• 제목/요약/키워드: Communication Broadcasting Convergence

검색결과 1,416건 처리시간 0.021초

불법 주정차 단속을 위한 딥러닝 기반 이미지 인식 모델 (A Deep Learning-Based Image Recognition Model for Illegal Parking Enforcement)

  • 조민규;김민준;김재환;김진욱;황병선;이승우;선준호;김진영
    • 한국인터넷방송통신학회논문지
    • /
    • 제24권1호
    • /
    • pp.59-64
    • /
    • 2024
  • 최근 다양한 산업 분야에서 드론과 인공지능 기술이 융합된 연구 사례가 진행되고 있다. 본 논문에서는 딥러닝 기반 객체 인식과 객체 판별 알고리즘을 이용하여 불법 주정차 차량 인식 모델을 제안한다. 객체 인식 알고리즘은 YOLOv8를 사용하였으며, 객체 판별 알고리즘은 ResNet18을 사용하였다. 제안된 모델은 일반 도로 상황에서 수집한 이미지 데이터를 이용하여 모델 학습을 수행하였고, 학습된 모델은 이미지 기반 불법 주정차를 판별하는데 높은 정확도를 보였다. 이를 통해 제안된 모델은 다양한 이미지로부터 불법 주정차 차량을 식별하기 위한 일반화 성능을 갖추고 있음을 확인하였다.

Implementation of Tone Control Module in Anchor System for Improved Audio Quality

  • Seungwon Lee;Soonchul Kwon;Seunghyun Lee
    • International Journal of Internet, Broadcasting and Communication
    • /
    • 제16권2호
    • /
    • pp.10-21
    • /
    • 2024
  • Recently, audio systems are changing the configuration of conventional sound reinforcement (SR) systems and public address (PA) systems by using audio over IP (AoIP), a technology that can transmit and receive audio signals based on internet protocol (IP). With the advancement of IP technology, AoIP technologies are leading the audio market and various technologies are being released. In particular, audio networks and control hierarchy over peer-to-peer (Anchor) technology based on AoIP is a system that transmits and receives audio signals over a wide bandwidth without an audio mixer, creating a novel paradigm for existing audio system configurations. Anchor technology forms an audio system by connecting audio sources and output equipment with On-site audio center (OAC), a device that can transmit and receive IP. Anchor's receiving OAC is capable of receiving and mixing audio signals transmitted from different IPs, making it possible to configure a novel audio system by replacing the conventional audio mixer. However, Anchor technology does not have the ability to provide audio effects to input devices such as microphones and instruments in the audio system configuration. Due to this, when individual control of each audio source is required, there is a problem of not being able to control the input signal, and it is impossible to individually affect a specific input signal. In this paper, we implemented a tone control module that can individually control the tone of the audio source of the input device using the audio processor core in the audio system based on Anchor technology, tone control for audio sources is possible through a tone control module connected to the transmitting OAC. As a result of the study, we confirmed that OAC receives the signal from the audio source, adjusts the tone and outputs it on the tone control module. Based on this, it was possible to solve problems that occurred in Anchor technology through transmitting OAC and tone control modules. In the future, we hope that the audio system configuration using Anchor technology will become established as the standard for audio equipment.

Convergence Point Adjustment Methods for Minimizing Visual Discomfort Due to a Stereoscopic Camera

  • Ha, Jong-Soo;Kim, Dae-Woong;Kim, Dong Hyun
    • Journal of information and communication convergence engineering
    • /
    • 제12권4호
    • /
    • pp.246-251
    • /
    • 2014
  • The recent rise of the three-dimensional television (3DTV) industry has led to a wide exploitation of the dual-lens stereoscopic camera. However, when the zoom-in function is used, it is possible that a camera object is magnified only with a fixed convergence point, thereby leading to visual discomfort. In this paper, we propose several methods based on which a convergence point can be adjusted to prevent visual discomfort during zoom-in for a dual-lens stereoscopic camera. Further, we produce 3D contents by applying the proposed methods to certain cases and at certain distances and conduct a subjective evaluation. On the subjective evaluation, 48 subjects watch the 3D contents created by the proposed methods and score the stages of the visual comfort. The result of the subjective evaluation shows that some of the proposed methods are more efficient than the others. We hope that these proposed high-efficiency methods can be applied to produce a dual-lens stereoscopic camera that allows convenient stereoscopic photography.

