• ETRI Journal
• /
• v.40 no.4
• /
• pp.499-510
• /
• 2018

A vector space based augmented structural kinematic (VSASK) feature descriptor is proposed for human activity recognition. An action descriptor is built by integrating the structural and kinematic properties of the actor using vector space based augmented matrix representation. Using the local or global information separately may not provide sufficient action characteristics. The proposed action descriptor combines both the local (pose) and global (position and velocity) features using augmented matrix schema and thereby increases the robustness of the descriptor. A multiclass support vector machine (SVM) is used to learn each action descriptor for the corresponding activity classification and understanding. The performance of the proposed descriptor is experimentally analyzed using the Weizmann and KTH datasets. The average recognition rate for the Weizmann and KTH datasets is 100% and 99.89%, respectively. The computational time for the proposed descriptor learning is 0.003 seconds, which is an improvement of approximately 1.4% over the existing methods.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2016.06a
• /
• pp.253-255
• /
• 2016

본 논문에서는 다중 지역 이진 패턴(Multi-scale Bock LBP, MB-LBP) 특징과 랜덤 포레스트에 기반한 새로운 기법의 머리 방향 분류 기법을 제안한다. 제안 기법에서는 occlusion 과 조명의 변화에 강인한 분류 정확도를 얻기 위해서 랜덤화된 트리를 학습하는 것을 목표로 한다. 우선, 얼굴 이미지로부터 많은 MB-LBP 특징을 추출하고, 얼굴 영상들을 랜덤하게 입력하고 MB-LBP 크기 파라미터와 같은 랜덤 특징과 블록 좌표들을 사용하여 트리를 생성한다. 게다가 각 노드에서 정보 이득을 최대화 하는 트리의 내부 노드를 생성하기 위해서 uniform LBP 의 특성을 고려한 분할 함수를 개발한다. 랜덤화된 트리는 랜덤 포레스트에 포함되어 있으며 마지막 결정단계에서 Maximum-A-Posteriori criterion 으로 최종 결정을 한다. 실험 결과는 제안 기법이 다양한 조명, 자세, 표현, occlusion 상황에서 기존의 방법보다 개선된 성능으로 머리 방향을 분류 할 수 있음을 보여준다.

• Proceedings of the KIEE Conference
• /
• 2002.11c
• /
• pp.370-373
• /
• 2002

This paper proposes a new measurement system determine the localization and the type of object which use only three ultrasonic sensors, one the transmitter, one the receiver and one transduce doing both transmitter and receiver. this system can classifies the type and determines the pose of the target object. it used the method of Pseudoamplitude Scan. So it significantly simple the sensing system and reduce the signal processing time so that the working environment can be recognized in real time.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.3
• /
• pp.971-976
• /
• 2000

In this paper a new method for head gesture recognition is proposed. A the first stage, face image data are transformed into low dimensional vectors by principal component analysis (PCA), which utilizes the high correlation between face pose images. The a self organization map(SM) is trained by the transformed face vectors, in such a that the nodes at similar locations respond to similar poses. A sequence of poses which comprises each model gesture goes through PCA and SOM, and the result is stored in the database. At the recognition stage any sequence of frames goes through the PCA and SOM, and the result is compared with the model gesture stored in the database. To improve robustness of classification, probabilistic relaxation labeling(PRL) is used, which utilizes the contextural information imbedded in the adjacent poses.

• Journal of the Institute of Convergence Signal Processing
• /
• v.22 no.1
• /
• pp.30-37
• /
• 2021

Accurate 3D object tracking with camera images is a key enabling technology for augmented reality applications. Motivated by the impressive success of convolutional neural networks (CNNs) in computer vision tasks such as image classification, object detection, image segmentation, recent studies for 3D object tracking have focused on leveraging deep learning. In this paper, we review deep learning approaches for 3D object tracking. We describe key methods in this field and discuss potential future research directions.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.5
• /
• pp.177-183
• /
• 2022

Recently, interesting of home training is getting bigger due to COVID-19. Accordingly, research on applying HAR(human activity recognition) technology to home training has been conducted. However, existing paper of HAR proposed static activity instead of dynamic activity. In this paper, the deep learning model where dynamic exercise posture can be analyzed and the accuracy of the user's exercise posture can be shown is proposed. Fitness images of AI-hub are analyzed by blaze pose. The experiment is compared with three types of deep learning model: RNN(recurrent neural network), LSTM(long short-term memory), CNN(convolution neural network). In simulation results, it was shown that the f1-score of RNN, LSTM and CNN is 0.49, 0.87 and 0.98, respectively. It was confirmed that CNN is more suitable for human activity recognition than other models from simulation results. More exercise postures can be analyzed using a variety learning data.

