• Smart Structures and Systems
• /
• v.29 no.4
• /
• pp.625-640
• /
• 2022

Maglev rail joints are vital components serving as connections between the adjacent F-type rail sections in maglev guideway. Damage to maglev rail joints such as bolt looseness may result in rough suspension gap fluctuation, failure of suspension control, and even sudden clash between the electromagnets and F-type rail. The condition monitoring of maglev rail joints is therefore highly desirable to maintain safe operation of maglev. In this connection, an online damage detection approach based on three-dimensional (3D) convolutional neural network (CNN) and time-frequency characterization is developed for simultaneous detection of multiple damage of maglev rail joints in this paper. The training and testing data used for condition evaluation of maglev rail joints consist of two months of acceleration recordings, which were acquired in-situ from different rail joints by an integrated online monitoring system during a maglev train running on a test line. Short-time Fourier transform (STFT) method is applied to transform the raw monitoring data into time-frequency spectrograms (TFS). Three CNN architectures, i.e., small-sized CNN (S-CNN), middle-sized CNN (M-CNN), and large-sized CNN (L-CNN), are configured for trial calculation and the M-CNN model with excellent prediction accuracy and high computational efficiency is finally optioned for multiple damage detection of maglev rail joints. Results show that the rail joints in three different conditions (bolt-looseness-caused rail step, misalignment-caused lateral dislocation, and normal condition) are successfully identified by the proposed approach, even when using data collected from rail joints from which no data were used in the CNN training. The capability of the proposed method is further examined by using the data collected after the loosed bolts have been replaced. In addition, by comparison with the results of CNN using frequency spectrum and traditional neural network using TFS, the proposed TFS-CNN framework is proven more accurate and robust for multiple damage detection of maglev rail joints.

• Journal of Broadcast Engineering
• /
• v.25 no.2
• /
• pp.151-156
• /
• 2020

In this paper, we propose a watermark extraction method of omnidirectional images using CNN (Convolutional Neural Network) to improve the extracted watermark accuracy of the previous deterministic method that based on algorithm. This CNN consists of a restoration process of extracting watermarks by correcting distortion during omnidirectional image generation and/or malicious attacks, and a classification process of classifying which watermarks are extracted watermarks. Experiments with various attacks confirm that the extracted watermarks are more accurate than the previous methods.

• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1337-1342
• /
• 2019

The Automatic Korean license plate recognition (AKLPR) is used in many fields. For many applications, high recognition rate and fast processing speed of ALPR are important. Recent advances in deep learning have improved the accuracy and speed of object detection and recognition, and CNN (Convolutional Neural Network) has been applied to ALPR. The ALPR is divided into the stage of detecting the LP region and the stage of detecting and recognizing the character in the LP region, and each step is implemented with separate CNN. In this paper, we propose a single stage CNN architecture to recognize license plate characters at high speed while keeping high recognition rate.

• Journal of Multimedia Information System
• /
• v.8 no.3
• /
• pp.147-158
• /
• 2021

Versatile Video Coding (VVC) is the latest video coding standard developed by Joint Video Exploration Team (JVET). In VVC, the quadtree plus multi-type tree (QT+MTT) structure of coding unit (CU) partition is adopted, and its computational complexity is considerably high due to the brute-force search for recursive rate-distortion (RD) optimization. In this paper, we aim to reduce the time complexity of inter-picture prediction mode since the inter prediction accounts for a large portion of the total encoding time. The problem can be defined as classifying the split mode of each CU. To classify the split mode effectively, a novel convolutional neural network (CNN) called multi-level tree (MLT-CNN) architecture is introduced. For boosting classification performance, we utilize additional information including inter-picture information while training the CNN. The overall algorithm including the MLT-CNN inference process is implemented on VVC Test Model (VTM) 11.0. The CUs of size 128×128 can be the inputs of the CNN. The sequences are encoded at the random access (RA) configuration with five QP values {22, 27, 32, 37, 42}. The experimental results show that the proposed algorithm can reduce the computational complexity by 11.53% on average, and 26.14% for the maximum with an average 1.01% of the increase in Bjøntegaard delta bit rate (BDBR). Especially, the proposed method shows higher performance on the sequences of the A and B classes, reducing 9.81%~26.14% of encoding time with 0.95%~3.28% of the BDBR increase.

