• Proceedings of the Korean Society for Language and Information Conference
• /
• 2007.11a
• /
• pp.303-310
• /
• 2007

Named Entity Recognition (NER) is always limited by its lower recall resulting from the asymmetric data distribution where the NONE class dominates the entity classes. This paper presents an approach that exploits non-local information to improve the NER recall. Several kinds of non-local features encoding entity token occurrence, entity boundary and entity class are explored under Conditional Random Fields (CRFs) framework. Experiments on SIGHAN 2006 MSRA (CityU) corpus indicate that non-local features can effectively enhance the recall of the state-of-the-art NER systems. Incorporating the non-local features into the NER systems using local features alone, our best system achieves a 23.56% (25.26%) relative error reduction on the recall and 17.10% (11.36%) relative error reduction on the F1 score; the improved F1 score 89.38% (90.09%) is significantly superior to the best NER system with F1 of 86.51% (89.03%) participated in the closed track.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.42 no.4
• /
• pp.892-898
• /
• 2017

This paper describes the development of AVN(Audio Video Navigation) software for vehicle information analysis and gesture recognition. The module that examine the CAN(Controller Area Network) data of vehicle in the designed software analyzes the driving state. Using classified information, the AVN software converge vehicle information and hand gesture information. As the result, the derived data is used to match the service step and to perform the service. The designed AVN software was implemented in HW platform that common used in vehicles. And we confirmed the operation of vehicle analysing module and gesture recognition in a simulated environment that is similar with real world.

• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.3
• /
• pp.223-228
• /
• 2011

In this paper, we proposed a technique to recognize three states in stance phase of gait cycle. Walking assistive devices are used to help the elderly people walk or to monitor walking behavior of the disabled persons. For the effective assistance, they adopt an intelligent sensor system to understand user's current state in walking. There are three states in stance phase; Loading Response, Midstance, and Terminal Stance. We developed a foot pressure sensor using 24 FSRs (Force Sensing/Sensitive Resistors). The foot pressure patterns were integrated through the interpolation of FSR cell array. The pressure patterns were processed to get the trajectories of COM (Center of Mass). Using the trajectories of COM of foot pressure, we can recognize the three states of stance phase. The experimental results show the effective recognition of stance phase and the possibility of usage on the walking assistive device for better control and/or foot pressure monitoring.

• Journal of Arbitration Studies
• /
• v.27 no.3
• /
• pp.53-76
• /
• 2017

Korean businesses, and their legal representatives, have observed the improvements of enforcement of commercial judgments through arbitration over traditional collections litigation in U.S. Courts-due to quicker proceedings, exceptional cost savings and more predictable outcomes-in attaching assets within U.S. jurisdictions. But how are the 2016 interim measures implemented by the Arbitration Act of Korea utilized to avoid jurisdictional and procedure pitfalls of enforcement proceedings in the Federal Courts of the United States? Authors examine the necessary prerequisites of the U.S. Federal Arbitration Act as adopted through the New York Convention, to which Korea and the U.S. are signatories, as distinguished from the Panama Convention. Five common U.S. arbitration institutions address U.S. "domestic" disputes, preempting U.S. state law arbitrations, while this article focuses on U.S. enforcement of "international" arbitration awards. Seeking U.S. recognition and enforcement of Korean arbitral awards necessitates avoiding common defenses involving due process, public policy or documentary formality challenges. Provisional and conservatory injunctive relief measures are explored. A variety of U.S. cases involving Korean litigants are examined to illustrate the legal challenges involving non?domestic arbitral awards, foreign arbitral awards and injunctive relief. Suggestions aimed toward further research are focused on typical Korean business needs such as motions to confirm foreign arbitration awards, enforce such awards or motions to compel arbitration.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2010.05a
• /
• pp.985-987
• /
• 2010

In this paper, we propose a maximum entropy-based emotion recognition model using individual average difference. In order to accurately recognize an user' s emotion, the proposed model utilizes the difference between the average of the given input physiological signals and the average of each emotion state' signals rather than only the input signal. For the purpose of alleviating data sparse -ness, the proposed model substitutes two simple symbols such as +(positive number)/-(negative number) for every average difference value, and calculates the average of physiological signals based on a second rather than the longer total emotion response time. With the aim of easily constructing the model, it utilizes a simple average difference calculation technique and a maximum entropy model, one of well-known machine learning techniques.

