• Journal of Biosystems Engineering
• /
• v.27 no.2
• /
• pp.151-160
• /
• 2002

Cow image processing technique would be useful not only for recognizing an individual but also for establishing the image database and analyzing the shape of cows. A cow (Holstein) has usually the unique speckle pattern. In this study, the individual recognition of cow was carried out using the speckle pattern and the content-based image retrieval technique. Sixty cow images of 16 heads were captured under outdoor illumination, which were complicated images due to shadow, obstacles and walking posture of cow. Sixteen images were selected as the reference image for each cow and 44 query images were used for evaluating the efficiency of individual recognition by matching to each reference image. Run-lengths and positions of runs across speckle area were calculated from 40 horizontal line profiles for ROI (region of interest) in a cow body image after 3 passes of 5$\times$5 median filtering. A similarity measure for recognizing cow individuals was calculated using Euclidean distance of normalized G-frame histogram (GH). normalized speckle run-length (BRL), normalized x and y positions (BRX, BRY) of speckle runs. This study evaluated the efficiency of individual recognition of cow using Recall(Success rate) and AVRR(Average rank of relevant images). Success rate of individual recognition was 100% when GH, BRL, BRX and BRY were used as image query indices. It was concluded that the histogram as global property and the information of speckle runs as local properties were good image features for individual recognition and the developed system of individual recognition was reliable.

• Journal of Electrical Engineering and Technology
• /
• v.9 no.2
• /
• pp.739-748
• /
• 2014

In this paper, we propose pyramid appearance and global structure action descriptors on both RGB and depth motion history images and a model-free method for human action recognition. In proposed algorithm, we firstly construct motion history image for both RGB and depth channels, at the same time, depth information is employed to filter RGB information, after that, different action descriptors are extracted from depth and RGB MHIs to represent these actions, and then multimodality information collaborative representation and recognition model, in which multi-modality information are put into object function naturally, and information fusion and action recognition also be done together, is proposed to classify human actions. To demonstrate the superiority of the proposed method, we evaluate it on MSR Action3D and DHA datasets, the well-known dataset for human action recognition. Large scale experiment shows our descriptors are robust, stable and efficient, when comparing with the-state-of-the-art algorithms, the performances of our descriptors are better than that of them, further, the performance of combined descriptors is much better than just using sole descriptor. What is more, our proposed model outperforms the state-of-the-art methods on both MSR Action3D and DHA datasets.

• Journal of Korea Water Resources Association
• /
• v.39 no.3 s.164
• /
• pp.215-226
• /
• 2006

In this study, we performed verification of VSRF (Very Short Range Forecast of precipitation) model and application of NWSPC (National Weather Service PC) rainfall-runoff model in Kyoungan-chun basin. We used two methods for verification of VSRF model. The first method is a meteorological verification that evaluates the special quality feature for rain amount between AWS and VSRF model over Kyoungan-chun basin, while second method is a hydrological verification that compares the calculated Mean Area Precipitation (MAP) between AWS and VSRF Quantitatively. This study examines the usefulness of VSRF precipitation forecasting model data in NWSPC hydrological model. As a result, correlation coefficient is over 0.6 within 3 hour lead time. It represents that the forecast results from VSRF are useful for water resources application.

• Journal of Intelligence and Information Systems
• /
• v.13 no.4
• /
• pp.79-90
• /
• 2007

The image quality measure is an important issue in the image processing. Several methods which measure the image quality have been proposed and these are based on the mathematical point of view. However, there is difference between the mathematicalmeasure and the measure based on the human visual system and a new measure has to be proposed because the final target of the image is a human visual system In this paper, a statistical image quality measure which is considered the human visual feature was suggested. The human visual system is using the global quality of the image and the local quality of the image and the local quality is more important to human visual system. In this paper, the image divided into several segments and the image qualities were calculated respectively. After then, the statistical method using scoring was applied to the image qualities. The result of the image quality measure was similar to the result of measure based on the human visual system.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.5
• /
• pp.488-495
• /
• 2021

We propose a speech recognition system based on conformer. Conformer is known to be convolution-augmented transformer, which combines transfer model for capturing global information with Convolution Neural Network (CNN) for exploiting local feature effectively. The baseline system is developed to be a transfer-based speech recognition using Long Short-Term Memory (LSTM)-based language model. The proposed system is a system which uses conformer instead of transformer with transformer-based language model. When Electronics and Telecommunications Research Institute (ETRI) speech corpus in AI-Hub is used for our evaluation, the proposed system yields 5.7 % of Character Error Rate (CER) while the baseline system results in 11.8 % of CER. Even though speech corpus is extended into other domain of AI-hub such as NHNdiguest speech corpus, the proposed system makes a robust performance for two domains. Throughout those experiments, we can prove a validation of the proposed system.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.1
• /
• pp.106-112
• /
• 2019

The Korea Copyright Commission has pursued copyright technology R&D projects to prevent illegal copying of comics and Webtoons. We developed a feature-based scanned comic filtering technology in order to apply technical measures to specific types of online service providers. We also developed technologies in order to monitor and identify illegally distributed comics on webhard sites and to monitor and identify illegally distributed webtoons. Even though all comic books posted on webhard sites are illegal, it is no trouble to download and access popular comics by accessing websites in foreign countries. Even under these circumstances, the comic and webtoon copyright protection technologies developed over the past six years have been used at all. In this paper, we examine what the problems are and find solutions to propose a copyright protection method for webtoons.

