• International Journal of Internet, Broadcasting and Communication
• /
• v.16 no.3
• /
• pp.212-220
• /
• 2024

A recently the advancement of society, AI technology has made significant strides, especially in the fields of computer vision and voice recognition. This study introduces a system that leverages these technologies to recognize users through a camera and relay commands within a vehicle based on voice commands. The system uses the YOLO (You Only Look Once) machine learning algorithm, widely used for object and entity recognition, to identify specific users. For voice command recognition, a machine learning model based on spectrogram voice analysis is employed to identify specific commands. This design aims to enhance security and convenience by preventing unauthorized access to vehicles and IoT devices by anyone other than registered users. We converts camera input data into YOLO system inputs to determine if it is a person, Additionally, it collects voice data through a microphone embedded in the device or computer, converting it into time-domain spectrogram data to be used as input for the voice recognition machine learning system. The input camera image data and voice data undergo inference tasks through pre-trained models, enabling the recognition of simple commands within a limited space based on the inference results. This study demonstrates the feasibility of constructing a device management system within a confined space that enhances security and user convenience through a simple real-time system model. Finally our work aims to provide practical solutions in various application fields, such as smart homes and autonomous vehicles.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.47 no.3
• /
• pp.212-219
• /
• 2019

As unmanned aerial vehicle (UAV) has been utilized for military operation, efficient ways for controlling UAV has been necessary. In particular, instead of conventional approach using console control, speech recognition based UAV control is essential for military environments in which rapid command operation is required. But research on this novel approach is not actively studied yet. In this study, we introduce efficient ways of speech recognition system operation for voice-driven UAV control, focusing on mission command control from manned aircraft rather than ground control center. We propose an efficient way of system operation for UAV control in cooperation of aircraft and UAV, and verify its efficiency via speech recognition experiment.

• Journal of the Architectural Institute of Korea Planning & Design
• /
• v.34 no.3
• /
• pp.3-10
• /
• 2018

The purpose of this study is to suggest the direction and recognition for applying to component of Emotion of the elementary school space with characteristics of child development. For the accomplishment of the study is to deduce types of emotional component and characteristics of child development based on literature and advanced research related to 'Child development and behavior', 'The elementary school space', and concept of 'children' and 'emotion'. In addition, The level of recognition of teachers and students about creation plan of school space by types of emotion component and preference and relationships of students on emotion component of elementary school space is investigated. The space environment has great influence in childhood going through big changes in physical, cognitive, emotional and social ways, Providing space environment built with emotion component such as 'affordance', 'diversity', 'territoriality', and 'relationships' considering characteristics of child development is most important of all, In particular, when building indoor space in elementary schools where students going through various development stages live, providing friendly environments for emotion of children put top priority on students in the decision-making process and guaranteed the participation of students is expected.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.9
• /
• pp.2289-2298
• /
• 1997

This paper proposes a new method to recognize map. In order to minimize the ripple effect of one recognition result affecting another, the structural information is represented with a hierarchical model. and the model is used in both the recognition and verification process. Furthermore, lines related to an entity are searched in a used in both the recognition and verification process. Furthermore, lines related to an entity are searched in a reduced search space by defining some relations between lines. When there is a mis-recognition after verificaiton, recognition process will be retired. In the process, the accurate result can obtained through the change of the parameter values used in the algorithm. As a result, the search space is reduced effectively, and even objects that embodies the broken lines and the crossed lines are recognized.

• Korean Institute of Interior Design Journal
• /
• v.20 no.2
• /
• pp.64-71
• /
• 2011

There are a method to analyze reactions or psychological condition and a method to observe visitor's behavioral reaction as methods to measure and evaluate humans' recognition behavior reaction of humans. The measure of the eye movement as a method to living body's reaction and psychological condition has an advantage to measure the information acceptance reaction of view recognition of the stimulus of view composition factors which has been used since a long time ago in other research areas, but almost not studies have been made on the exhibition views in museums. Therefore, on the premise of such recognition, this study aimed at obtaining various types of information through vision angles of visitor in exhibition space of an museum and judging space information, at measuring the condition of information acceptance through attention experiments and observation investigation of and finding out the disposition and characteristics so as to verify the relationship between the exhibition space and exhibition Method.

