model showed better performance than the comparative models.

• Journal of Intelligence and Information Systems
• /
• v.29 no.2
• /
• pp.115-127
• /
• 2023

Video stabilization is one of the camera technologies that the importance is gradually increasing as the personal media market has recently become huge. For deep learning-based video stabilization, existing methods collect pairs of video datas before and after stabilization, but it takes a lot of time and effort to create synchronized datas. Recently, to solve this problem, unsupervised learning method using only unstable video data has been proposed. In this paper, we propose a network structure that learns the stabilized trajectory only with the unstable video image without the pair of unstable and stable video pair using the Convolutional Auto Encoder structure, one of the unsupervised learning methods. Optical flow data is used as network input and output, and optical flow data was mapped into grid units to simplify the network and minimize noise. In addition, to generate a stabilized trajectory with an unsupervised learning method, we define the loss function that smoothing the input optical flow data. And through comparison of the results, we confirmed that the network is learned as intended by the loss function.

• KIPS Transactions on Software and Data Engineering
• /
• v.6 no.12
• /
• pp.573-580
• /
• 2017

Deep Neural Network (DNN) is a large layered neural network which is consisted of a number of layers of non-linear units. Deep Learning which represented as DNN has been applied very successfully in various applications. However, many issues in DNN have been identified through past researches. Among these issues, generalization is the most well-known problem. A Recent study, Dropout, successfully addressed this problem. Also, Dropout plays a role as noise, and so it helps to learn robust feature during learning in DNN such as Denoising AutoEncoder. However, because of a large computations required in Dropout, training takes a lot of time. Since Dropout keeps changing an inter-layer representation during the training session, the learning rates should be small, which makes training time longer. In this paper, using mutation operation, we reduce computation and improve generalization performance compared with Dropout. Also, we experimented proposed method to compare with Dropout method and showed that our method is superior to the Dropout one.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.12
• /
• pp.1890-1897
• /
• 2021

People counting is an important technology to provide application services such as smart home, smart building, smart car, etc. Due to the social distancing of COVID-19, the people counting technology attracted public attention. People counting system can be implemented in various ways such as camera, sensor, wireless, etc. according to service requirements. People counting system using WiFi AP uses WiFi CSI data that reflects multipath information. This technology is an effective solution implementing indoor with low cost. The conventional WiFi CSI-based people counting technologies have low accuracy that obstructs the high quality service. This paper proposes a deep learning people counting system based on WiFi CSI data. Data preprocessing using auto-encoder, data augmentation that transform WiFi CSI data, and a proposed deep learning model improve the accuracy of people counting. In the experimental result, the proposed approach shows 89.29% accuracy in 6 subjects.

• Journal of IKEEE
• /
• v.27 no.4
• /
• pp.464-471
• /
• 2023

In this paper, we propose a method to reduce age distortion in facial expression image generation using StyleGAN Encoder. The facial expression image generation process first creates a face image using StyleGAN Encoder, and changes the expression by applying the learned boundary to the latent vector using SVM. However, when learning the boundary of a smiling expression, age distortion occurs due to changes in facial expression. The smile boundary created in SVM learning for smiling expressions includes wrinkles caused by changes in facial expressions as learning elements, and it is determined that age characteristics were also learned. To solve this problem, the proposed method calculates the correlation coefficient between the smile boundary and the age boundary and uses this to introduce a method of adjusting the age boundary at the smile boundary in proportion to the correlation coefficient. To confirm the effectiveness of the proposed method, the results of an experiment using the FFHQ dataset, a publicly available standard face dataset, and measuring the FID score are as follows. In the smile image, compared to the existing method, the FID score of the smile image generated by the ground truth and the proposed method was improved by about 0.46. In addition, compared to the existing method in the smile image, the FID score of the image generated by StyleGAN Encoder and the smile image generated by the proposed method improved by about 1.031. In non-smile images, compared to the existing method, the FID score of the non-smile image generated by the ground truth and the method proposed in this paper was improved by about 2.25. In addition, compared to the existing method in non-smile images, it was confirmed that the FID score of the image generated by StyleGAN Encoder and the non-smile image generated by the proposed method improved by about 1.908. Meanwhile, as a result of estimating the age of each generated facial expression image and measuring the estimated age and MSE of the image generated with StyleGAN Encoder, compared to the existing method, the proposed method has an average age of about 1.5 in smile images and about 1.63 in non-smile images. Performance was improved, proving the effectiveness of the proposed method.