A Novel Method for Improvement of Optical Space Surveillance Systems
조태환 Taehwan Cho , 이성섭 Soungsub Lee
25(2) 125-131, 2021
조태환 Taehwan Cho , 이성섭 Soungsub Lee
DOI: JANT Vol.25(No.2) 125-131, 2021
The optical space surveillance systems are used for image capture and location tracking of space objects including satellites. Such an optical space surveillance systems have the advantage of capable of capturing high-level satellite images, but have a disadvantage in that its location tracking performance is lower than that of laser or radar space surveillance systems. Therefore, in this paper, we proposed a method to improve the location tracking performance of the satellite by reducing the location tracking error of the optical space surveillance systems. In this method, two Kalman filters were used to model the constant velocity circular motion and the acceleration motion, and the performance was improved by applying them to the data of the optical space surveillance systems. The proposed method in this paper shows an average of 17.34% improvement in performance.
|
Air Traffic Controller’s Acceptance and Use of New Technology from ASBU using Technology Acceptance Model and Theory of Planned Behavior
이진숙 Jin-sook Lee , 김기웅 Kee-woong Kim , 박성식 Sung-sik Park
25(2) 132-142, 2021
이진숙 Jin-sook Lee , 김기웅 Kee-woong Kim , 박성식 Sung-sik Park
DOI: JANT Vol.25(No.2) 132-142, 2021
This research has analyzed the air traffic controllers’ perception on ASBU (aviation system block upgrade) and their acceptance and intention to use of new technologies. Based on the integrated model of technology acceptance model and theory of planned behavior, it was empirically analyzed of air traffic controllers’ acceptance and behavioral intention to use. According to the analysis, it was proven the usefulness and self-efficacy of air traffic controllers have a significant positive effect on their attitude to use and intention to use. From the usefulness perspective, it was statistically proven air traffic controllers have both acceptance and intention to use new technologies. However, from the ease to use perspective, it shows ease to use has nothing to do with an attitude to use. That is, air traffic controllers perceived ASBU is useful technologies but not easy to use them.
|
Analysis on the Integrity Monitoring Methods of High Accuracy Service of Global Navigation Satellite Systems
손표웅 Pyo-woong Son , 박슬기 Sul Gee Park , 박상현 Sang Hyun Park , 신유진 Yujin Shin , 송재영 Jaeyoung Song , 김의호 Euiho Kim
25(2) 143-149, 2021
손표웅 Pyo-woong Son , 박슬기 Sul Gee Park , 박상현 Sang Hyun Park , 신유진 Yujin Shin , 송재영 Jaeyoung Song , 김의호 Euiho Kim
DOI: JANT Vol.25(No.2) 143-149, 2021
International Maritime Organization (IMO) recommends maintaining a positioning accuracy of global satellite navigation system (GNSS) within the range of 10 cm and monitoring its integrity. In order to provide the positioning accuracy of 10 cm or better, high-precision correction information must be used based on the carrier phase of the satellite signal. Furthermore, in order to monitor the integrity of this, the errors should be overbounded in the process of generating correction. In this paper, we analyze the GNSS integrity monitoring methods in high-precision position service currently in operation. The integrity monitoring technique of the wide area augmentation system (WAAS) of the U.S. is studied. As a conclusion, we propose the integrity monitoring method of Korean high-precision GNSS service in maritime area.
|
One-dimensional Object Tracking using LSTM Neural Network
박선배 Sun-bae Park , 유도식 Do-sik Yoo
25(2) 150-155, 2021
박선배 Sun-bae Park , 유도식 Do-sik Yoo
DOI: JANT Vol.25(No.2) 150-155, 2021
Object tracking is a technique of signal processing that estimates objects locations based on past locations and present time observed data. While, Kalman filter and particle filter are among the most notable object tracking schemes, these filters need to know the system model to achieve optimal performance. The recursive neural network (RNN) with a feedback loop added to the perceptron neural network can be used for object tracking. Also, RNN evolved into long-short term memory (LSTM) that solved the long-term dependence problem and is being used in various fields. In this paper, in order to study the tracking performance of LSTM, we consider a simple problem of one-dimensional object tracking, and compare the tracking performance with Kalman and particle filters. In order to test the tracking performance in diverse observation environments, various noise models such as Gaussian, Laplace, exponential, and uniformly distributed noises are considered. Under the various circumstances, we observe that LSTM neural network achieves fairly stable performance without knowing the system model.
