• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.53-62
• /
• 2022

The deployment of advanced structural health monitoring (SHM) systems in large-scale civil structures collects large amounts of data. Note that these data may contain multiple types of anomalies (e.g., missing, minor, outlier, etc.) caused by harsh environment, sensor faults, transfer omission and other factors. These anomalies seriously affect the evaluation of structural performance. Therefore, the effective analysis and mining of SHM data is an extremely important task. Inspired by the deep learning paradigm, this study develops a novel generative adversarial network (GAN) and convolutional neural network (CNN)-based data anomaly detection approach for SHM. The framework of the proposed approach includes three modules : (a) A three-channel input is established based on fast Fourier transform (FFT) and Gramian angular field (GAF) method; (b) A GANomaly is introduced and trained to extract features from normal samples alone for class-imbalanced problems; (c) Based on the output of GANomaly, a CNN is employed to distinguish the types of anomalies. In addition, a dataset-oriented method (i.e., multistage sampling) is adopted to obtain the optimal sampling ratios between all different samples. The proposed approach is tested with acceleration data from an SHM system of a long-span bridge. The results show that the proposed approach has a higher accuracy in detecting the multi-pattern anomalies of SHM data.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.4
• /
• pp.587-594
• /
• 2018

The purpose of this study is to propose a power demand volatility evaluation system based on LSTM and not to verify the accuracy of the demand module which is a core module, but to recognize the sudden change of power pattern by using deeplearning in the actual power demand monitoring system. Then we confirm the availability of the module. Also, we tried to provide a visualized report so that the manager can determine the fluctuation of the power usage patten by applying it as a module to the web based system. It is confirmed that the power consumption data shows a certain pattern in the case of government offices and hospitals as a result of implementation of the volatility evaluation system. On the other hand, in areas with relatively low power consumption, such as residential facilities, it was not appropriate to evaluate the volatility.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.6
• /
• pp.1127-1137
• /
• 2022

Cryptographic analysis and decryption technology utilizing the parallel operation of GPU has been studied in the direction of shortening the computation time of the password analysis system. These studies focus on optimizing the code to improve the speed of cryptographic analysis operations on a single GPU or simply increasing the number of GPUs to enhance parallel operations. However, using a large number of GPUs without optimization for data transmission causes longer data transmission latency than using a single GPU and increases the overall computation time of the cryptographic analysis system. In this paper, we investigate GPUDirect RDMA and related technologies for high-performance data processing in deep learning or HPC research fields in GPU clustering environments. In addition, we present a method of designing a high-performance cryptanalysis system using the relevant technologies. Furthermore, based on the suggested system topology, we present a method of implementing a cryptanalysis system using password cracking and GPU reduction. Finally, the performance evaluation results are presented according to demonstration of high-performance technology is applied to the implemented cryptanalysis system, and the expected effects of the proposed system design are shown.

• Convergence Security Journal
• /
• v.21 no.1
• /
• pp.115-120
• /
• 2021

An O2O-based real estate brokerage web sites or apps are increasing explosively. As a result, the environment has been changed from the existing offline-based real estate brokerage environment to the online-based environment, and consumers are getting very good feelings in terms of time, cost, and convenience. However, behind the convenience of online-based real estate brokerage services, users often suffer time and money damage due to false information or malicious false information. Therefore, in this study, in order to reduce the damage to consumers that may occur in the O2O-based real estate brokerage service, we designed a false property information filtering system that can determine the authenticity of registered property information using artificial intelligence technology. Through the proposed research method, it was shown that not only the authenticity of the property information registered in the online real estate service can be determined, but also the temporal and financial damage of consumers can be reduced.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.6
• /
• pp.536-543
• /
• 2023

Speaker diarization, which labels for "who spoken when?" in speech with multiple speakers, has been studied on a deep neural network-based end-to-end method for labeling on speech overlap and optimization of speaker diarization models. Most deep neural network-based end-to-end speaker diarization systems perform multi-label classification problem that predicts the labels of all speakers spoken in each frame of speech. However, the performance of the multi-label-based model varies greatly depending on what the threshold is set to. In this paper, it is studied a speaker diarization system using single-label classification so that speaker diarization can be performed without thresholds. The proposed model estimate labels from the output of the model by converting speaker labels into a single label. To consider speaker label permutations in the training, the proposed model is used a combination of Permutation Invariant Training (PIT) loss and cross-entropy loss. In addition, how to add the residual connection structures to model is studied for effective learning of speaker diarization models with deep structures. The experiment used the Librispech database to generate and use simulated noise data for two speakers. When compared with the proposed method and baseline model using the Diarization Error Rate (DER) performance the proposed method can be labeling without threshold, and it has improved performance by about 20.7 %.

