• International Journal of Computer Science & Network Security
• /
• v.22 no.10
• /
• pp.406-412
• /
• 2022

Scene classification of very high-resolution (VHR) imagery can attribute semantics to land cover in a variety of domains. Real-world application requirements have not been addressed by conventional techniques for remote sensing image classification. Recent research has demonstrated that deep convolutional neural networks (CNNs) are effective at extracting features due to their strong feature extraction capabilities. In order to improve classification performance, these approaches rely primarily on semantic information. Since the abstract and global semantic information makes it difficult for the network to correctly classify scene images with similar structures and high interclass similarity, it achieves a low classification accuracy. We propose a VHR remote sensing image classification model that uses extracts the global feature from the original VHR image using an EfficientNet-V2L CNN pre-trained to detect similar classes. The image is then classified using a multilayer perceptron (MLP). This method was evaluated using two benchmark remote sensing datasets: the 21-class UC Merced, and the 38-class PatternNet. As compared to other state-of-the-art models, the proposed model significantly improves performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.12
• /
• pp.3364-3382
• /
• 2023

Remote sensing image segmentation plays an important role in realizing intelligent city construction. The current mainstream segmentation networks effectively improve the segmentation effect of remote sensing images by deeply mining the rich texture and semantic features of images. But there are still some problems such as rough results of small target region segmentation and poor edge contour segmentation. To overcome these three challenges, we propose an improved semantic segmentation model, referred to as MRU-Net, which adopts the U-Net architecture as its backbone. Firstly, the convolutional layer is replaced by BasicBlock structure in U-Net network to extract features, then the activation function is replaced to reduce the computational load of model in the network. Secondly, a hybrid multi-scale recognition module is added in the encoder to improve the accuracy of image segmentation of small targets and edge parts. Finally, test on Massachusetts Buildings Dataset and WHU Dataset the experimental results show that compared with the original network the ACC, mIoU and F1 value are improved, and the imposed network shows good robustness and portability in different datasets.

• Journal of KIISE
• /
• v.45 no.1
• /
• pp.94-98
• /
• 2018

We propose Wide Inception ResNet (WIR Net) an optimized neural network architecture as a real-time semantic segmentation method for autonomous driving. The neural network architecture consists of an encoder that extracts features by applying a residual connection and inception module, and a decoder that increases the resolution by using transposed convolution and a low layer feature map. We also improved the performance by applying an ELU activation function and optimized the neural network by reducing the number of layers and increasing the number of filters. The performance evaluations used an NVIDIA Geforce GTX 1080 and TX1 boards to assess the class and category IoU for cityscapes data in the driving environment. The experimental results show that the accuracy of class IoU 53.4, category IoU 81.8 and the execution speed of $640{\times}360$, $720{\times}480$ resolution image processing 17.8fps and 13.0fps on TX1 board.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.7
• /
• pp.1884-1896
• /
• 1999

In this paper, we propose an ATM API for Java pplication programming. The proposed Java ATM API is an extended form of java.net package of the Java Core API. Our Java ATM API is defined based on the ATM Forum's semantic standard for native ATM services, "Native ATM Service : Semantic Description, Version 1.0". IN order to provide native ATM services, we defined several new classes within the java.net package, including AtmAddress for AtmAddress, AtmSocket, ATM BLLI and AtmBHLI for Atm BLLI and BHLI information, AtmServerSocket, AtmMulticastSocket, AtmSocketImpl for socket programming over native ATM communication, and AtmConnAttr for native ATM connection characteristics. Software structure for construcing the Java ATM API over Winsock 2 environment and its implementation method are also presented.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.991-1005
• /
• 2022

Satellite imaging is an effective supplementary data source for detecting and verifying nuclear activity. It is also highly beneficial in regions with limited access and information, such as nuclear installations. Time series analysis, in particular, can identify the process of preparing for the conduction of a nuclear experiment, such as relocating equipment or changing facilities. Differences in the semantic segmentation findings of time series photos were employed in this work to detect changes in meaningful items connected to nuclear activity. Building, road, and small object datasets made of KOMPSAT 3/3A photos given by AIHub were used to train deep learning models such as U-Net, PSPNet, and Attention U-Net. To pick relevant models for targets, many model parameters were adjusted. The final change detection was carried out by including object information into the first change detection, which was obtained as the difference in semantic segmentation findings. The experiment findings demonstrated that the suggested approach could effectively identify altered pixels. Although the suggested approach is dependent on the accuracy of semantic segmentation findings, it is envisaged that as the dataset for the region of interest grows in the future, so will the relevant scope of the proposed method.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.04b
• /
• pp.333-335
• /
• 2000