• Journal of the Korean housing association
• /
• v.25 no.5
• /
• pp.63-71
• /
• 2014

Recently, many researches on the supply of public rental-housings that utilize modular houses have been conducted for the expanded supply of public rental house and consumer-customized supply. For this reason, a research on recognition and satisfaction of residents who live in modular houses was conducted based on their evaluations on public rental houses utilizing actual modular houses. Consequently planning guidelines that would increase residency satisfaction was drawn. Surveys and interviews of 'The Gongneung Hee-mang Housing' were conducted, and analyses of descriptive statistics, multiple regression analysis and importance-satisfaction analysis were conducted. The post occupancy evaluations inform that to gain extra recognition by promoting advantages of modular houses and engaging in educational activities are needed. Dealing with issues such as external noises and vibrations, proper common facility, interior space plan, taking account of facility plan is also needed. This study would contribute to establish modular house planning, reflecting the residents' practical demands.

• Journal of Information Processing Systems
• /
• v.14 no.6
• /
• pp.1464-1479
• /
• 2018

In face recognition, sometimes the number of available training samples for single category is insufficient. Therefore, the performances of models trained by convolutional neural network are not ideal. The small sample face recognition algorithm based on novel Siamese network is proposed in this paper, which doesn't need rich samples for training. The algorithm designs and realizes a new Siamese network model, SiameseFacel, which uses pairs of face images as inputs and maps them to target space so that the $L_2$ norm distance in target space can represent the semantic distance in input space. The mapping is represented by the neural network in supervised learning. Moreover, a more lightweight Siamese network model, SiameseFace2, is designed to reduce the network parameters without losing accuracy. We also present a new method to generate training data and expand the number of training samples for single category in AR and labeled faces in the wild (LFW) datasets, which improves the recognition accuracy of the models. Four loss functions are adopted to carry out experiments on AR and LFW datasets. The results show that the contrastive loss function combined with new Siamese network model in this paper can effectively improve the accuracy of face recognition.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.2
• /
• pp.143-149
• /
• 2019

Floating holograms are technologies that provide rich 3D stereoscopic images in a wide space such as advertisement, concert. In addition, It is possible to reduce the 3D glasses inconvenience, eye strain, and space distortion, and to enjoy 3D images with excellent realism and existence. Therefore, this paper implements a user gesture recognition system for manipulating a floating hologram characters that can be used in a small space devices. The proposed method detects face region using haar feature-based cascade classifier, and recognizes the user gestures using a user gesture-occurred position information that is acquired from the gesture difference image in real time. And Each classified gesture information is mapped to the character motion in floating hologram for manipulating a character action. In order to evaluate the performance of the proposed user gesture recognition system for manipulating a floating hologram character, we make the floating hologram display devise, and measures the recognition rate of each gesture repeatedly that includes body shaking, walking, hand shaking, and jumping. As a results, the average recognition rate was 88%.

• Journal of Internet Computing and Services
• /
• v.19 no.3
• /
• pp.43-48
• /
• 2018

In this paper, we propose an indoor-space recognition using DNN and super-pixel. In order to recognize the indoor space from the image, segmentation process is required for dividing an image Super-pixel is performed algorithm which can be divided into appropriate sizes. In order to recognize each segment, features are extracted using a proposed method. Extracted features are learned using DNN, and each segment is recognized using the DNN model. Experimental results show the performance comparison between the proposed method and existing algorithms.

• Journal of the Korean Institute of Educational Facilities
• /
• v.3 no.2
• /
• pp.25-35
• /
• 1996

This study is to find countermeasures of facility space according to the development of new educational program and the opening of education market from the renewal of existing facilities. The renewal is consider as the worldwide trend for enhancing added value of the facilities in economic side of educational facilities as well as educational contents. This study selects six from 94 elementary schools in Kwangju-city to examine our educational conditions, understand using conditions and recognition of existing space and possibility of the recycle. The hypothesis that recognition of existing facility space in school may be varied according to physical conditions of children is made in four sides, six elements and two trends are extracted and then life pattern, size, sense of materials and cognition of three types of passage space are analyzed. Accordingly, this study obtains such results that passage space is life space having strong place concept and there is the possibility of renewal of existing school building facilities corresponding to multipurpose of school facilities.