• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.72-75
• /
• 2020

조립 부품 이미지에 해당하는 3D CAD 모델 매칭 기술은 최근 로봇 조립 기술의 발전으로 필요성이 대두되고 있다. 이미지 기반 3 차원 모델 매칭 연구는 진행되어 왔지만 가구 부품 이미지와는 특성이 다른 RGB[5] 이미지나 스케치 이미지를 다루는[1] 접근들이었다. 딥러닝을 사용하는 스케치 이미지 기반 3 차원 물제 검색 연구에서는 대부분 3 차원 이미지를 다각도에서 렌더링한 view 이미지들에서 feature를 추출하고 pooling 하여 하나의 feature를 출력한다. 그러나 기존의 view pooling 방식은 단순한 평균 방식으로, 부품 이미지에 따른 view를 반영하기에는 한계가 있었다. 따라서 본 논문에서는 조립 부품 이미지 기반 3 차원 물체 검색을 위해 query 부품 이미지에 따라 다른 view 이미지에 집중할 수 있는 방식의 attentional view pooling을 제안한다. 또한 조립 부품 데이터의 특성 상 class 당 CAD 모델이 하나인 상황이므로 학습 데이터가 터무니없이 부족하여 이를 해결하기 위한 학습 데이터 증강 방법을 제안한다. 실험은 의자 부품 11가지에 대해 진행하였고 이를 통해 제안하는 방식의 성능을 입증하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.894-895
• /
• 2023

증강현실(AR)이란, 현실의 이미지나 배경에 3차원 가상 이미지를 입혀 실시간으로 출력되는 컴퓨터 그래픽 기법이다. 본 논문에서는 AR 서비스를 이용한 여행 기록 애플리케이션을 제안한다. 현재 제공되는 서비스에서는 장소에 대해 제공하는 정보가 매우 한정적이다. 따라서 공유 커뮤니티를 형성할 수 있도록 하여 다양한 정보를 전달하고 직접적인 시각 자료를 제공받을 수 있는 애플리케이션이 필요하다고 생각하여 기획하게 되었다. ReactNative와 SpringBoot를 사용하며, MySQL을 이용하여 데이터를 저장한다. Google Map API를 활용하여 사용자가 기록하고자 하는 장소에 대한 위도와 경도 정보를 받을 수 있도록 하며, AR 기능은 Unity로 구현하였다. AR을 포함한 혼합 현실 서비스를 제공하여 사용자들이 본인의 경험에만 의존하는 것이 아닌 풍부한 몰입력과 현장감을 통해 보다 가시적이고 생생한 다이어리 기록을 만들 수 있다.

• Analytical Science and Technology
• /
• v.30 no.6
• /
• pp.303-311
• /
• 2017

A new method for obtaining the photoluminescence of footwear impression by using 1,2-indandione (1,2-IND) solution, which is a latent fingerprint-developing reagent, was studied. A binary complex of DL-alanine and 1,2-IND was prepared by spraying a DL-alanine solution and the 1,2-IND solution (an amino acid sensitive reagent) onto dry or wet origin footwear impression deposited on the surface of printed A4 paper. This binary complex reacts with the trace metal component in the footwear impression to form a ternary complex that exhibits photoluminescence. However, when 5-methylthioninhydrin (5-MTN) solution was used instead of 1,2-IND, no consistent photoluminescence was observed even under identical treatment conditions. In addition, when footwear impressions treated with DL-alanine and 1,2-IND solutions were stored under various temperature conditions (30, 40 and $50^{\circ}C$) and various humidity conditions (30 %, 40 %, 50 % and 60 % RH), the contrast between the footwear impression and the background decreased. Optimal footwear impression photoluminescence was obtained when the footwear impressions treated with DL-alanine and 1,2-IND solutions were stored at $30^{\circ}C$ and 30 % RH for 1 h. The sensitivity of the developed method was ccompared with the sensitivities of three known methods - black gelatin lifting, 2,2'-dipyridyl treatment, and 8-hydroxyquinoline treatment. The results showed that the sensitivity of the developed method was worse than that of the black gelatin lifting method but better than that of 2,2'-dipyridyl or 8-hydroxyquinoline treatment method.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.21 no.12
• /
• pp.1352-1370
• /
• 2010

In this paper, we investigated the shadow area of radio programs which are currently being broadcast via FM 102.5 MHz in Jeollabuk-do consisting of 6 cities and 8-guns. The field strength data of FM 102.5 MHz was used with criterion of shadow area. As a result, the field strength of the eastern mountainous area in Jeollabuk-do was very low. We also measured, compared and analyzed the field strength of 102.5 MHz(TBN-Jeonju), 100.7 MHz(KBS) and 99.1 MHz (MBC) in Jeollabuk-do. Based on the field strength data, we tried to find what correlation exists and how much influence is between difference of antenna output power and solution of shadow area. In result, enhancement of antenna output power is not solution in terms of reducing the shadow Area of 102.5 MHz(TBN-Jeonju). In addition, we measured, compared and analyzed the field strength of 102.5 MHz(TBN-Jeonju) and 106.9 MHz(EBS). In fact, the relay stations with 104.1 MHz, 104.9 MHz, 105.3 MHz and 107.5 MHz of EBS FM has a number of FM relay stations in the eastern mountainous area of Jeollabuk-do. As a result, we confirmed that EBS almost did not have shadow area in the eastern mountainous area of Jeollabuk-do based on the field strength data. As in the case of EBS and KBS, we suggest installing TBN-Jeonju FM relay station with the best solution to reducing the shadow area to extend the audible range of TBN-Jeonju.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.23 no.4
• /
• pp.491-498
• /
• 2012

Recently, many papers have been proposed in order to improve the OFDM system performance in T-DMB DOCR (Digital On Channel Repeater), by using removing the feedback signal so that the transmitter power can be increased or by using the equalizer to remove ICI. Despite these efforts, however, signal quality at the receiving terminal has not been improved because of constellation smearing in T-DMB DOCR. In this paper, in order to suppress constellation smearing, we propose an effective equalizer algorithm that can improve system performance. We perform adaptive channel estimation and non-coherent decision directed noise cancellation method that can estimate the channel subsequently during data symbols period in the frequency domain. So we can obtain better quality of the signal at the receiving terminal. In order to secure QoS(Quality of Service) required in T-DMB handsets, we evaluate SNR and BER in T-DMB DOCR(Digital On Channel Repeater) and verified by simulation. In this simulation results, this system is satisfied the performance of BER=$10^{-5}$ at less than SNR=14 dB at the receiver after compensation of phase noise -18 dBc.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.3
• /
• pp.59-66
• /
• 2021

Humans mainly recognize surrounding objects using visual and auditory information among the five senses (sight, hearing, smell, touch, taste). Major research related to the latest object recognition mainly focuses on analysis using image sensor information. In this paper, after emitting various chirp audio signals into the observation space, collecting echoes through a 2-channel receiving sensor, converting them into spectral images, an object recognition experiment in 3D space was conducted using an image learning algorithm based on deep learning. Through this experiment, the experiment was conducted in a situation where there is noise and echo generated in a general indoor environment, not in the ideal condition of an anechoic room, and the object recognition through echo was able to estimate the position of the object with 83% accuracy. In addition, it was possible to obtain visual information through sound through learning of 3D sound by mapping the inference result to the observation space and the 3D sound spatial signal and outputting it as sound. This means that the use of various echo information along with image information is required for object recognition research, and it is thought that this technology can be used for augmented reality through 3D sound.