• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.39 no.2
• /
• pp.235-240
• /
• 2015

Active Safety system is requested on the market of the medium and heavy duty commercial vehicle over 4.5ton beside the market of passenger car with advancement of the digital equipment proportionally. Unlike the passenger car, the mounting position of camera in case of the medium and heavy duty commercial vehicle is relatively high, it is disadvantaged conditions for lane recognition in contradiction to passenger car. In this work, we show the method of lane recognition through the Sobel edge, based on the spatial domain processing, Hough transform and color conversion correction. Also we suggest the low error method of front vehicles recognition in order to reduce the detection error through Haar-like, Adaboost, SVM and Template matching, etc., which are the object recognition methods by frontal camera vision. It is verified that the reliability over 98% on lane recognition is obtained through the vehicle test.

• Journal of Advanced Navigation Technology
• /
• v.18 no.1
• /
• pp.14-21
• /
• 2014

In this paper, we devise a new ontology-based context-aware system to recognize the smart car information, in which driver's intent is utilized by information of car, driver, environment as well as driving state, driver state. So proposed system can handle dynamically risk changes by adding real-time situational awareness information. We utilize the camera image recognition technology for context-aware intelligent vehicle driving information, and implement information acquisition scheme OBD-II protocol to acquire vehicle's information. Experiments confirm that the proposed advanced driver safety assist system outperforms the conventional system, which only utilizes the information of vehicle, driver, and environmental information, to support the service of a high-speed driving, lane-departure service and emergency braking situation awareness.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.1
• /
• pp.6-11
• /
• 2011

In this paper, we propose a gait-based human identification system using eigenfeature regularization and extraction (ERE). First, a gait feature for human identification which is called gait energy image (GEI) is generated from walking sequences acquired from a camera sensor. In training phase, regularized transformation matrix is obtained by applying ERE to the gallery GEI dataset, and the gallery GEI dataset is projected onto the eigenspace to obtain galley features. In testing phase, the probe GEI dataset is projected onto the eigenspace created in training phase and determine the identity by using a nearest neighbor classifier. Experiments are carried out on the CASIA gait dataset A to evaluate the performance of the proposed system. Experimental results show that the proposed system is better than previous works in terms of correct classification rate.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.3
• /
• pp.164-173
• /
• 2014

In underwater robotics, vision would be a key element for recognition in underwater environments. However, due to turbidity an underwater optical camera is rarely available. An underwater imaging sonar, as an alternative, delivers low quality sonar images which are not stable and accurate enough to find out natural objects by image processing. For this, artificial landmarks based on the characteristics of ultrasonic waves and their recognition method by a shape matrix transformation were proposed and were proven in Part 1. But, this is not working properly in undulating and dynamically noisy sea-bottom. To solve this, we propose a framework providing a selection phase of likelihood candidates, a selection phase for final candidates, recognition phase and tracking phase in sequence images, where a particle filter based selection mechanism to eliminate fake candidates and a mean shift based tracking algorithm are also proposed. All 4 steps are running in parallel and real-time processing. The proposed framework is flexible to add and to modify internal algorithms. A pool test and sea trial are carried out to prove the performance, and detail analysis of experimental results are done. Information is obtained from tracking phase such as relative distance, bearing will be expected to be used for control and navigation of underwater robots.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.2072-2072
• /
• 2009

최근 연구들에서, 증강 현실(Augmented Reality; AR) 환경에서의 촉각 상호작용에 대한 가능성이 논의되었다. 비젼 기반의 트래킹을 기초로 한 증강 현실 기술은 미리 정의된 2차원 마커(marker)를 이용하여, 카메라로부터 획득된 실시간 영상 위에 가상 물체를 증강한다. 그러나, 카메라로부터 획득된 데이터는 몇몇 오차 요인들, 예를 들어 마커의 위치를 인식하는데 나타나는 오차, 카메라 안에 존재하는 센서 잡음 등으로 인해서 마커 잡음(마커를 인식하면서 나타나는 잡음)이 불가결하게 발생하게 된다. 이러한 이유로 인해서, 사용자가 한 손에는 마커를, 다른 한 손으로는 촉감 장치를 이용하여, 마커에 증강된 물체를 만질 때, 마커 잡음은 힘의 떨림(force trembling)을 발생시킨다. 심지어, 이러한 현상은 정지된 마커에 증강된, 마커가 움직이지 않는 상황에서도 발생한다. 게다가, 마커 위에 증강된 물체가 약간 빠른 속도로 이동하게 될 경우, 측정된 이동 거리는 연속적인 프레임(frame)들 간의 불연속적일 수 있다. 만약 사용자가, 대략 30Hz로 위치와 방향이 갱신되는 가상물체를 촉각적으로 상호작용하려 한다면, 계산되는 반력은 급작스런 힘의 변화를 생성하게 될 수도 있다. 이러한 현상을 극복하기 위해서, 마커 잡음을 최소화하기 위해서 정적 임계값(constant threshold)을 이용할 뿐만 아니라, 보간법을 같이 사용한 방법이 있었다. 하지만, 이러한 방법은 정적 임계값을 이용하고, 영상 프레임 갱신 속도와(video frame rate)와 촉각 프레임 갱신 속도가 일정하다는 가정을 사용하였기 때문에, 여전히 힘의 불연속적인 발생이 나타난다. 따라서, 이 논문에서는 두 가지 방법을 이용하여 증강 현실 내에서, 발생할 수 있는 힘의 불연속적인 변화를 보정하는 두 가지 방법, 잡음 제거를 위한 확장된 칼만 필터(Extend Kalman Filter)와 영상과 촉각 갱신 속도 차이에 따른 갑작스런 힘의 변화를 제거하기 위한 적응적 외삽법(Adaptive Extrapolation method)을 제안한다.

