• 한국지능시스템학회논문지
• /
• 제19권3호
• /
• pp.317-322
• /
• 2009

로봇은 일정 속도 이상의 움직임을 가지면 필연적으로 진동을 하기 때문에 장착된 카메라의 경우 이미지가 흔들려 더 이상 로봇의 눈 역할을 수행하지 못하게 된다. 따라서 흔들리는 영상의 안정화를 위한 방법을 연구가 필요해졌다. 영상에서 글로벌 모션 벡터를 계산하여 안정화하는 방법이 존재하지만 이는 프로세서가 처리해야 하는 데이터양이 많아지기 때문에 임베디드 로봇의 사양의 한계로 인하여 실시간으로 영상을 송출하는데 큰 어려움이 있다. 이를 보완하기 위하여 모션벡터를 사용하지 않는 모션센서를 통한 영상 안정화를 제안한다. 모션센서를 통하여 로봇의 진행과 관계없는 움직임을 추출하고 추출한 움직임을 영상에서 제거하는 방법이다.

• 한국전자통신학회논문지
• /
• 제10권12호
• /
• pp.1329-1336
• /
• 2015

저전력 CMOS 카메라 기술의 발전으로 농업용 모니터링, 자연환경 감시 등의 다양한 비디오 센서네트워크 응용들에 대한 연구가 활발히 진행되고 있다. 이러한 응용들에서 핵심 기술은 영상을 어떻게 압축하고 전송할 것인가에 대한 것이다. 일반 센서 데이터에 비해 영상 데이터는 양이 크기 때문에 특히 트래픽에 대한 정확한 예측이 이루어져야만 광범위한 네트워크 자원을 효과적으로 관리할 수 있다. 본 논문에서는 비디오 센서 네트워크 환경에서 비디오 트래픽을 정확하게 예측하는 방법을 제안한다. 제안하는 방법은 영상의 복잡도를 측정하고 이 값을 적응적으로 트래픽 예측에 적용함으로써 기존의 방법들 보다 정확하게 압축 영상의 트래픽 양을 예측할 수 있다. 실험결과는 적응적 복잡도 예측을 이용한 제안하는 방법이 기존 방법에 비해 12% 이상 정확하게 결과 비트량을 예측하는 것을 보여준다.

• 대한임베디드공학회논문지
• /
• 제6권2호
• /
• pp.68-75
• /
• 2011

This paper describes the development of an open source motion sensor emulator. It helps developers to understand the motion sensor and its data better. Through this emulator, the realtime or stored motion sensor data can be applied to the applications that utilize motion sensors. The data of motion sensors which include accelerometer sensor, magnetic field sensor, gyro sensor, GPS, and so on, can be collected by using smart phones or motion sensors. We also describe a visualizer which shows various graphs, video and 3D animations based on the data sent by the emulator. It helps developers to understand motion sensors and how to use the sensors. The developed emulator is compatible with Android sensor simulator.

• IEIE Transactions on Smart Processing and Computing
• /
• 제4권2호
• /
• pp.110-114
• /
• 2015

This paper reports a novel object-detection technique in video sequences. The proposed algorithm consists of detection of objects in illuminated and occluded videos by using object features and a neural network technique. It consists of two functional modules: region-based object feature extraction and continuous detection of objects in video sequences with region features. This scheme is proposed as an enhancement of the Lowe's scale-invariant feature transform (SIFT) object detection method. This technique solved the high computation time problem of feature generation in the SIFT method. The improvement is achieved by region-based feature classification in the objects to be detected; optimal neural network-based feature reduction is presented in order to reduce the object region feature dataset with winner pixel estimation between the video frames of the video sequence. Simulation results show that the proposed scheme achieves better overall performance than other object detection techniques, and region-based feature detection is faster in comparison to other recent techniques.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권4호
• /
• pp.1146-1165
• /
• 2022

A huge amount of data in the form of videos and images is being produced owning to advancements in sensor technology. Use of low performance commodity hardware coupled with resource heavy image processing and analyzing approaches to infer and extract actionable insights from this data poses a bottleneck for timely decision making. Current approach of GPU assisted and cloud-based architecture video analysis techniques give significant performance gain, but its usage is constrained by financial considerations and extremely complex architecture level details. In this paper we propose a data pipeline system that uses open-source tools such as Apache Spark, Kafka and OpenCV running over commodity hardware for video stream processing and image processing in a distributed environment. Experimental results show that our proposed approach eliminates the need of GPU based hardware and cloud computing infrastructure to achieve efficient video steam processing for face detection with increased throughput, scalability and better performance.

