• 한국전기전자재료학회논문지
• /
• 제33권3호
• /
• pp.239-244
• /
• 2020

This study examines the feasibility of the image deep learning method using convolution neural networks (CNNs) to maintain a porcelain insulator. Data augmentation is performed to prevent over-fitting, and the classification performance is evaluated by training the age, material, region, and pollution level of the insulator using image data in which the background and labelling are removed. Based on the results, it was difficult to predict the age, but it was possible to classify 76% of the materials, 60% of the pollution level, and more than 90% of the regions. From the results of this study, we identified the potential and limitations of the CNN classification for the four groups currently classified. However, it was possible to detect discoloration of the porcelain insulator resulting from physical, chemical, and climatic factors. Based on this, it will be possible to estimate the corrosion of the cap and discoloration of the porcelain caused by environmental deterioration, abnormal voltage, and lightning.

• 한국음향학회지
• /
• 제40권5호
• /
• pp.488-495
• /
• 2021

본 논문에서는 콘포머 기반 한국어 음성인식 시스템을 제안한다. 콘포머는 트랜스포머 모델에 콘볼루션신경망(Convolution Neural Network, CNN) 기능을 보강한 구조이며 광역 정보를 잘 표현할 수 있는 트랜스포머와 지역 정보를 잘 표현할 수 있는 CNN을 결합한 신경망이다. 음성인식 기본 시스템으로 트랜스포모에 기반한 음성인식시스템을 개발하였으며 언어모델로는 Long Short-Term Memory(LSTM)을 사용하였다. 콘포머 기반 음성인식시스템은 트랜스포머 대신에 콘포머를 사용하였고 언어모델로는 트랜스포머를 이용하였다. 성능 평가를 위해 AI-hub에 있는 Electronics and Telecommunications Research Institute(ETRI) 음성코퍼스를 활용하였으며 트랜스포머 기반 음성인식 시스템은 오인식률이 11.8 %이 되었으며 콘포머 기반 음성인식시스템은 오인식률이 5.7 %가 되었다. AI-hub에 있는 다른 영역의 NHN다이퀘스트 음성 코퍼스를 추가해도 유사한 성능이 유지가 되어 제안된 콘포머 음성인식시스템의 유효성을 입증하였다.

• 전기학회논문지
• /
• 제67권7호
• /
• pp.865-870
• /
• 2018

A machine vision based industrial inspection includes defects detection and classification. Fast inspection is a fundamental problem for many applications of real-time vision systems. It requires little computation time and localizing defects robustly with high accuracy. Deep learning technique have been known not to be suitable for real-time applications. Recently a couple of fast region-based CNN algorithms for object detection are introduced, such as Faster R-CNN, and YOLOv2. We apply these methods for an industrial inspection problem. Three CNN based detection algorithms, VOV based CNN, Faster R-CNN, and YOLOv2, are experimented for defect detection on metal surface. The results for inspection time and various performance indices are compared and analysed.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권12호
• /
• pp.3868-3888
• /
• 2022

A widely used social networking service like Twitter has the ability to disseminate information to large groups of people even during a pandemic. At the same time, it is a convenient medium to share irrelevant and unverified information online and poses a potential threat to society. In this research, conventional machine learning algorithms are analyzed to classify the data as either non-rumor data or rumor data. Machine learning techniques have limited tuning capability and make decisions based on their learning. To tackle this problem the authors propose a deep learning-based Rumor Detection Neural Network model to predict the rumor tweet in real-world events. This model comprises three layers, AttCNN layer is used to extract local and position invariant features from the data, AttBi-LSTM layer to extract important semantic or contextual information and HPOOL to combine the down sampling patches of the input feature maps from the average and maximum pooling layers. A dataset from Kaggle and ground dataset #gaja are used to train the proposed Rumor Detection Neural Network to determine the veracity of the rumor. The experimental results of the RDNN Classifier demonstrate an accuracy of 93.24% and 95.41% in identifying rumor tweets in real-time events.

• 디지털산업정보학회논문지
• /
• 제18권4호
• /
• pp.1-11
• /
• 2022

Recently, research on license plate recognition, which is a core technology of an intelligent transportation system(ITS), is being actively conducted. In this paper, we propose a method to extract numbers and characters from low-quality license plate images by applying the YOLOv4 algorithm. YOLOv4 is a one-stage object detection method using convolution neural network including BACKBONE, NECK, and HEAD parts. It is a method of detecting objects in real time rather than the previous two-stage object detection method such as the faster R-CNN. In this paper, we studied a method to directly extract number and character regions from low-quality license plate images without additional edge detection and image segmentation processes. In order to evaluate the performance of the proposed method we experimented with 500 license plate images. In this experiment, 350 images were used for training and the remaining 150 images were used for the testing process. Computer simulations show that the mean average precision of detecting number and character regions on vehicle license plates was about 93.8%.

• Journal of Information Processing Systems
• /
• 제17권2호
• /
• pp.385-398
• /
• 2021

The complexity of deep learning models affects the real-time performance of gesture recognition, thereby limiting the application of gesture recognition algorithms in actual scenarios. Hence, a residual learning neural network based on a deep convolutional neural network is proposed. First, small convolution kernels are used to extract the local details of gesture images. Subsequently, a shallow residual structure is built to share weights, thereby avoiding gradient disappearance or gradient explosion as the network layer deepens; consequently, the difficulty of model optimisation is simplified. Additional convolutional neural networks are used to accelerate the refinement of deep abstract features based on the spatial importance of the gesture feature distribution. Finally, a fully connected cascade softmax classifier is used to complete the gesture recognition. Compared with the dense connection multiplexing feature information network, the proposed algorithm is optimised in feature multiplexing to avoid performance fluctuations caused by feature redundancy. Experimental results from the ISOGD gesture dataset and Gesture dataset prove that the proposed algorithm affords a fast convergence speed and high accuracy.