• Safety and Health at Work
• /
• v.14 no.3
• /
• pp.272-278
• /
• 2023

Background: Some working conditions may pose a higher physical or psychological demand to pregnant women leading to increased risks of pregnancy complications. Objectives: We assessed the association of woman's employment status and the industrial classification with obstetric complications. Methods: We conducted a national population study using the National Health Information Service database of Republic of Korea. Our analysis encompassed 1,316,310 women who experienced first-order live births in 2010-2019. We collected data on the employment status and the industrial classification of women, as well as their diagnoses of preeclampsia (PE) and gestational diabetes mellitus (GDM) classified as A1 (well controlled by diet) or A2 (requiring medication). We calculated odds ratios (aORs) of complications per employment, and each industrial classification was adjusted for individual risk factors. Results: Most (64.7%) were in employment during pregnancy. Manufacturing (16.4%) and the health and social (16.2%) work represented the most prevalent industries. The health and social work exhibited a higher risk of PE (aOR = 1.11, 95% confidence interval [CI]: 1.03-1.21), while the manufacturing industry demonstrated a higher risk of class A2 GDM (1.20, 95% CI: 1.03-1.41) than financial intermediation. When analyzing both classes of GDM, women who worked in public administration and defense/social security showed higher risk of class A1 GDM (1.04, 95% CI: 1.01, 1.07). When comparing high-risk industries with nonemployment, the health and social work showed a comparable risk of PE (1.02, 95% CI: 0.97, 1.07). Conclusion: Employment was associated with overall lower risks of obstetric complications. Health and social service work can counteract the healthy worker effect in relation to PE. This highlights the importance of further elucidating specific occupational risk factors within the high-risk industries.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.10
• /
• pp.67-76
• /
• 2023

Although deep learning models are making innovative achievements in the field of computer vision, the problem of vulnerability to adversarial examples continues to be raised. Adversarial examples are attack methods that inject fine noise into images to induce misclassification, which can pose a serious threat to the application of deep learning models in the real world. In this paper, we propose a model that detects adversarial examples using differences in predictive values between edge-learned classification models and underlying classification models. The simple process of extracting the edges of the objects and reflecting them in learning can increase the robustness of the classification model, and economical and efficient detection is possible by detecting adversarial examples through differences in predictions between models. In our experiments, the general model showed accuracy of {49.9%, 29.84%, 18.46%, 4.95%, 3.36%} for adversarial examples (eps={0.02, 0.05, 0.1, 0.2, 0.3}), whereas the Canny edge model showed accuracy of {82.58%, 65.96%, 46.71%, 24.94%, 13.41%} and other edge models showed a similar level of accuracy also, indicating that the edge model was more robust against adversarial examples. In addition, adversarial example detection using differences in predictions between models revealed detection rates of {85.47%, 84.64%, 91.44%, 95.47%, and 87.61%} for each epsilon-specific adversarial example. It is expected that this study will contribute to improving the reliability of deep learning models in related research and application industries such as medical, autonomous driving, security, and national defense.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.2
• /
• pp.31-41
• /
• 2024

Deep learning models show excellent performance in tasks such as image classification and object detection in the field of computer vision, and are used in various ways in actual industrial sites. Recently, research on improving robustness has been actively conducted, along with pointing out that this deep learning model is vulnerable to hostile examples. A hostile example is an image in which small noise is added to induce misclassification, and can pose a significant threat when applying a deep learning model to a real environment. In this paper, we tried to confirm the robustness of the edge-learning classification model and the performance of the adversarial example detection model using it for adversarial examples of various algorithms. As a result of robustness experiments, the basic classification model showed about 17% accuracy for the FGSM algorithm, while the edge-learning models maintained accuracy in the 60-70% range, and the basic classification model showed accuracy in the 0-1% range for the PGD/DeepFool/CW algorithm, while the edge-learning models maintained accuracy in 80-90%. As a result of the adversarial example detection experiment, a high detection rate of 91-95% was confirmed for all algorithms of FGSM/PGD/DeepFool/CW. By presenting the possibility of defending against various hostile algorithms through this study, it is expected to improve the safety and reliability of deep learning models in various industries using computer vision.

• Management & Information Systems Review
• /
• v.39 no.3
• /
• pp.1-16
• /
• 2020

The current tariff return system requires tax officials to calculate tax amount by themselves and pay the tax amount on their own responsibility. In other words, in principle, the duty and responsibility of reporting payment system are imposed only on the taxee who is required to calculate and pay the tax accurately. In case the tax payment system fails to fulfill the duty and responsibility, the additional tax is imposed on the taxee by collecting the tax shortfall and imposing the tax deduction on For this reason, item classifications, together with tariff assessments, are the most difficult and could pose a significant risk to entities if they are misclassified. For this reason, import reports are consigned to customs officials, who are customs experts, while paying a substantial fee. The purpose of this study is to classify HS items to be reported upon import declaration and to indicate HS codes to be recorded on import declaration. HS items were classified using the attached image in the case of item classification based on the case of the classification of items by the Korea Customs Service for classification of HS items. For image classification, CNN was used as a deep learning algorithm commonly used for image recognition and Vgg16, Vgg19, ResNet50 and Inception-V3 models were used among CNN models. To improve classification accuracy, two datasets were created. Dataset1 selected five types with the most HS code images, and Dataset2 was tested by dividing them into five types with 87 Chapter, the most among HS code 2 units. The classification accuracy was highest when HS item classification was performed by learning with dual database2, the corresponding model was Inception-V3, and the ResNet50 had the lowest classification accuracy. The study identified the possibility of HS item classification based on the first item image registered in the item classification determination case, and the second point of this study is that HS item classification, which has not been attempted before, was attempted through the CNN model.