• 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.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.9
• /
• pp.407-418
• /
• 2023

The Multi-Object Goal Visual Navigation(MultiOn) is a visual navigation task in which an agent must visit to multiple object goals in an unknown indoor environment in a given order. Existing models for the MultiOn task suffer from the limitation that they cannot utilize an integrated view of multimodal context because use only a unimodal context map. To overcome this limitation, in this paper, we propose a novel deep neural network-based agent model for MultiOn task. The proposed model, MCFMO, uses a multimodal context map, containing visual appearance features, semantic features of environmental objects, and goal object features. Moreover, the proposed model effectively fuses these three heterogeneous features into a global multimodal context map by using a point-wise convolutional neural network module. Lastly, the proposed model adopts an auxiliary task learning module to predict the observation status, goal direction and the goal distance, which can guide to learn the navigational policy efficiently. Conducting various quantitative and qualitative experiments using the Habitat-Matterport3D simulation environment and scene dataset, we demonstrate the superiority of the proposed model.

• Journal of the Korean Society for Library and Information Science
• /
• v.48 no.4
• /
• pp.71-90
• /
• 2014

This study aims at examining the ways which a school librarian uses SNS when performing tasks and the use of SNS with the features of the task and the organizational system run by one person affect task performance. In this study, for objective explanation of the interrelation among a librarian's tasks and the use of SNS and performance, the model of a Librarian's Task-SNS Technology Fit was suggested through the combination of the Technology Acceptance Model and the Task-Technology Fit. Based on the result of the survey analysis, their utilization of SNS is not just limited to personal taste nor interferes with their work. Rather, it enhances task performance owing to the characteristics of the tasks or the organization such as the organizational system run by one person.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.05a
• /
• pp.419-421
• /
• 2018

Cloud vehicular networks are a promising paradigm to improve vehicular through distributing computation tasks between remote clouds and local vehicular terminals. Software-Defined Network(SDN) can bring advantages to Intelligent Transportation System(ITS) through its ability to provide flexibility and programmability through a logically centralized controlled cluster that has a full comprehension of view of the network. However, as the SDN paradigm is currently studied in vehicular ad hoc networks(VANETs), adapting it to work on cloud-based vehicular network requires some changes to address particular computation features such as task computation of applications of cloud-based vehicular networks. There has been initial work on briging SDN concepts to vehicular networks to reduce the latency by using the fog computing technology, but most of these studies do not directly tackle the issue of task computation. This paper proposes a Software-Defined Cloud-based vehicular Network called SDCVN framework. In this framework, we study the effectiveness of task computation of applications of cloud-based vehicular networks with vehicular cloud and roadside edge cloud. Considering the edge cloud service migration due to the vehicle mobility, we present an efficient roadside cloud based controller entity scheme where the tasks are adaptively computed through vehicular cloud mode or roadside computing predictive trajectory decision mode. Simulation results show that our proposal demonstrates a stable and low route setup time in case of installing the forwarding rules of the routing applications because the source node needs to contact the controller once to setup the route.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.05a
• /
• pp.238-240
• /
• 2018

Cloud vehicular networks are a promising paradigm to improve vehicular through distributing computation tasks between remote clouds and local vehicular terminals. Software-Defined Network(SDN) can bring advantages to Intelligent Transportation System(ITS) through its ability to provide flexibility and programmability through a logically centralized controlled cluster that has a full comprehension of view of the network. However, as the SDN paradigm is currently studied in vehicular ad hoc networks(VANETs), adapting it to work on cloud-based vehicular network requires some changes to address particular computation features such as task computation of applications of cloud-based vehicular networks. There has been initial work on briging SDN concepts to vehicular networks to reduce the latency by using the fog computing technology, but most of these studies do not directly tackle the issue of task computation. This paper proposes a Software-Defined Cloud-based vehicular Network called SDCVN framework. In this framework, we study the effectiveness of task computation of applications of cloud-based vehicular networks with vehicular cloud and roadside edge cloud. Considering the edge cloud service migration due to the vehicle mobility, we present an efficient roadside cloud based controller entity scheme where the tasks are adaptively computed through vehicular cloud mode or roadside computing predictive trajectory decision mode. Simulation results show that our proposal demonstrates a stable and low route setup time in case of installing the forwarding rules of the routing applications because the source node needs to contact the controller once to setup the route.

