• Journal of the Korea Computer Graphics Society
• /
• v.23 no.3
• /
• pp.31-38
• /
• 2017

In this paper, we propose a system which estimates Manhattan coordinate systems for urban scene images using a convolutional neural network (CNN). Estimating the Manhattan coordinate system from an image under the Manhattan world assumption is the basis for solving computer graphics and vision problems such as image adjustment and 3D scene reconstruction. We construct a CNN that estimates Manhattan coordinate systems based on GoogLeNet [1]. To train the CNN, we collect about 155,000 images under the Manhattan world assumption by using the Google Street View APIs and calculate Manhattan coordinate systems using existing calibration methods to generate dataset. In contrast to PoseNet [2] that trains per-scene CNNs, our method learns from images under the Manhattan world assumption and thus estimates Manhattan coordinate systems for new images that have not been learned. Experimental results show that our method estimates Manhattan coordinate systems with the median error of $3.157^{\circ}$ for the Google Street View images of non-trained scenes, as test set. In addition, compared to an existing calibration method [3], the proposed method shows lower intermediate errors for the test set.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.11 no.9
• /
• pp.923-928
• /
• 2016

The various technologies related to the monitoring human health in real-time for the emergency situations are developing these days. Mostly the human pulse is used for measuring as the vital signs so far, but the EEG became a major research trend now. However, there are some problems measuring and sending EEG signals of all the people walking down the street to the dedicated server. Especially, there are some restrictions for collecting and sending EEG signals in 2-dimensional space in real-time. Therefore, I suggests an efficient network model using 3-dimensional space of drones to avoid the restrictions. The models are designed, simulated, and evaluated with the Opnet simulator.

• ETRI Journal
• /
• v.41 no.2
• /
• pp.235-241
• /
• 2019

In this paper, we propose a rank-weighted reconstruction feature to improve the robustness of a feed-forward deep neural network (FFDNN)-based acoustic model. In the FFDNN-based acoustic model, an input feature is constructed by vectorizing a submatrix that is created by slicing the feature vectors of frames within a context window. In this type of feature construction, the appropriate context window size is important because it determines the amount of trivial or discriminative information, such as redundancy, or temporal context of the input features. However, we ascertained whether a single parameter is sufficiently able to control the quantity of information. Therefore, we investigated the input feature construction from the perspectives of rank and nullity, and proposed a rank-weighted reconstruction feature herein, that allows for the retention of speech information components and the reduction in trivial components. The proposed method was evaluated in the TIMIT phone recognition and Wall Street Journal (WSJ) domains. The proposed method reduced the phone error rate of the TIMIT domain from 18.4% to 18.0%, and the word error rate of the WSJ domain from 4.70% to 4.43%.

• Journal of architectural history
• /
• v.20 no.6
• /
• pp.169-184
• /
• 2011

The following research of the road network of Jeju-Eupseong during Daehan Empire period has a twofold purpose: to study some characteristics of the road network at that time; and, to restore it to the original form of that period before a newly constructed road, called Shinjakro, has been established. As an attempt to trace the old shape of Jeju-Eupseong, this study analyzed some historical factors based on the first land cadastral map which was made in 1914, including outskirts of Jeju-Eupseong; such as castle itself, castle gate, road, bridge, lots of land, etc. Then this study also tried to restore the old road network of Jeju-Eupseong, through finding the original land-lot shape in the land cadastral map. There was five Shinjakro made between 1914 and 1917. The road network before then was composed of the double east-west axes and the single north-south axis. These axes was connected to some important place of the inside of Jeju-Eupseong; such as castle gates, fountains, Gaek-sa, etc. There were many branch lines between these main axes at about 80-120m intervals. Also there was an outer road along the outer wall of castle, connected with each castle gates. Especially, the north-west axis was the baseline which divided into two large parts, a government office area and non-government area (housing and commercial street for the people). Finally, this paper examines that the road network of Jeju-Eupseong was the true result for the efficient function of the city, especially considering natural geographical conditions and environment of living of that time.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.435-438
• /
• 2003

A townscape has been a main factor in urban-development problems in Japan. In the townscape, keeping harmony with environment is a common goal. But useful and meaningful goals are expressing individuality and impression of the town in the townscape. In this paper, we propose the colony planning support system system to improve the townscape. The system finds propositional colour combinations based on three elements, town image, colour harmony, and cost. The targets of this model are mostly townscapes in residential areas that already exist, In this paper, we introduce the construction of a Kansei evaluation model to quantify the impression. First, we conducted computer-based evaluational experiments for 20 subjects using the SD method to clarify the relationship between town image and street colours. We chose 16 adjective words related to town image and prepared 100 colour picture samples for the evaluation. After the experiments, we constructed the model using a neural network for each word. We chose 62 experimental results for the training data of the neural network and 20 results for the testing data. Each colour in the data was selected to have unique hue, brightness or saturation attributes, After the construction, we tested the model for accuracy. We input the testing data into the constructed model and calculated errors between the output from the model and the experimental results. Testing of the model showed that the model worked well for more than 80％ of the samples. The model demonstrated influences of colours on the town image.