• Smart Structures and Systems
• /
• v.30 no.6
• /
• pp.613-626
• /
• 2022

Guaranteeing the quality and integrity of structural health monitoring (SHM) data is very important for an effective assessment of structural condition. However, sensory system may malfunction due to sensor fault or harsh operational environment, resulting in multiple types of data anomaly existing in the measured data. Efficiently and automatically identifying anomalies from the vast amounts of measured data is significant for assessing the structural conditions and early warning for structural failure in SHM. The major challenges of current automated data anomaly detection methods are the imbalance of dataset categories. In terms of the feature of actual anomalous data, this paper proposes a data anomaly detection method based on data-level and deep learning technique for SHM of civil engineering structures. The proposed method consists of a data balancing phase to prepare a comprehensive training dataset based on data-level technique, and an anomaly detection phase based on a sophisticatedly designed network. The advanced densely connected convolutional network (DenseNet) and Transformer encoder are embedded in the specific network to facilitate extraction of both detail and global features of response data, and to establish the mapping between the highest level of abstractive features and data anomaly class. Numerical studies on a steel frame model are conducted to evaluate the performance and noise immunity of using the proposed network for data anomaly detection. The applicability of the proposed method for data anomaly classification is validated with the measured data of a practical supertall structure. The proposed method presents a remarkable performance on data anomaly detection, which reaches a 95.7% overall accuracy with practical engineering structural monitoring data, which demonstrates the effectiveness of data balancing and the robust classification capability of the proposed network.

• International Journal of Computer Science & Network Security
• /
• v.22 no.5
• /
• pp.175-181
• /
• 2022

COVID-19 is an acute respiratory syndrome that affects the host's breathing and respiratory system. The novel disease's first case was reported in 2019 and has created a state of emergency in the whole world and declared a global pandemic within months after the first case. The disease created elements of socioeconomic crisis globally. The emergency has made it imperative for professionals to take the necessary measures to make early diagnoses of the disease. The conventional diagnosis for COVID-19 is through Polymerase Chain Reaction (PCR) testing. However, in a lot of rural societies, these tests are not available or take a lot of time to provide results. Hence, we propose a COVID-19 classification system by means of machine learning and transfer learning models. The proposed approach identifies individuals with COVID-19 and distinguishes them from those who are healthy with the help of Deep Visual Embeddings (DVE). Five state-of-the-art models: VGG-19, ResNet50, Inceptionv3, MobileNetv3, and EfficientNetB7, were used in this study along with five different pooling schemes to perform deep feature extraction. In addition, the features are normalized using standard scaling, and 4-fold cross-validation is used to validate the performance over multiple versions of the validation data. The best results of 88.86% UAR, 88.27% Specificity, 89.44% Sensitivity, 88.62% Accuracy, 89.06% Precision, and 87.52% F1-score were obtained using ResNet-50 with Average Pooling and Logistic regression with class weight as the classifier.

• Journal of Advanced Navigation Technology
• /
• v.27 no.2
• /
• pp.182-189
• /
• 2023

Depth estimation is a key technology in 3D map generation for autonomous driving of vehicles, robots, and drones. The existing sensor-based method has high accuracy but is expensive and has low resolution, while the camera-based method is more affordable with higher resolution. In this study, we propose self-attention-based unsupervised monocular depth estimation for UAV camera system. Self-Attention operation is applied to the network to improve the global feature extraction performance. In addition, we reduce the weight size of the self-attention operation for a low computational amount. The estimated depth and camera pose are transformed into point cloud. The point cloud is mapped into 3D map using the occupancy grid of Octree structure. The proposed network is evaluated using synthesized images and depth sequences from the Mid-Air dataset. Our network demonstrates a 7.69% reduction in error compared to prior studies.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.4
• /
• pp.357-363
• /
• 2023

In underwater signal processing, separating individual signals from mixed signals has long been a challenge due to low signal quality. The common method using Short-time Fourier transform for spectrogram analysis has faced criticism for its complex parameter optimization and loss of phase data. We propose a Triple-path Recurrent Neural Network, based on the Dual-path Recurrent Neural Network's success in long time series signal processing, to handle three-dimensional tensors from multi-channel sensor input signals. By dividing input signals into short chunks and creating a 3D tensor, the method accounts for relationships within and between chunks and channels, enabling local and global feature learning. The proposed technique demonstrates improved Root Mean Square Error and Scale Invariant Signal to Noise Ratio compared to the existing method.