• Annual Conference on Human and Language Technology
• /
• 2017.10a
• /
• pp.330-332
• /
• 2017

개체명 인식(Named Entity Recognition, 이하 NER)은 인명(PS), 기관명(OG), 장소(LC), 날짜(DT), 시간(TI) 등에 해당하는 개체명에 일정한 태깅 값을 주어 그 정보를 가시화하는 작업이다. 한국어 개체명 인식은 아직 그 자질이 충분히 밝혀져 있지 않아 자연어 처리 분야의 발전을 더디게 하는 한 요소로 작용하고 있다. 한국어가 음절 기반으로 단어를 형성하고 비교적 어순이 자유롭다는 특성이 있기에, 이런 특징을 잘 포착할 수 있는 "음절 기반의 Convolutional Neural Network(CNN)"의 아키텍쳐를 제안하여 66.80%의 성능을 보였다. 이 방법을 사용하면 형태소 분석등 개체명 이전 단계에서 발생하는 오류에 의해 개체명 인식(NER)의 성능이 떨어지는 문제를 해결할 수 있고, 조사나 어미 등을 제거하기 위한 후처리를 생략할 수 있다.

• 한국어정보학회:학술대회논문집
• /
• 2017.10a
• /
• pp.330-332
• /
• 2017

개체명 인식(Named Entity Recognition, 이하 NER)은 인명(PS), 기관명(OG), 장소(LC), 날짜(DT), 시간(TI) 등에 해당하는 개체명에 일정한 태깅 값을 주어 그 정보를 가시화하는 작업이다. 한국어 개체명 인식은 아직 그 자질이 충분히 밝혀져 있지 않아 자연어 처리 분야의 발전을 더디게 하는 한 요소로 작용하고 있다. 한국어가 음절 기반으로 단어를 형성하고 비교적 어순이 자유롭다는 특성이 있기에, 이런 특징을 잘 포착할 수 있는 "음절 기반의 Convolutional Neural Network(CNN)"의 아키텍쳐를 제안하여 66.80%의 성능을 보였다. 이 방법을 사용하면 형태소 분석등 개체명 이전 단계에서 발생하는 오류에 의해 개체명 인식(NER)의 성능이 떨어지는 문제를 해결할 수 있고, 조사나 어미 등을 제거하기 위한 후처리를 생략할 수 있다.

• Journal of Information Processing Systems
• /
• v.16 no.5
• /
• pp.1105-1112
• /
• 2020

Gait recognition, as a promising biometric, can be used in video-based surveillance and other security systems. However, due to the complexity of leg movement and the difference of external sampling conditions, gait recognition still faces many problems to be addressed. In this paper, an improved convolutional neural network (CNN) based on Gabor filter is therefore proposed to achieve gait recognition. Firstly, a gait feature extraction layer based on Gabor filter is inserted into the traditional CNNs, which is used to extract gait features from gait silhouette images. Then, in the process of gait classification, using the output of CNN as input, we utilize metric learning techniques to calculate distance between two gaits and achieve gait classification by k-nearest neighbors classifiers. Finally, several experiments are conducted on two open-accessed gait datasets and demonstrate that our method reaches state-of-the-art performances in terms of correct recognition rate on the OULP and CASIA-B datasets.

• Journal of Korea Multimedia Society
• /
• v.22 no.6
• /
• pp.665-675
• /
• 2019

In this paper, we compare and analyze the classification performance of deep learning algorithm Convolutional Neural Network(CNN) ac cording to ensemble generation and combining techniques. We used several CNN models(VGG16, VGG19, DenseNet121, DenseNet169, DenseNet201, ResNet18, ResNet34, ResNet50, ResNet101, ResNet152, GoogLeNet) to create 10 ensemble generation combinations and applied 6 combine techniques(average, weighted average, maximum, minimum, median, product) to the optimal combination. Experimental results, DenseNet169-VGG16-GoogLeNet combination in ensemble generation, and the product rule in ensemble combination showed the best performance. Based on this, it was concluded that ensemble in different models of high benchmarking scores is another way to get good results.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.11a
• /
• pp.112-115
• /
• 2022

The process of deep learning usually needs to deal with massive data which has greatly limited the development of deep learning technologies today. Convolutional Neural Network (CNN) structure is often used to solve image classification problems. However, a large number of images may be required in order to train an image in CNN, which is a heavy burden for existing computer systems to handle. If the image data can be compressed under the premise that the computer hardware system remains unchanged, it is possible to train more datasets in deep learning. However, image compression usually adopts the form of lossy compression, which will lose part of the image information. If the lost information is key information, it may affect learning performance. In this paper, we will analyze the effect of image compression on deep learning performance on CNN-based cat and dog classification. Through the experiment results, we conclude that the compression of images does not have a significant impact on the accuracy of deep learning.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.557-558
• /
• 2021

In this paper, we propose a new image classification method based on Convolutional Neural Network (CNN), which is mainly used to solve model overfitting and non-convergence and to improve classification accuracy in image classification tasks on small datasets.