• Journal of KIISE:Software and Applications
• /
• v.34 no.1
• /
• pp.52-58
• /
• 2007

In this paper we propose a new robust method to extract accurate human silhouettes indoors with active stereo camera. A prime application is for gesture recognition of mobile robots. The segmentation of distant moving objects includes many problems such as low resolution, shadows, poor stereo matching information and instabilities of the object and background color distributions. There are many object segmentation methods based on color or stereo information but they alone are prone to failure. Here efficient color, stereo and image segmentation methods are fused to infer object and background areas of high confidence. Then the inferred areas are incorporated in graph cut to make human silhouette extraction robust and accurate. Some experimental results are presented with image sequences taken using pan-tilt stereo camera. Our proposed algorithms are evaluated with respect to ground truth data and proved to outperform some methods based on either color/stereo or color/contrast alone.

• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.5
• /
• pp.562-573
• /
• 2002

In this paper, we propose a method for measuring the volume of packages for the efficient handling of the packages. Using the geometrical characteristics of the rectangular parallelepiped type objects, the method measures the volume of packages with one camera only in real time. In preprocessing of volume measurement, the method extracts outer lines of the object and then crossing points of the lines as feature points or vertexes. From these cross points(-feature points-), the volume of the package is calculated. Compared to the direct feature extraction, the proposed method shows especially the blurring robust result by using the line for feature extraction. Additionally, the method can get the stable result by considering object's direction. From experimental results, it is demonstrated that this method is very effective for the real time volume measurement of the rectangular parallelepiped.

• Journal of Platform Technology
• /
• v.12 no.3
• /
• pp.62-70
• /
• 2024

Recent surveillance systems employ multiple sensors, such as cameras and radars, to enhance the accuracy of intrusion detection. However, object recognition through camera (RGB, Thermal) sensors may not always be accurate during nighttime, in adverse weather conditions, or when the intruder is camouflaged. In such situations, it is possible to detect intruders by utilizing the trajectories of objects extracted from camera or radar sensors. This paper proposes a method to detect intruders using only trajectory information in environments where object recognition is challenging. The proposed method involves training an LSTM-Attention based trajectory classification model using normal and abnormal (intrusion, loitering) trajectory data of animals and humans. This model is then used to identify abnormal human trajectories and perform intrusion detection. Finally, the validity of the proposed method is demonstrated through experiments using real data.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.07a
• /
• pp.201-204
• /
• 2010

물체 인식 및 추적 기술은 기계가 영상 정보를 기반으로 주변을 인지하고 정황을 파악하는 컴퓨터 비전 분야의 매우 중요한 연구 영역 중 하나이다. 현재까지 이러한 물체 인식/추적에 대한 다양한 연구들이 있어 왔고, 최근 증강현실에 대한 높은 관심을 바탕으로 증강현실을 위한 영상 정보 기반의 정확하고 정교한 추적 기술에 대한 관심 또한 매우 높아졌다. 본 논문에서는 텍스쳐가 없는 단색의 블록에 대해 증강현실을 실현하기 위한 물체 추적 방식을 제안한다. 제안하는 방식은 다수의 블록들을 조합하여 구성하고, 이 조합으로부터 추출한 특징점에 색상 정보 기반의 지역 서술자를 정의함으로써 사전에 정의된 서술자와 의 비교를 통해 물체를 추적하는 방식이다. 제안된 추적 방식은 사전에 기준이 되는 지역 서술자를 정의함에 있어서 기준 영상에 다양한 어파인 변환을 적용함으로써 카메라와 대상물과의 각도가 큰 입력 영상에 대해서도 추적에 실패하지 않는다. 실험을 통해 제안된 방식을 집 모양으로 구성한 블록 조합에 적용하여 3차원 가상 콘텐츠를 증강시켜 봄으로써 제안된 방식의 유용성을 확인하였다. 제안된 방식은 텍스쳐가 없는 환경에서 사용자의 상호작용으로 텍스쳐를 구성하고 이를 추적하는 방식으로 향후 아이들을 위한 교육 프로그램, 모바일 기기에서의 응용 프로그램 등으로 적용 가능하다.

• Journal of Satellite, Information and Communications
• /
• v.12 no.2
• /
• pp.56-60
• /
• 2017

This paper describes the design and implementation of fall-down detection algorithm based on image processing. The fall-down detection algorithm separates objects by using background subtraction and binarization after grayscale conversion of the input image acquired by the camera, and recognizes the human body by using labeling operation. The recognized human body can be monitored on the display image, and an alarm is generated when fall-down is detected. By using computer simulation, the proposed algorithm has shown a detection rate of 90%. We verify the feasibility of the proposed system by verifying the function by using the prototype test implemented on the DSP image processing board.