• International Journal of Computer Science & Network Security
• /
• 제21권11호
• /
• pp.17-22
• /
• 2021

Recently, image super-resolution techniques used in convolutional neural networks (CNN) have led to remarkable performance in the research area of digital image processing applications and computer vision tasks. Convolutional layers stacked on top of each other can design a more complex network architecture, but they also use more memory in terms of the number of parameters and introduce the vanishing gradient problem during training. Furthermore, earlier approaches of single image super-resolution used interpolation technique as a pre-processing stage to upscale the low-resolution image into HR image. The design of these approaches is simple, but not effective and insert the newer unwanted pixels (noises) in the reconstructed HR image. In this paper, authors are propose a novel single image super-resolution architecture based on synchronized depthwise separable convolution with Dense Skip Connection Block (DSCB). In addition, unlike existing SR methods that only rely on single path, but our proposed method used the synchronizes path for generating the SISR image. Extensive quantitative and qualitative experiments show that our method (SDCN) achieves promising improvements than other state-of-the-art methods.

• 한국해안·해양공학회논문집
• /
• 제33권3호
• /
• pp.101-109
• /
• 2021

항해용 X-band 레이다를 이용한 파랑관측은 해수면에 후방산란 된 전자기파 이미지를 분석하여 이루어진다. 1분당 42개의 해수면 시계열 이미지로부터 3차원 FFT를 계산하고 변조전달함수(Modulation Transfer Function)를 구하여 파랑정보를 추출한다. 따라서 레이다 파고계로 계측한 유의파고의 정확도는 X-band 레이다 영상의 상태에 따라 결정된다. 2020년 여름 태풍 마이삭과 하이선 내습 시 강릉 안인 해안에 설치된 X-band 레이다 파고계로 관측한 유의파고의 오차가 크게 발생하였다. 이는 태풍 내습 시 급격히 유의파고가 증가하는 한편 강한 강우가 동반되어 X-band 레이다 영상의 품질이 저하되었기 때문이다. 최대 오차 발생 이전까지 많은 강우가 있었음이 확인된다. 본 연구에서는 convolution neural network(CNN)을 이용하여 레이다 이미지로부터 강우 여부를 분류하고 강우여부에 따라 강우시 인공신경망 모델을 적용하여 태풍 시 유의파고 관측 정확도를 향상시켰다. 폭우를 동반한 태풍 시 레이다 자료 특성에 기반하여 인공신경망 유의파고 산출 알고리즘을 개선하고 이를 통해 X-band 레이다 파고계의 정확도를 향상시키는 방법을 제시하였다.

• 방송공학회논문지
• /
• 제23권3호
• /
• pp.345-350
• /
• 2018

CNN (convolutional neural network)은 최근 가장 주목받는 인공지능 기법 중 하나이며 특히 영상 분류에서 기존의 기법에 비해 월등한 성능을 보인다. 본 논문에서는 CNN을 이용하여 다양한 거리 사진을 분류하고, 분류 결과를 이용하여 해당 거리에 대한 안전도의 평가 방법을 제안한다. 제안하는 기법은 CNN을 이용하여 총 네 가지 유형의 거리 사진에 대하여 학습을 수행하는 과정과 학습된 네트워크 모델을 바탕으로 해당 거리 사진의 분류와 안전도를 평가하는 과정을 포함한다. 거리 사진의 학습 과정에서는 네 가지 유형의 거리 사진 데이터셋을 수집하고 이 데이터를 증강시킨 후 CNN 학습을 수행한다. 학습된 CNN 모델은 주어진 입력 영상의 분류를 정확히 수행하고, 거리의 안전도는 각 유형에 대한 확률을 조합하여 정량적으로 계산한다.

• 농업과학연구
• /
• 제47권4호
• /
• pp.1109-1122
• /
• 2020

The aim of this study was to develop a marbling classification and prediction model using small parts of sirloin images based on a deep learning algorithm, namely, a convolutional neural network (CNN). Samples were purchased from a commercial slaughterhouse in Korea, images for each grade were acquired, and the total images (n = 500) were assigned according to their grade number: 1++, 1+, 1, and both 2 & 3. The image acquisition system consists of a DSLR camera with a polarization filter to remove diffusive reflectance and two light sources (55 W). To correct the distorted original images, a radial correction algorithm was implemented. Color images of sirloins of Hanwoo (mixed with feeder cattle, steer, and calf) were divided and sub-images with image sizes of 161 × 161 were made to train the marbling prediction model. In this study, the convolutional neural network (CNN) has four convolution layers and yields prediction results in accordance with marbling grades (1++, 1+, 1, and 2&3). Every single layer uses a rectified linear unit (ReLU) function as an activation function and max-pooling is used for extracting the edge between fat and muscle and reducing the variance of the data. Prediction accuracy was measured using an accuracy and kappa coefficient from a confusion matrix. We summed the prediction of sub-images and determined the total average prediction accuracy. Training accuracy was 100% and the test accuracy was 86%, indicating comparably good performance using the CNN. This study provides classification potential for predicting the marbling grade using color images and a convolutional neural network algorithm.