• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.2
• /
• pp.855-870
• /
• 2020

Windows are key components of building facades. Detecting windows, crucial to 3D semantic reconstruction and scene parsing, is a challenging task in computer vision. Early methods try to solve window detection by using hand-crafted features and traditional classifiers. However, these methods are unable to handle the diversity of window instances in real scenes and suffer from heavy computational costs. Recently, convolutional neural networks based object detection algorithms attract much attention due to their good performances. Unfortunately, directly training them for challenging window detection cannot achieve satisfying results. In this paper, we propose an approach for window detection. It involves an improved Faster R-CNN architecture for window detection, featuring in a window region proposal network, an RoI feature fusion and a context enhancement module. Besides, a post optimization process is designed by the regular distribution of windows to refine detection results obtained by the improved deep architecture. Furthermore, we present a newly collected dataset which is the largest one for window detection in real street scenes to date. Experimental results on both existing datasets and the new dataset show that the proposed method has outstanding performance.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.11 no.3
• /
• pp.43-51
• /
• 2008

In this paper, we present WebGIS techniques that can acquire more information from users who offered information and location of region's crops. So that it can be search the information based on Ontology defining Metadata for understand and control more accurately. And this paper shows how to implement about prototype of crops information system for obtaining information of location. Our object is to offer results Service form search, to analyze question of user and to show the exact geographic information about question of user. So this paper can be provide convenience to users that can be show Semantic WebGIS system.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2017.05a
• /
• pp.93-94
• /
• 2017

This study have derived the big issue of overseas construction through big data analysis. For identification of big issues on overseas construction, domestic online articles, 30 daily newspapers like the JoongAng Ilbo, 7 construction related articles including construction economy and 1,759 local newspapers and small media companies were analyzed from October 1st, 2015 to September 30th, 2016. 13,884 cases in total were used for big data analyses and big issue candidates were identified. The analysis result is as shown below. First, looking into major issues on overseas construction for a year, construction orders in the Middle East decreased because of the drop in oil prices. Accordingly, there were discussions on concerns and crises we may face as profitabilities worsened in overseas construction. Second, analyzing main concern based on 8 key words on overseas construction among construction issues for the last one year, it was found as following: Region (29.4%), Business environment (21.4%), Group (15.8%), Profitability (14.5%), Policy and Institution (7.8%), Market environment (4.2%), Business (project) (4.15%), and Education (3.2%). Third, among 30 issues on 8 key words, 10 key issues that are likely to spread and continue were identified. Then, a semantic network map among key words and centrality were analyzed.

• Proceedings of the Korean Institute of Landscape Architecture Conference
• /
• 2007.10b
• /
• pp.214-219
• /
• 2007

Psychological and physiological effects of plants were studied by investigating human responses while observing plants. Eighteen healthy adult male(aged between $19{\sim}25$ years) participated in this study. Semantic differential method(SD method) and multi-channel near-infrared spectroscopy(NIRS) were used to survey verbal and non-verbal response, respectively. Cerebral hemodynamics as a new evaluation index of brain activity was recorded for right brain hemisphere where visual information is mainly delivered. Thirty seconds of cerebral blood flow in forty seven channels were calculated when watching five types of picture images with different rates of hedge against gray block wall; 0:10, 3:7, 5:5, 7:3, 10:0. In the SD results, similar evaluations were found in all subjects. However, the change of cerebral hemodynamics as a non-verbal response varied among subjects. Largely two patterns of hemodynamics change were found with increasing plants rate in picture images; group A showed significant decreases of blood flow volume in many cortical regions, Group B had significant increase of blood flow volume in the occipital region for the scenes seen comparatively more plant. Our findings on the cerebral hemodynamics may indicate that there are two patterns of brain activity during observation of plants; group A in which brain areas associated with visual information and thinking work simultaneously to the visual stimuli; group B in which brain areas associated only with visual information work.

• Journal of Biomedical Engineering Research
• /
• v.27 no.4
• /
• pp.170-179
• /
• 2006

The purpose of this study was to investigate the spatiotemporal cortical activation pattern and functional connectivity during visual perception of words. 61 channel recordings of electroencephalogram were obtained from 15 subjects while they were judging the meaning of Korean, English, and Chinese words with concrete meanings. We examined event-related potentials (ERP) and applied independent component analysis (ICA) to find and separate simultaneously activated neural sources. Spectral analysis was also performed to investigate the gamma-band activity (GBA, 30-50 Hz) which is known to reflect feature binding. Five significant ERP components were identified and left hemispheric dominance was observed for most sites. Meaningful differences of amplitudes and latencies among languages were observed. It seemed that familiarity with each language and orthographic characteristics affected the characteristics of ERP components. ICA helped confirm several prominent sources corresponding to some ERP components. The results of spectral and time-frequency analyses showed distinct GBAs at prefrontal, frontal, and temporal sites. The GBAs at prefrontal and temporal sites were significantly correlated with the LPC amplitude and response time. The differences in spatiotemporal patterns of GBA among languages were not prominent compared to the inter-individual differences. The gamma-band coherence revealed short-range connectivity within frontal region and long-range connectivity between frontal, posterior, and temporal sites.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.11
• /
• pp.5555-5567
• /
• 2018