• Korean Institute of Interior Design Journal
• /
• v.13 no.2
• /
• pp.64-71
• /
• 2004

This study attempts to clarify inter-relationship between design expression in the surface of contemporary multi-family housing and physical trends in this digital age. The study examines characteristics of housing design influenced from new digital technology in the forms of metaphor such as dematerialization, fluidity and mobility. Currently constructed multi-family housing in Europe after 1990 has been analyzed to find digital reflections in the surfaces. As a conclusion, findings are as follows; In this digital age, a house is considered as a base in the forms of boxes and containers to network to outer worlds. In a complex of multi-family housing, public spaces move to the inside and become parts of private domain. Therefore, a house directly faces a street/outer world without transitional areas and devices. The facades of housing become delicate skins to control private domain in the inside and the world in the outside. In this digital age, the surfaces of housing are designed by ways of wrapping and screening. Printing and mapping methods are used to surface design because they are manipulative and flexible methods in design like graphic design. Screening devices using louvers and shutters are powerful tools to make random and unpredictable images. If units of a louver and a shutter were pixels, a whole surface would be a digital screen. In such assumption, the facade of buildings reflects digital screen to interface while the louvers in operation reflect clicking to network. As a conclusion, design expression in surfaces in-between of multi-family housing is a metaphor of digital mechanism.

• Journal of the Korean housing association
• /
• v.24 no.5
• /
• pp.57-67
• /
• 2013

The purpose of this study is to find out distributional principles of the shop-houses in residential district through physical composition of residential area. Through the research that are based on literature and field survey, it analysed the relation between physical composition of residential area and the shop-houses of selected four sites under similar urban condition where land readjustment project were held. The consequences of this study are summarized as follows; 1) A road that is not penetrating the inside a residential area is seemed to have low occupancy of shops comparing to other roads with similar hierarchy. And the shops functions as convenience facilities. It can be inferred from the statement above that even a road with higher hierarchy is not easy to form commercial area if without penetrating the residential area. 2) According to the size and the arranged method of Inner blocks of residential area towards major roads, distribution of shop-houses appear to be different. In other words, when shop-houses are located along the roads, the proportion of shop was higher than when located perpendicular to the roads. 3) When the number of average lots are little and the size of block is small, usually, street ratio and the number of corners gets higher, which eventually increases the number of shop-houses.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.42 no.5
• /
• pp.1038-1046
• /
• 2017

In this paper, we propose the design of LED Streetlight Platform with capabilities of weather monitoring and color temperature control. Several previous works are focused on the energy efficiency or data management of streetlight system, but no work has been done on the lighting performance, especially when natural phenomenon such as fog or haze appears on the street and obstructs the visibility of drivers and pedestrians. To solve such issue, we propose the use of two LED lamps with different correlated color temperature, which will be activated interchangeably according to the condition on the street. We also present the design of communication scheme between each devices in the system. Moreover, our experimental results show the LED Streetlight Platform can perform well and the data can be displayed properly at the website.

• The Journal of the Korea Contents Association
• /
• v.19 no.6
• /
• pp.334-346
• /
• 2019

This study aims to investigate the appearance of Gyeongridangil and its related commercial streets, which seemingly serve as a place brand. Internet trend data, locational conditions, and commercial compositions were analyzed to determine the characteristics of their formation. Result showed that, as the commercial streets developed because of social network services, the street names tended to be considered as place images. In terms of locational conditions, they chose rather remote places, risking the disadvantages for their own spatial style. Interpreted within the framework of place branding, these streets seemed to identify themselves with a unique streetscape using spatial restraints as an advantage. They positioned in a niche area to avoid conflict with other commercial types and attract consumers with the image of "suppliers of trendy lifestyle", thus successfully becoming a place brand.

• International Journal of Computer Science & Network Security
• /
• v.23 no.12
• /
• pp.235-245
• /
• 2023

Smart applications based on the Network of Everything also known as Internet of Everything (IoE) are increasing popularity as network connectivity requires rise further. As a result, there will be a greater need for developing 6G technologies for wireless communications in order to overcome the primary limitations of visible 5G networks. Furthermore, implementing neural networks into 6G will bring remedies for the most complex optimizing networks challenges. Future 6G mobile phone networks must handle huge applications that require data and an increasing amount of users. With a ten-year time skyline from thought to the real world, it is presently time for pondering what 6th era (6G) remote correspondence will be just before 5G application. In this article, we talk about 6G dreams to clear the street for the headway of 6G and then some. We start with the conversation of imaginative 5G organizations and afterward underline the need of exploring 6G. Treating proceeding and impending remote organization improvement in a serious way, we expect 6G to contain three critical components: cell phones super broadband, very The Web of Things (or IoT and falsely clever (artificial intelligence). The 6G project is currently in its early phases, and people everywhere must envision and come up with its conceptualization, realization, implementation, and use cases. To that aim, this article presents an environment for Presented Distributed Artificial Intelligence as-a-Services (DAIaaS) supplying in IoE and 6G applications. The case histories and the DAIaaS architecture have been evaluated in terms of from end to end latency and bandwidth consumption, use of energy, and cost savings, with suggestion to improve efficiency.