Comparative Analysis of the Status of Restaurant Start-ups Before and After the Lifting of Social Distancing Through Big Data Analysis

  • Jong-Hyun Park;Yang-Ja Bae;Jun-Ho Park;Gi-Hwan Ryu
    • International journal of advanced smart convergence
    • /
    • 제12권4호
    • /
    • pp.353-360
    • /
    • 2023
  • This paper explores notable shifts in the restaurant startup market following the lifting of social distancing measures. Key trends identified include an escalated interest in startups, a heightened focus on the quality and diversity of food, a relative decline in the importance of delivery services, and a growing interest in specific industry sectors. The study's data collection spanned three years, from April 2021 to May 2023, encompassing the period before and after social distancing. Data were sourced from a range of online platforms, including blogs, news sites, cafes, web documents, and intellectual forums, provided by Naver, Daum, and Google. From this collected data, the top 50 words were identified through a refinement process. The analysis was structured around the social distancing application period, comparing data from April 2021 to April 2022 with data from May 2022 to May 2023. These observed trend changes provide founders with valuable insights to seize new market opportunities and formulate effective startup strategies. In summary, We offer crucial insights for founders, enabling them to comprehend the evolving dynamics in food service startups and to adapt their strategies to the current market environment.

3D Object Generation and Renderer System based on VAE ResNet-GAN

  • Min-Su Yu;Tae-Won Jung;GyoungHyun Kim;Soonchul Kwon;Kye-Dong Jung
    • International journal of advanced smart convergence
    • /
    • 제12권4호
    • /
    • pp.142-146
    • /
    • 2023
  • We present a method for generating 3D structures and rendering objects by combining VAE (Variational Autoencoder) and GAN (Generative Adversarial Network). This approach focuses on generating and rendering 3D models with improved quality using residual learning as the learning method for the encoder. We deep stack the encoder layers to accurately reflect the features of the image and apply residual blocks to solve the problems of deep layers to improve the encoder performance. This solves the problems of gradient vanishing and exploding, which are problems when constructing a deep neural network, and creates a 3D model of improved quality. To accurately extract image features, we construct deep layers of the encoder model and apply the residual function to learning to model with more detailed information. The generated model has more detailed voxels for more accurate representation, is rendered by adding materials and lighting, and is finally converted into a mesh model. 3D models have excellent visual quality and accuracy, making them useful in various fields such as virtual reality, game development, and metaverse.

하이브리드 방송 환경 하에서 HTML5 기반 장면구성 기술 (Scene Composition Technology Based on HTML5 in Hybrid Broadcasting Environment)

  • 조민우;박정욱;김규헌
    • 방송공학회논문지
    • /
    • 제18권2호
    • /
    • pp.237-248
    • /
    • 2013
  • 하이브리드 방송(Hybrid Broadcasting) 환경은 다수의 전송경로를 통한 방송 통신 융합 환경이다. 이러한 하이브리드 방송 환경은 기존 방송망의 대역폭 제한으로 소수의 미디어만을 전달하던 것과 달리, 타 통신 전송망을 동시에 활용함으로써 다수의 미디어를 전달할 수 있다. 현재, 스마트TV를 필두로 방송망과 IP망을 결합한 하이브리드 방송 환경이 조성되어 다양한 서비스가 나타나고 있으며, 스마트폰 및 태블릿PC와 같은 타 스마트기기에서 또한 머지않아 하이브리드 방송 환경을 통한 서비스가 나타날 것으로 전망된다. 이렇게 다수의 미디어를 동시에 소비할 수 있는 하이브리드 방송 환경을 효과적으로 활용하는 방법의 하나는 장면구성을 이용하는 것이다. 장면구성은 미디어가 소비되는 시간과 화면상의 공간을 특정함으로써 다수의 미디어를 복합적으로 소비하는 방법이다. 이에 본 논문에서는 하이브리드 방송 환경에 적합하고, 스마트기기에 쉽게 적용될 수 있도록 HTML5를 기반으로 한 장면구성 기술을 제안한다. 하지만 HTML5의 스크립트 언어 및 스타일 언어를 활용한 미디어의 공간구성 및 시간구성은 처리의 복잡성을 높이고, 사용 가능한 단말기의 제한을 줄 수 있다. 또한, 하나의 HTML5 문서는 하나의 장면만 표현할 수 있다. 따라서 제안하는 장면구성 기술은 마크업 언어를 통해 미디어의 공간구성 및 시간구성과 다수 장면을 표현할 수 있도록 HTML5를 확장한다. 더불어 하이브리드 방송환경에 적용을 위한 확장을 포함한다. 이를 위해 HTML5 기술 및 제안하는 장면구성 기술을 소개하고, 구현 및 실험을 통해 본 제안을 검증한다.

Thermal Characteristics of the Optimal Design on 20W COB LED Down Light Heat Sink

  • Kwon, Jae-Hyun;Lee, Jun-Myung;Huang, Wei;Park, Keon-Jun;Kim, Yong-Kab
    • International journal of advanced smart convergence
    • /
    • 제2권2호
    • /
    • pp.19-22
    • /
    • 2013
  • As the demand of the LED for lighting that emits light by p-n junction is increasing, studies on heatproof plate technology is being conducted to minimize the temperature of the LED lighting. As for the temperature of the LED devices, their light emitting efficiency decreases and the maximum lifespan drops down to 1/5. Therefore there are heat dissipation studies going on to minimize the heat. For LED heat dissipation, aluminum heat sink plates are mostly used. For this paper, we designed heat sink that fits residential 20W COB LED Down Light; packaged the heat sink and 20W COB and analyzed and evaluated the thermal properties through a Solidworks flow simulation. We are planning to design the optimal heat sink plate to solve the thermal agglomeration considering TIM(Thermal Interface material).

