• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.05a
• /
• pp.793-795
• /
• 2008

최근 유비쿼터스 컴퓨팅과 유비쿼터스 네트워크를 활용하여 새로운 서비스들을 개발하려는 노력이 진행 중이며, 이에 관련된 기술의 중요성도 급증하고 있다. 특히 감시정찰 센서네트워크의 핵심 구성요소인 저가의 경량 센서노드에서 측정한 미가공 데이터(raw data)를 사용하여 침입 물체의 실시간 탐지, 식별, 추적 및 예측하기 위한 디지털 신호처리 기술은 주요 기술 중 하나이다. 본 논문에서는 감시정찰 센서네트워크의 핵심 구성요소인 자기 센서노드에서 측정한 자기 미가공 데이터를 사용하여 사람과 차량을 탐지할 수 있는 자기 센서 디지털 신호처리 알고리즘을 설계한다. 알고리즘의 주 목표는 감시정찰용 센서노드의 탐지 신뢰성을 높이기 위한 높은 침입물체 탐지 성공률(success rate)과 낮은 허위신고(false alarm) 횟수를 가지는 것이다.

• Korea Multimedia Society
• /
• v.13 no.4
• /
• pp.33-39
• /
• 2009

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2014.01a
• /
• pp.251-252
• /
• 2014

본 논문에서는 차영상 기법을 이용하여 이동하는 객체를 탐지하고 계수하는 시스템을 제안한다. 제안된 시스템은 카메라를 통해 들어온 입력 영상과 배경의 차이를 통해 객체를 탐지하고 객체의 움직임을 분석하여 이동 객체를 계수한다. 실험 결과를 통해 물체의 이동 객체의 탐지 및 계수가 이루어짐을 확인 할 수 있다.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.5
• /
• pp.55-65
• /
• 2016

Detecting an object which is located at seabed is an important issue for various areas. This paper presents an approach to detection of an object that is placed at seabed in the shallow water. A conventional scheme is to employ a side-scan sonar to obtain images of a detection area and to use image processing schemes to recognize an object. Since this approach relies on high frequency signals to get clear images, its detection range becomes shorter and the processing time is getting longer. In this paper, we consider an active sonar system that is repeatedly sending a linear frequency modulated signal of 6~20 kHz in the shallow water of 100m depth. The proposed approach is to model consecutively received reflected signals and to measure their modeling error magnitudes which decide the existence of an object placed on seabed depending on relative magnitude with respect to threshold value. The feature of this approach is to only require an assumption that the seabed consists of an homogeneous sediment, and not to require a prior information on the specific properties of the sediment. We verify the proposed approach in terms of detection probability through computer simulation.

• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.233-247
• /
• 2023

Recently, research on detecting objects in hidden spaces beyond the direct line-of-sight of observers has received attention. Most studies use optical equipment that utilizes the directional of light, but sound that has both diffraction and directional is also suitable for non-line-of-sight(NLOS) research. In this paper, we propose a novel method of detecting objects in non-line-of-sight (NLOS) areas using acoustic signals in the audible frequency range. We developed a deep learning model that extracts information from the NLOS area by inputting only acoustic signals and predicts the properties and location of hidden objects. Additionally, for the training and evaluation of the deep learning model, we collected data by varying the signal transmission and reception location for a total of 11 objects. We show that the deep learning model demonstrates outstanding performance in detecting objects in the NLOS area using acoustic signals. We observed that the performance decreases as the distance between the signal collection location and the reflecting wall, and the performance improves through the combination of signals collected from multiple locations. Finally, we propose the optimal conditions for detecting objects in the NLOS area using acoustic signals.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.1
• /
• pp.25-31
• /
• 2023