• Journal of Korea Multimedia Society
• /
• v.22 no.9
• /
• pp.1000-1010
• /
• 2019

Image semantic segmentation and dehazing are key tasks in the computer vision. In recent years, researches in both tasks have achieved substantial improvements in performance with the development of Convolutional Neural Network (CNN). However, most of the previous works for semantic segmentation assume the images are captured in clear weather and show degraded performance under hazy images with low contrast and faded color. Meanwhile, dehazing aims to recover clear image given observed hazy image, which is an ill-posed problem and can be alleviated with additional information about the image. In this work, we propose a deep multi-task network for simultaneous semantic segmentation and dehazing. The proposed network takes single haze image as input and predicts dense semantic segmentation map and clear image. The visual information getting refined during the dehazing process can help the recognition task of semantic segmentation. On the other hand, semantic features obtained during the semantic segmentation process can provide cues for color priors for objects, which can help dehazing process. Experimental results demonstrate the effectiveness of the proposed multi-task approach, showing improved performance compared to the separate networks.

• Journal of Korea Multimedia Society
• /
• v.16 no.7
• /
• pp.820-828
• /
• 2013

Extracting unique features from an image is a fundamental issue when making panorama images, acquiring stereo images, recognizing objects and analyzing images. Generally, the task to compare features to other images requires much computing time because some features are formed as a vector which has many elements. In this paper, we present a method that compares features after reducing the feature dimension extracted from an image using PCA(principal component analysis) and sorting the features in a linked list. SURF(speeded up robust features) is used to describe image features. When the dimension reduction method is applied, we can reduce the computing time without decreasing the matching accuracy. The proposed method is proved to be fast and robust in experiments.

• Proceedings of the KIEE Conference
• /
• 2005.07d
• /
• pp.2489-2491
• /
• 2005

Currently, Programmable Logic Contorller(PLC) uses Real Time Operation System(RTOS) as basic OS. RTOS executes defined results as to defined time. General features of RTOS emphasize the priority in each task, high-speed process of external interrupt, task scheduling, synchronization in task, the limitation of memory capacity. For safety critical placement, PLC software needs Verification and Validation(V&V). For example, nuclear power plant. In this paper, PLC RTOS is verified by formal methods. Particularly, formal method V&V uses verification tool called 'STATEMATE', and shows the results.

• International journal of advanced smart convergence
• /
• v.5 no.3
• /
• pp.1-7
• /
• 2016

In existing Convolutional Neural Network (CNNs) for object recognition task, there are only few efforts known to reduce the noises from the images. Both convolution and pooling layers perform the features extraction without considering the noises of the input image, treating all pixels equally important. In computer vision field, there has been a study to weight a pixel importance. Seam carving resizes an image by sacrificing the least important pixels, leaving only the most important ones. We propose a new way to combine seam carving approach with current existing CNN model for object recognition task. We attempt to remove the noises or the "unimportant" pixels in the image before doing convolution and pooling, in order to get better feature representatives. Our model shows promising result with CIFAR-10 dataset.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1994.10a
• /
• pp.559-564
• /
• 1994

Recently, a visual servoing for an eye-in-hand robot has become an interesting problem. A distance between a camera and a task object is very useful information for visual servoing. In the previous works for visual servoing, the distance can be obtained from the difference between a reference and a measured feature value of the object such as area on image plane. However, since this feature depends on the object, the reference feature value must be changed when other task object is taken. To overcome this difficulty, this paper presents a novel method for visual servoing. In the proposed method, a blur is used to obtain the distance. The blur, one of the most important features, depends on the focal length of camera. Since it is not affected by the change of object, the reference feature value is not changed although other task object is taken. In this paper, we show a relationship between the distance and the blur, and define the feature jacobian matrix based on camera defocusing to operate the robot. A series of experiments is performed to verify the proposed method.