• Journal of the Korean Society of Industry Convergence
• /
• v.24 no.5
• /
• pp.547-553
• /
• 2021

Dynamic object recognition is an important task for autonomous vehicles. Since dynamic objects exhibit a higher collision risk than static objects, our own trajectories should be planned to match the future state of moving elements in the scene. Time information such as optical flow can be used to recognize movement. Existing optical flow calculations are based only on camera sensors and are prone to misunderstanding in low light conditions. In this regard, to improve recognition performance in low-light environments, we applied a normalization filter and a correction function for Gamma Value to the input images. The low light quality improvement algorithm can be applied to confirm the more accurate detection of Object's Bounding Box for the vehicle. It was confirmed that there is an important in object recognition through image prepocessing and deep learning using YOLO.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2019.05a
• /
• pp.439-441
• /
• 2019

Deep learning has become a powerful and robust algorithm in Artificial Intelligence. One of the most impressive forms of Deep learning tools is that of the Convolutional Neural Networks (CNN). CNN is a state-of-the-art solution for object recognition. For instance when we utilize CNN with MNIST handwritten digital dataset, mostly the result is well. Because, in MNIST dataset, all digits are centralized. Unfortunately, the real world is different from our imagination. If digits are shifted from the center, it becomes a big issue for CNN to recognize and provide result like before. To solve that issue, we have created frequency images from spatial images by a Fast Fourier Transform (FFT).

• ETRI Journal
• /
• v.44 no.2
• /
• pp.286-299
• /
• 2022

Human activity recognition in real time is a challenging task. Recently, a plethora of studies has been proposed using deep learning architectures. The implementation of these architectures requires the high computing power of the machine and a massive database. However, handcrafted features-based machine learning models need less computing power and very accurate where features are effectively extracted. In this study, we propose a handcrafted model based on three-dimensional sequential skeleton data. The human body skeleton movement over a frame is computed through joint positions in a frame. The joints of these skeletal frames are projected into two-dimensional space, forming a "movement polygon." These polygons are further transformed into a one-dimensional space by computing amplitudes at different angles from the centroid of polygons. The feature vector is formed by the sampling of these amplitudes at different angles. The performance of the algorithm is evaluated using a support vector machine on four public datasets: MSR Action3D, Berkeley MHAD, TST Fall Detection, and NTU-RGB+D, and the highest accuracies achieved on these datasets are 94.13%, 93.34%, 95.7%, and 86.8%, respectively. These accuracies are compared with similar state-of-the-art and show superior performance.

• Journal of the Korean Society for Precision Engineering
• /
• v.10 no.4
• /
• pp.142-151
• /
• 1993

An intelligent mulitiple monitoring system to monitor tool/machining states synthetically was proposed and developed. It consists of 2 fundamental subsystems : the multiple sensor detection unit and the intellignet integrated diagnosis unit. Three signals, that is, spindle motor current, Z-axis motor current, and machining sound were adopted to detect tool/machining states more reliably. Based on the multiple sensor information, the diagnosis unit judges either tool breakage or degree of tool wear state using fuzzy reasoning. Tool breakage is diagnosed by the level of spindle/z-axis motor current. Tool wear is diagnosed by both the result of fuzzy pattern recognition for motor currents and the result of pattern matching for machining sound. Fuzzy c-means algorithm was used for fuzzy pattern recognition. Experiments carried out for drill operation in the machining center have shown that the developed system monitors abnormal drill/states drilling very reliably.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.8
• /
• pp.2103-2118
• /
• 2024

Chinese herbal slices (CHS) automated recognition based on computer vision plays a critical role in the practical application of intelligent Chinese medicine. Due to the complexity and similarity of herbal images, identifying Chinese herbal slices is still a challenging task. Especially, easily-confused CHS have higher inter-class and intra-class complexity and similarity issues, the existing deep learning models are less adaptable to identify them efficiently. To comprehensively address these problems, a novel tiny easily-confused CHS dataset has been built firstly, which includes six pairs of twelve categories with about 2395 samples. Furthermore, we propose a ResNeSt-CHS model that combines multilevel perception fusion (MPF) and perceptive sparse fusion (PSF) blocks for efficiently recognizing easilyconfused CHS images. To verify the superiority of the ResNeSt-CHS and the effectiveness of our dataset, experiments have been employed, validating that the ResNeSt-CHS is optimal for easily-confused CHS recognition, with 2.1% improvement of the original ResNeSt model. Additionally, the results indicate that ResNeSt-CHS is applied on a relatively small-scale dataset yet high accuracy. This model has obtained state-of-the-art easily-confused CHS classification performance, with accuracy of 90.8%, far beyond other models (EfficientNet, Transformer, and ResNeSt, etc) in terms of evaluation criteria.