In order to estimate the radial speed of an underwater object so-called target with active sonar, Continuous Wave (CW) pulse is generally used, but if a target is slow and at near distance, it is not easy to estimate the radial velocity of the target due to acoustic reverberation in the ocean. In 2017, Wang et al. utilized broadband signal of two Hyperbolic Frequency Modulation (HFM) pulses, which is known as a doppler-invariant pulse, with equal frequency band and in opposite sweep directions to overcome this problem and successfully estimate the radial speed of slow-moving nearby target. They demonstrated the estimation of the radial velocity with computer simulation using the parameters of two HFM starting time differences and receiving times. However, for it uses two HFM pulses with equal frequency, cross-correlation between the two pulses negatively affect the detection performance. To mitigate this cross-correlation effect, we suggest using two different band HFM with the opposite sweep directions. In this paper, a method of radial velocity estimation is derived and simulated using two HFM pulses with the pulse length of 1 second and bandwidth of 400 Hz. Applying the suggested method, the radial velocity was estimated with approximately 6 % of relative error in the simulation.

• Journal of Advanced Navigation Technology
• /
• v.14 no.2
• /
• pp.284-288
• /
• 2010

Video surveillance is widely used in establishing the societal security network. In this paper, intrusion detection based on visual information acquired by static camera is proposed. Proposed approach uses background model constructed by approximated median filter(AMF) to find a foreground candidate, and detected object is calculated by analyzing motion information. Motion detection is determined by the relative size of 2D object in RGB space, finally, the threshold value for detecting object is determined by heuristic method. Experimental results showed that the performance of intrusion detection is better one when the spatio-temporal candidate informations change abruptly.

• Journal of KIISE:Software and Applications
• /
• v.30 no.10
• /
• pp.973-982
• /
• 2003

This paper presents an efficient method for automatically detecting objects in a given image. The GHT is a robust template matching algorithm for automatic object detection in order to find objects of various shapes. Many different templates are applied by the GHT in order to find objects of various shapes and size. Every boundary detected by the GHT scan be used as an initial outline for more precise contour-finding techniques. The main weakness of the GHT is the excessive time and memory requirements. In order to overcome this drawback, the proposed algorithm uses a multiresolution search by scaling down the original image to half-sized and quarter-sized images. Using the information from the first iterative GHT on a quarter-sized image, the range of nuclear sizes is determined to limit the parameter space of the half-sized image. After the second iterative GHT on the half-sized image, nuclei are detected by the fine search and segmented with edge information which helps determine the exact boundary. The experimental results show that this method gives reduction in computation time and memory usage without loss of accuracy.

• Journal of Internet Computing and Services
• /
• v.20 no.4
• /
• pp.39-46
• /
• 2019

In recently years, it has become important to detect underground objects of various marterials including metals, such as detecting the location of sink holes and pipe. For this reason, ground penetrating radar(GPR) technology is attracting attention in the field of underground detection. GPR irradiates the radar wave to find the position of the object buried underground and express the reflected wave from the object as image. However, it is not easy to interpret GPR images because the features reflected from various objects underground are similar to each other in GPR images. Therefore, in order to solve this problem, in this paper, to estimate the piping position in the GRP image according to the threshold value using the CNN (Convolutional Neural Network) model based on deep running, which is widely used in the field of image recognition, As a result of the experiment, it is proved that the pipe position is most reliably detected when the threshold value is 7 or 8.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.555-558
• /
• 2020

본 논문에서는 입력 영상에 담긴 다양한 물체들과 그들 간의 관계를 효과적으로 탐지하여, 하나의 장면 그래프로 표현해내는 새로운 심층 신경망 모델을 제안한다. 제안 모델에서는 물체와 관계의 효과적인 탐지를 위해, 합성 곱 신경망 기반의 시각 맥락 특징들뿐만 아니라 언어 맥락 특징들을 포함하는 다양한 멀티 모달 맥락 정보들을 활용한다. 또한, 제안 모델에서는 관계를 맺는 두 물체 간의 상호 의존성이 그래프 노드 특징값들에 충분히 반영되도록, 그래프 신경망을 이용해 맥락 정보를 임베딩한다. 본 논문에서는 Visual Genome 벤치마크 데이터 집합을 이용한 비교 실험들을 통해, 제안 모델의 효과와 성능을 입증한다.