• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.6
• /
• pp.1833-1848
• /
• 2022

Named entity recognition (NER) is an important basic task in the field of Natural Language Processing (NLP). Recently deep learning approaches by extracting word segmentation or character features have been proved to be effective for Chinese Named Entity Recognition (CNER). However, since this method of extracting features only focuses on extracting some of the features, it lacks textual information mining from multiple perspectives and dimensions, resulting in the model not being able to fully capture semantic features. To tackle this problem, we propose a novel Multi-view Semantic Feature Fusion Model (MSFM). The proposed model mainly consists of two core components, that is, Multi-view Semantic Feature Fusion Embedding Module (MFEM) and Multi-head Self-Attention Mechanism Module (MSAM). Specifically, the MFEM extracts character features, word boundary features, radical features, and pinyin features of Chinese characters. The acquired font shape, font sound, and font meaning features are fused to enhance the semantic information of Chinese characters with different granularities. Moreover, the MSAM is used to capture the dependencies between characters in a multi-dimensional subspace to better understand the semantic features of the context. Extensive experimental results on four benchmark datasets show that our method improves the overall performance of the CNER model.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.3
• /
• pp.740-753
• /
• 2023

Text recognition in natural scene images is a challenging problem in computer vision. The accurate identification of ship number characters can effectively improve the level of ship traffic management. However, due to the blurring caused by motion and text occlusion, the accuracy of ship number recognition is difficult to meet the actual requirements. To solve these problems, this paper proposes a dual-branch network based on the CRNN identification network. The network couples image restoration and character recognition. The CycleGAN module is used for blur restoration branch, and the Pix2pix module is used for character occlusion branch. The two are coupled to reduce the impact of image blur and occlusion. Input the recovered image into the text recognition branch to improve the recognition accuracy. After a lot of experiments, the model is robust and easy to train. Experiments on CTW datasets and real ship maps illustrate that our method can get more accurate results.

• Journal of Korea Game Society
• /
• v.15 no.6
• /
• pp.7-16
• /
• 2015

To improve the sense of unity between game players and game world avatars, we propose a method to measure game player's physical status and manipulate game character appearances and behavior using the measurements. To measure the status, we implemented a hardware module that perceive the body temperature. We implemented a 3D horror game whose game space is a common subway space ordinarily encountered in daily life so that ordinary people can easily get used to the game environment. The experiments of applying the measurement data from the game player to the game world avatar shows that the player's physical status can be effectively utilized to improve the sense of reality in gameplay.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.8B
• /
• pp.805-811
• /
• 2009

Firstly, this dissertation suggests the tag ID transfer method using two-dimensional arrangement. Secondly, provide better and stable security system compare to existing one by transferring tag; ID using established two-dimensional arrangement. Thirdly, provide operating module, which possible to descramble two-dimensional arrangement, with a character set when descrambling in a tag and a server. Lastly, suggest safe key transfer using a character set and two-dimensional arrangement. In order to embody suggested system and assess, transferred two-dimensional arrangement several times to carry out the experiment. Confirmed that it is impossible for suggested system to decode key patterns compare to existing RFID systems.

• Journal of the Institute of Convergence Signal Processing
• /
• v.2 no.1
• /
• pp.1-11
• /
• 2001

This paper presents a learning-based approach for the construction of license Plate recognition system. The system consist of three modules. They are respectively, car detection module, license plate recognition module and recognition module. Car detection module detects a car in the given image sequence obtained from the camera with simple color-based approach. Segmentation module extracts the license plate in detect car image using neural network as filters for analyzing the color and texture properties of license plate. Recognition module then reads characters in detected license plate with support vector machine (SVM)-based characters recognizer. The system has been tested from parking lot and tollgate, etc. and have show the following performances on average: Car detect rate 100%, segmentation rate 97.5%, and character recognition rate about 97.2%. Overall system performances is 94.7% and processing time is one sec. Then our propose system does well using real world.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.5 no.4
• /
• pp.29-38
• /
• 1997

In this paper, brake steering performance of electric-driven special-purpose vehicles is investigated. A 14 DOF model is developed considering nonlinear character- istics of the suspension and tire. Based on the model, cornering performance is analyzed for brake steering, acceleration steering and pivoting, respectively. Simulation results are obtained based on the developed SIMULINK module. This analysis about the non steady state cornering performance is particularly important for armored vehicles because the projected route of the vehicle at emergency should be predicted accuracy.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.5
• /
• pp.1130-1139
• /
• 1998

This paper presents an integrated solution for the content-based news video indexing and the retrieval. Currently, it is impossible to automatically index a general video, but we can index a specific structural video such as news videos. Our proposed model extracts automatically the key frames by using the structured knowledge of news and consists of the news item segmentation, caption recognition and search browser modules. We present above three modules in the following: the news event segmentation module recognizes an anchor-person shot based on face recognition, and then its news event are divided by the anchor-person's frame information. The caption recognition module detects the caption-frames with the caption characteristics, extracts their character region by the using split-merge method, and then recognizes characters with OCR software. Finally, the search browser module could make a various of searching mechanism possible.

• Proceedings of the IEEK Conference
• /
• 2002.06c
• /
• pp.223-226
• /
• 2002

If one can easily modify the existing motion data to a new motion in making an animation movie, he can save a lot of time for graphic design. To implement this kind of system, we propose a PC-based system composed of low cost commercial animation tool (3D Studio Max) for visualization of the animation and motion editing module that handles optimization process during the motion transform. Researchers studying advanced motion transform techniques only have to focus on the mathematical manipulation of the motion data

• Journal of the Korean Mathematical Society
• /
• v.36 no.3
• /
• pp.593-607
• /
• 1999

Let denote the orthosymplectic Lie superalgebra spo (2m,1). For each irreducible -module, we describe its character in terms of tableaux. Using this result, we decompose kV, the k-fold tensor product of the natural representation V of , into its irreducible -submodules, and prove that the Brauer algebra Bk(1-2m) is isomorphic to the centralizer algebra of spo(2m, 1) on kV for m .

• Journal of Broadcast Engineering
• /
• v.9 no.2
• /
• pp.151-163
• /
• 2004

In this paper, we propose a content-based video indexing and retrieval system using MPEG-7 standard to retrieve and manage videos efficiently. The proposed system consists of video indexing module for a video DB and video retrieval module to allow various query methods on a web environment. Video indexing module stores metadata such as manually typed in keywords, automatically recognized character names, and MPEG-7 visual descriptors extracted by indexing module into a DB in a sever side. A user can access to retrieval module by a web and retrieve desired videos through various query methods like keywords, faces, example and sketch. For this retrieval system, we propose ATC(Adaptive Twin Comparison) as a cut detection method for efficient video indexing and QBME(Query By Modified Example) as an improved content-based query method for the convenience of users. Experimental results show that the proposed ATC method detects cuts well and the proposed QBME method provides the conveniences better than existing query methods such as QBE(Query By Example) and QBS(Query By Sketch).