컴퓨터는 자연언어로 된 문장을 올바르게 이해하기 위해 의미지식을 필요로 하며 이러한 의미지식을 정확하게 구축하기 위해서는 수작업을 필요로 한다. 그러나 수작업에 의한 의미지식 구축은 많은 비용과 시간을 필요로 하고, 작성자의 주관이 개입되며, 응용 도메인에 따라 의미지표 테이블이 수정되면 의미지표 사전의 재구축이 불가피하다. 이러한 문제점을 해결하기 위해 본 논문에서는 영어 WordNet과 한영 사전을 이용한 한국어 명사 의미지표 사전의 자동 구축 방법을 제안한다.

• Language and Information
• /
• v.14 no.1
• /
• pp.165-196
• /
• 2010

The primary goal of this paper is to propose methods of enriching two heterogeneous ontologies: Sejong Semantic Classes (SJSC) and KorLexNoun 1.5 (KLN). In order to achieve this goal, this study introduces the pros and cons of two ontologies, and analyzes the error patterns found during the fine-grained manual mapping processes between them. Error patterns can be classified into four types: (1) structural defectives involved in node branching, (2) errors in assigning the semantic classes, (3) deficiency in providing linguistic information, and (4) lack of the lexical units representing specific concepts. According to these error patterns, we propose different solutions in order to correct the node branching defectives and the semantic class assignment, to complement the deficiency of linguistic information, and to increase the number of lexical units suitably allotted to their corresponding concepts. Using the results of this study, we can obtain more enriched ontologies by correcting the defects and errors in each ontology, which will lead to the enhancement of practicality for syntactic and semantic analysis.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.2
• /
• pp.276-282
• /
• 2011

CoreNet is a valuable resource to use in the domain of natural language processing including Korean-Chinese-Japanese multilingual text analysis, and translation among natural languages. CoreNet is mapped to SUMO in order to encourage its application in broader fields and enhance its international status as a multilingual lexical semantic network. To do this, indirect and direct mapping methodologies are used. Through the indirect mapping among CoreNet-KorLex-PWN-SUMO, we alleviate the difficulty of translating CoreNet concept terms in Korean into SUMO concepts in English, and maximize recall of SUMO concepts corresponding to the concept of CoreNet.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.12
• /
• pp.97-105
• /
• 2016

Network-Centric Operational Environment(NCOE) service becomes critical in today's military environment network because reusability of service and interaction are being increasingly important as well in business process. However, the refinement of service by semantic similarity and functional similarity at the business process was not detailed yet. In order to enhance accuracy of refining of business service, in this study, the authors introduce a method for refining service by semantic similarity and functional similarity in BPMN model. The business process are designed in a BPMN model. In this model, candidated services are refined through binding related activities by the analysis result of semantic similarity based on word-net and functional similarity based on properties specification between activities. Then, the services are identified through refining the candidated service. The proposed method is expected to enhance the service identification with accuracy and modularity. It also can accelerate more standardized service refinement developments by the proposed method.

• Nuclear Engineering and Technology
• /
• v.56 no.1
• /
• pp.340-356
• /
• 2024

Bridging traditional experts' disciplinary boundaries is important for nuclear knowledge management systems. However, expert competences are often described in unstructured texts and require substantial human effort to link related competences across disciplines. The purpose of this research is to develop and evaluate a natural language processing approach, based on Latent Semantic Analysis, to enable the automatic linking of related competences across different disciplines and communities of practice. With datasets of unstructured texts as input training data, our results show that the algorithm can readily identify nuclear domain-specific semantic links between words and concepts. We discuss how our results can be utilized to generate a quantitative network of links between competences across disciplines, thus acting as an enabler for identifying and bridging communities of practice, in nuclear and beyond.