A Prediction of Nutrition Water for Strawberry Production using Linear Regression

  • Venkatesan, Saravanakumar;Sathishkumar, VE;Park, Jangwoo;Shin, Changsun;Cho, Yongyun
    • International journal of advanced smart convergence
    • /
    • 제9권1호
    • /
    • pp.132-140
    • /
    • 2020
  • It is very important to use appropriate nutrition water for crop growth in hydroponic farming facilities. However, in many cases, the supply of nutrition water is not designed with a precise plan, but is performed in a conventional manner. We proposes a forecasting technique for nutrition water requirements based on a data analysis for optimal strawberry production. To do this, the proposed forecasting technique uses linear regression for correlating strawberry production, soil condition, and environmental parameters with nutrition water demand for the actual two-stage strawberry production soil. Also, it includes predicting the optimal amount of nutrition water requires according to the heterogeneous cultivation environment and variety by comparing the amount of nutrition water needed for the growth and production of different kinds of strawberries. We suggested study uses two types of section beds that are compared to find out the best section bed production of strawberry growth. The dataset includes 233 samples collected from a real strawberry greenhouse, and the four predicted variables consist of the total amounts of nutrition water, average temperature, humidity, and CO2 in the greenhouse.

16-QAM 신호에서 Slice 가중치에 의한 S-MMA 적응 등화 알고리즘의 성능 비교 (Performance Comparison of S-MMA Adaptive Equalization Algorithm by Slice Weighting Value in 16-QAM Signal)

  • 임승각
    • 한국인터넷방송통신학회논문지
    • /
    • 제13권3호
    • /
    • pp.55-61
    • /
    • 2013
  • 본 논문에서는 통신 채널에서 발생되는 찌그러짐과 잡음의 영향을 최소화하기 위하여 사용되는 S-MMA 적응 등화 알고리즘에서 Slice 가중치에 따른 성능을 비교하였다. 기존의 MMA 알고리즘에서는 등화기의 출력 신호와 송신 신호의 dispersion 상수만을 이용하지만, S-MMA에서는 등화기 출력 신호와 dispersion 상수외에 결정 장치의 출력 신호를 slice 상수만큼 고려하여 채널의 진폭과 위상 찌그러짐을 동시에 보상할 수 있다. 이때 slice 가중치가 적응 등화 알고리즘의 성능에 미치는 영향을 컴퓨터 시뮬레이션을 통해 확인하며, 성능 지수로는 등화기 출력 신호 성상도, 수렴 특성을 나타내는 잔류 isi, 최대 찌그러짐, MSE와 채널의 신호대 잡음비에 따른 SER을 사용하였다. 시뮬레이션 결과 slice 가중치가 적으면 잔류isi, 최대 찌그러짐과 MSE 성능이 우월하며, 가중치가 큰 경우 SER 성능이 우월함을 확인하였다.

Multilevel QAM 신호 전송에서 RMMA와 SCA 적응 등화 알고리즘의 성능 비교 (A Performance Comparison of RMMA and SCA Adaptive Equalization Algorithm in Multilevel QAM Signal Transmission)

  • 임승각
    • 한국인터넷방송통신학회논문지
    • /
    • 제18권2호
    • /
    • pp.111-116
    • /
    • 2018
  • 본 논문은 Multilevel QAM 신호의 전송시 채널에서 발생되는 부호간 간섭을 최소화시키기 위한 RMMA (Region-based MMA)와 SCA (Square Contour Algorithm) 적응 등화 알고리즘의 성능 비교에 관한 것이다. RMMA는 Multilevel QAM 신호의 등화시 기존 MMA 알고리즘에서 안정성과 4-level의 constant modulus로 변환하여 성능을 개선키 위하여 사용되며, SCA는 기존 CMA와 RCA을 조합하여 등화 성능을 개선키 위하여 사용된다. 이들은 서로 상이한 원리를 적용하여 등화 성능을 개선하고 있으므로, 논문에서는 이들의 성능을 동일한 채널 환경에 적용하여 비교하였다. 이를 위한 성능 지수로는 등화기 출력 성상도, 잔류 isi, 최대 찌그러짐을 적용하였다. 컴퓨터 시뮬레이션 결과 모든 성능에서는 RMMA가 SCA보다 신호점 군집 능력과 수렴 속도에서 우월하며, misadjustment에 의한 등화 잡음이 감소함을 알 수 있었다.