State-of-art instance segmentation networks are successful at generating 2D segmentation mask for region proposals with highest classification score, yet 3D object segmentation task is limited to geocentric embedding or detector of Sliding Shapes. To this end, we propose an amodal 3D instance segmentation network called A3IS-CNN, which extends the detector of Deep Sliding Shapes to amodal 3D instance segmentation by adding a new branch of 3D ConvNet called A3IS-branch. The A3IS-branch which takes 3D amodal ROI as input and 3D semantic instances as output is a fully convolution network(FCN) sharing convolutional layers with existing 3d RPN which takes 3D scene as input and 3D amodal proposals as output. For two branches share computation with each other, our 3D instance segmentation network adds only a small overhead of 0.25 fps to Deep Sliding Shapes, trading off accurate detection and point-to-point segmentation of instances. Experiments show that our 3D instance segmentation network achieves at least 10% to 50% improvement over the state-of-art network in running time, and outperforms the state-of-art 3D detectors by at least 16.1 AP.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.06a
• /
• pp.322-325
• /
• 2018

영상의 관심 영역 검출은 영상처리 및 컴퓨터 비전 응용 분야에서 꾸준하게 사용되고 있는 기법이다. 특히, 근래 심층신경망 연구의 급격한 발전에 힘입어 심층신경망을 이용한 관심 영역 검출 기법에 대한 연구가 활발하게 진행되고 있다. 한편 Fully Convolutional Network(이하 FCN)은 본래 심층 예측(Dense Prediction)을 통한 의미론적 영상 분할(Semantic Segmentation)을 수행하기 위해 제안된 심층신경망 구조이다. FCN을 영상의 관심 영역 검출에 활용하여도 기존 관심 영역 검출 기법과 비교하여 충분히 좋은 성능을 발휘할 수 있다. 그러나 FCN에 사용되는 convolution 층의 수가 많고, 이에 따른 가중치(weight)의 개수도 기하급수적으로 늘어나 검출에 필요한 시간 복잡도가 매우 크다는 문제점이 있다. 따라서 본 논문에서는 기존 FCN이 가진 검출 시간 복잡도의 문제점을 convolution 층의 가중치 관점에서 해결하고자 이를 조절하여 FCN의 관심 영역 검출 속도를 향상시키는 방법을 제안한다. 적절한 convolution 층의 가중치를 조절함으로써, MSRA10K 데이터셋 환경에서 검출 정확도를 크게 저하시키지 않고도 최대 약 20.5%만큼 검출 속도를 향상시킬 수 있었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.10
• /
• pp.3390-3405
• /
• 2022

Long Short-Term Memory (LSTM) combined with attention mechanism is extensively used to generate semantic sentences of images in image captioning models. However, features of salient regions and spatial information are not utilized sufficiently in most related works. Meanwhile, the LSTM also suffers from the problem of underutilized information in a single time step. In the paper, two innovative approaches are proposed to solve these problems. First, the Synergy-Gated Attention (SGA) method is proposed, which can process the spatial features and the salient region features of given images simultaneously. SGA establishes a gated mechanism through the global features to guide the interaction of information between these two features. Then, the Recurrent Fusion LSTM (RF-LSTM) mechanism is proposed, which can predict the next hidden vectors in one time step and improve linguistic coherence by fusing future information. Experimental results on the benchmark dataset of MSCOCO show that compared with the state-of-the-art methods, the proposed method can improve the performance of image captioning model, and achieve competitive performance on multiple evaluation indicators.

• International journal of advanced smart convergence
• /
• v.13 no.1
• /
• pp.108-113
• /
• 2024

The purpose of this study is to analyze user perceptions of tourism platforms through big data. Data were collected from Naver, Daum, and Google as big data analysis channels. Using semantic network analysis with the keyword 'tourism platform,' a total of 29,265 words were collected. The collection period was set for two years, from August 31, 2021, to August 31, 2023. Keywords were analyzed for connected networks using TexTom and Ucinet programs for social network analysis. Keywords perceived by tourism platform users include 'travel,' 'diverse,' 'online,' 'service,' 'tourists,' 'reservation,' 'provision,' and 'region.' CONCOR analysis revealed four groups: 'platform information,' 'tourism information and products,' 'activation strategies for tourism platforms,' and 'tourism destination market.' This study aims to expand and activate services that meet the needs and preferences of users in the tourism field, as well as platforms tailored to the changing market, based on user perception, current status, and trend data on tourism platforms.

• Journal of the Korean Geographical Society
• /
• v.41 no.5 s.116
• /
• pp.582-599
• /
• 2006

Due to the growing popularity of mobile information technology, more people, especially in the general public, have access to computerized geospatial information systems for wayfinding tasks or urban navigation. One of the problems with the current services is that, whether the users are exploring or navigating, whether they are travelers who are totally new to a region or long-term residents who have a fair amount of regional knowledge, the same method is applied and the direction are given in the same way. However, spatial knowledge for a given urban region expands in proportion to residency. Urban navigation is highly dependent on cognitive mental images, which is developed through spatial experience and social communication. Thus, the wayfinding service for a regional community can be highly supported, using well-known regional places. This research is to develop the framework for urban navigation within a regional community. The concept of communal ontology is proposed to aid in urban regional navigation. The experimental work was implemented with case study to collect regional landmarks, develop the ontological model and represent it with formal structure. The final product of this study will provide the geographical information of a region to the other agent and be the fundamental information structure for cognitive urban regional navigation.