• Journal of Digital Convergence
• /
• v.19 no.2
• /
• pp.245-251
• /
• 2021

As the trend of steadily increasing the number of single or double household, there is a growing demand to see who is the outsider visiting the home during the free time. Various models of face recognition technology have been proposed through many studies, and Harr Cascade of OpenCV and Hog of Dlib are representative open source models. Among the two modes, Dlib's Hog has strengths in front of the indoor and at a limited distance, which is the focus of this study. In this paper, a face recognition visitor access system based on Dlib was designed and implemented. The whole system consists of a front module, a server module, and a mobile module, and in detail, it includes face registration, face recognition, real-time visitor verification and remote control, and video storage functions. The Precision, Specificity, and Accuracy according to the change of the distance threshold value were calculated using the error matrix with the photos published on the Internet, and compared with the results of previous studies. As a result of the experiment, it was confirmed that the implemented system was operating normally, and the result was confirmed to be similar to that reported by Dlib.

• Journal of Aerospace System Engineering
• /
• v.16 no.1
• /
• pp.49-55
• /
• 2022

The role of drones has been expanded to various fields such as agriculture, construction, and logistics. In particular, agriculture drones are emerging as an effective alternative to solve the problem of labor shortage and reduce the input cost. In this study therefore, we proposed the fruit recognition algorithm and harvesting mechanism for fruit harvesting drone system that can safely harvest fruits at high positions. In the fruit recognition algorithm, we employ "You-Only-Look-Once" which is a deep learning-based object detection algorithm and verify its feasibility by establishing a virtual simulation environment. In addition, we propose the fruit harvesting mechanism which can be operated by a single driving motor. The rotational motion of the motor is converted into a linear motion by the scotch yoke, and the opened gripper moves forward, grips a fruit and rotates it for harvesting. The feasibility of the proposed mechanism is verified by performing Multi-body dynamics analysis.

• Journal of KIISE
• /
• v.43 no.7
• /
• pp.795-800
• /
• 2016

In image classification, recent CNNs compete with human performance. However, there are limitations in more general recognition. Herein we deal with indoor images that contain too much information to be directly processed and require information reduction before recognition. To reduce the amount of data processing, typically variational inference or variational Bayesian methods are suggested for object detection. However, these methods suffer from the difficulty of marginalizing over the given space. In this study, we propose an image-text integrated recognition system using active vision based on Spatial Transformer Networks. The system attempts to efficiently sample a partial region of a given image for a given language information. Our experimental results demonstrate a significant improvement over traditional approaches. We also discuss the results of qualitative analysis of sampled images, model characteristics, and its limitations.

• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.143-174
• /
• 2023

With the development of deep learning technologies, Artificial Intelligence powered Optical Character Recognition (AI-OCR) has evolved to read multiple languages from various forms of images accurately. For the financial industry, where a large number of diverse documents are processed through manpower, the potential for using AI-OCR is great. In this study, we present a configuration and a design of an AI-OCR modality for use in the financial industry and discuss the platform construction with application cases. Since the use of financial domain data is prohibited under the Personal Information Protection Act, we developed a deep learning-based data generation approach and used it to train the AI-OCR models. The AI-OCR models are trained for image preprocessing, text recognition, and language processing and are configured as a microservice architected platform to process a broad variety of documents. We have demonstrated the AI-OCR platform by applying it to financial domain tasks of document sorting, document verification, and typing assistance The demonstrations confirm the increasing work efficiency and conveniences.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.5
• /
• pp.447-453
• /
• 2020

In this paper, we propose a system to extract effective speaker representations from a speech signal using a deep learning method. Based on the fact that speech signal contains identity unrelated information such as text content, emotion, background noise, and so on, we perform a training such that the extracted features only represent speaker-related information but do not represent speaker-unrelated information. Specifically, we propose an auto-encoder based disentanglement method that outputs both speaker-related and speaker-unrelated embeddings using effective loss functions. To further improve the reconstruction performance in the decoding process, we also introduce a discriminator popularly used in Generative Adversarial Network (GAN) structure. Since improving the decoding capability is helpful for preserving speaker information and disentanglement, it results in the improvement of speaker verification performance. Experimental results demonstrate the effectiveness of our proposed method by improving Equal Error Rate (EER) on benchmark dataset, Voxceleb1.