• IEIE Transactions on Smart Processing and Computing
• /
• 제4권6호
• /
• pp.379-383
• /
• 2015

This paper presents an approach to detect object distance with the use of the recently developed low-cost Kinect sensor. The technique is based on Kinect color depth-image processing and can be used to design various computer-vision applications, such as object recognition, video surveillance, and autonomous path finding. The proposed technique uses keypoint feature detection in the Kinect depth image and advantages of depth pixels to directly obtain the feature distance in the depth images. This highly reduces the computational overhead and obtains the pixel distance in the Kinect captured images.

• 대한임베디드공학회논문지
• /
• 제19권1호
• /
• pp.19-25
• /
• 2024

This paper proposes an artificial intelligence-based personal information area object detection optimization method in an embedded system to de-identify personal information in video data. As an object detection optimization method, first, in order to increase the detection rate for personal information areas when detecting objects, a gyro sensor is used to collect the shooting angle of the image data when acquiring the image, and the image data is converted into a horizontal image through the collected shooting angle. Based on this, each learning model was created according to changes in the size of the image resolution of the learning data and changes in the learning method of the learning engine, and the effectiveness of the optimal learning model was selected and evaluated through an experimental method. As a de-identification method, a shuffling-based masking method was used, and double-key-based encryption of the masking information was used to prevent restoration by others. In order to reuse the original image, the original image could be restored through a security key. Through this, we were able to secure security for high personal information areas and improve usability through original image restoration. The research results of this paper are expected to contribute to industrial use of data without personal information leakage and to reducing the cost of personal information protection in industrial fields using video through de-identification of personal information areas included in video data.

• Journal of information and communication convergence engineering
• /
• 제7권4호
• /
• pp.507-515
• /
• 2009

In this paper, we propose an image signal processor (ISP) for an ultra small HD-grade video sensor with 3A (AWB, AE, and AF) in camera phones that can process 720P/30fps videos. In order to enhance the video quality of the systems, it is necessary to achieve the high performance of the 3A. The proposed AWB algorithm multiplies the adjusted coefficients of color gains to the captured data of white objects. The proposed AE method adopts the index step moving based on the difference between an averaged Y luminance and a target luminance, together with IIR filters with variable time responses. The proposed AF technique controls the focus curve to find the lens position that maximizes the integrated high frequency components in luminance values by using highpass filters. Finally, we compare the image quality captured from our system to the quality of a commercial HD camcorder in order to evaluate the performance of the proposed ISP. The proposed ISP system is also fabricated with 0.18um CMOS flash memory process.

• 한국군사과학기술학회지
• /
• 제15권6호
• /
• pp.761-769
• /
• 2012

In this paper, we propose a low-power design strategy to minimize energy-consumption for surveillance and reconnaissance sensor networks. The sensor network consists of many different nodes with various operations such as target detection, packet relay, video monitoring, changing protocols, and etc. Each sensor node consists of sensing, computing, communication, and power components. These components are integrated on a single or multiple boards. Therefore, the power consumption of each component can be different on various operation types. First, we identified the list of components and measured power consumption for them from the first prototype nodes. Next, we focus on which components are the main sources of energy consumption. We propose many energy-efficient approaches to reduce energy consumption for each operation type.

• 한국주거학회:학술대회논문집
• /
• 한국주거학회 2008년 추계학술발표대회 논문집
• /
• pp.407-412
• /
• 2008

The purpose of this study is to find out the preference and the opinion of feminine seniors on the life support appliances(video phone, medication dispenser, activity monitor and sleep monitor). The appliances may be improved reflecting the result of the preferred functions and designs. The respondents were 141 feminine seniors of 65 years & older, living in Ulsan city or Cheongju city, Korea. Following is the result. 1) All the responses answered that the medication dispenser, video phone and activity monitor, sleep monitor are useful appliances. 2) All the seniors expressed prefer to function of the appliances, and were highly concerned about the price, safety and convenience. 3)The preferred designs were (1) white or red standing medication dispenser,(2) wall-mounted video phone working by voice, (3) metal activity monitor sensor like white or yellow bracelet.