|
Code Embedding and Detection for Camera Identifiable Photos
배준현 Joon-hyeon Bae , 홍원기 Won-kee Hong , 고재필 Jaepil Ko
25(2) 156-161, 2021
배준현 Joon-hyeon Bae , 홍원기 Won-kee Hong , 고재필 Jaepil Ko
DOI: JANT Vol.25(No.2) 156-161, 2021
Recently, applications that acquire additional information by taking a quick response (QR) code photo with a smart-phone camera is widely used. The QR codes can contain a lot of information, but in order to recognize a QR code, we need to take a photo so that all four corners of the code are covered. To overcome this limitation, we propose a method to embed a sequence of code bits to a digital image and detect from the captured image of its photos. The proposed scheme embed the code in the frequency domain so that the code is not lost despite severe image degradation issued during the print-camera process. The conventional frequency-domain approaches usually embed only 1~2 bits, however we designed to be able to contain 16 bits. In particular, we additionally embed alignment bits as well as code bits to facilitate the decoding process. In our experiments, we achieved an average accuracy of 95.2% for images partially captured under random rotation. This corresponds to only 0.76 bits in terms of the average number of error bits.
|
Implementing Parking Zone Management System for Disabled based on Deep Learning using Cloud Platform
황주훈 Ju-hoon Hwang , 김창복 Chang-bok Kim
25(2) 162-168, 2021
황주훈 Ju-hoon Hwang , 김창복 Chang-bok Kim
DOI: JANT Vol.25(No.2) 162-168, 2021
This study proposed a system that manages parking spaces for the disabled, this will lead to the promotion of welfare for those who are disabled by using deep learning and cloud platforms. Deep learning used you only look once (YOLO) for license plate detection concerning car images in parking areas, and convolutional neural network (CNN) was used for license plate character recognition from extracted numbers and text images. This system can be managed in real time, and it has been simplified so that it can be managed only with video. In addition, it is recognized and accurate by increasing the recognition rate of Korean characters compared to the existing optical character recognition (OCR), and it has the advantage of scalability in the management area by enabling parking management but only if closed circuit television (CCTV) is installed. This system requires a study to increase the accurate license plate recognition rate. This is an important factor, and a continuous study on the processing speed problem to execute YOLO and CNN algorithms in a somewhat low performance raspberry environment.
|
B777 Hard Landing Trend Analysis based on Quick Access Recorder (QAR) Data
정승섭 Chung Seung Sup , 김현덕 Kim Hyeon Deok
25(2) 169-176, 2021
정승섭 Chung Seung Sup , 김현덕 Kim Hyeon Deok
DOI: JANT Vol.25(No.2) 169-176, 2021
According to the International Air Transport Association (IATA) 2019 Safety Report, one accident involving hard landing has caused 41 deaths. The accident was analyzed to be caused by the pilots’ final judgement of making a forced the landing in unstabilized conditions. The factors leading to hard landings are unstabilized approach, misjudged flare, inappropriate planning, unexpected change in weather etc. This research aimed to look into detail the characteristics of hard landings by utilizing the quick access recorder (QAR) data of 24 recorded hard landing incidents of B777, such as descent rate, flight path angle, threshold passing altitude, initial flare altitude etc. The main causual factors were derived to be low vertical path and late flare. In order to promote proactive an precautionary measures, stabilized approach is emphasized as well as the continuous monitor of flight path angle below 200 ft to maintain proper vertical path.
|