• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.36 no.2
• /
• pp.43-50
• /
• 2018

The camera is a key element in image-based three-dimensional positioning, and camera calibration, which properly determines the internal characteristics of such a camera, is a necessary process that must be preceded in order to determine the three-dimensional coordinates of the object. In this study, a new methodology was proposed to determine interior orientation parameters of a camera semi-automatically without being influenced by size and shape of checkerboard for camera calibration. The proposed method consists of exterior orientation parameters estimation using quaternion, recognition of calibration target, and interior orientation parameter determination through bundle block adjustment. After determining the interior orientation parameters using the chessboard calibration target, the three-dimensional position of the small 3D model was determined. In addition, the horizontal and vertical position errors were about ${\pm}0.006m$ and ${\pm}0.007m$, respectively, through the accuracy evaluation using the checkpoints.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.3
• /
• pp.123-132
• /
• 2021

To determine 3D(three-dimensional) positions using a stereo-camera in close-range photogrammetry, camera calibration to determine not only the interior orientation parameters of each camera but also the relative orientation parameters between the cameras must be preceded. As time passes after performing camera calibration, in the case of non-metric cameras, the interior and relative orientation parameters may change due to internal instability or external factors. In this study, to evaluate the stability of the stereo-camera, not only the stability of two single cameras and a stereo-camera were analyzed, but also the three-dimensional position accuracy was evaluated using checkpoints. As a result of evaluating the stability of two single cameras through three camera calibration experiments over four months, the root mean square error was ±0.001mm, and the root mean square error of the stereo-camera was ±0.012mm ~ ±0.025mm, respectively. In addition, as the results of distance accuracy using the checkpoint were ±1mm, the interior and relative orientation parameters of the stereo-camera were considered stable over that period.

• Proceedings of the KIEE Conference
• /
• 2003.07d
• /
• pp.2564-2566
• /
• 2003

표정인식은 컴퓨터 비전 분야에서 중요한 부분을 차지하고 있으며, 현재 꾸준히 연구가 진행되고 있다. 표정인식 시스템은 크게 얼굴 영역 추출과 표정인식 부분으로 나눌 수 있으며, 얼굴 영역 추출은 전체 인식 시스템의 성능에 큰 영향을 미친다. 특히 표정인식 시스템은 일반 얼굴인식 시스템과 다르게 부분적으로나 전체적으로 형태의 변화가 큰 얼굴에 대해서 정확한 얼굴 영역이 확보되지 않으면 높은 인식성능을 기대하기 어렵다. 따라서 표정인식 시스템은 얼굴 영역 추출이 비중한 부분을 차지하고 있다. 본 논문에서는 영상에서 실시간으로 얼굴 영역을 찾아내고, 그 영역에서 얼굴의 특징점인 눈과 입의 위치를 검출하고, 이를 바탕으로 얼굴의 정확한 영역을 확정하는 일련의 과정을 서술한다.

• The KIPS Transactions:PartB
• /
• v.18B no.6
• /
• pp.375-384
• /
• 2011

This paper suggests a method to recognize facial expressions of animation characters by means of dominant colors and feature points. The proposed method defines a simplified mesh model adequate for the animation character and detects its face and facial components by using dominant colors. It also extracts edge-based feature points for each facial component. It then classifies the feature points into corresponding AUs(action units) through neural network, and finally recognizes character facial expressions with the suggested AU specification. Experimental results show that the suggested method can recognize facial expressions of animation characters reliably.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.41 no.2
• /
• pp.131-136
• /
• 2004

According to development of media, various information is recorded in media, expression is one during interesting information. Because expression includes of relationship of human inside. Intention of inside is expressed by gesture, but expression has more information. And, expression can manufacture voluntarily, include plan of inside on the man. Also, expression has unique character in a person, have alliance that do division possibility. In this paper, to analyze expression of USB camera animation, wish to detect facial building block. Because characteristic point by person's expression change exists on face component. For component detection, in animation one frame with Capture, grasp facial position, and separate face area, and detect characteristic points of face component.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2018.05a
• /
• pp.137-137
• /
• 2018

표면영상유속계는 매우 간편하고 신속하게 하천의 유속장을 측정하는 기법이지만 하천의 넓은 구역을 카메라로 촬영하기 때문에 영상왜곡이 필연적으로 발생한다. 이러한 왜곡을 보정하기위해 많이 사용되고 있는 2차원 투영좌표변환법을 이용하여 유속을 분석하였다. 하지만 2차원 투영좌표변환법의 경우 표정점이 수표면의 높이와 같은 위치에 존재하지 않으면 유속 분석 결과에 큰 오차를 유발시킨다. 홍수 시 하천의 수위가 급변할 경우 표정점을 수위 변화에 맞추어 이동시키면서 영상을 촬영한다는 것은 현실적으로 불가능하다. 이러한 문제점을 극복하기위해 하천의 수위 변화에 대응하는 영상왜곡 보정 기법 개발이 필요하다. 이에 본 연구에서는 기존의 2차원 투영좌표변환법을 개선하기 위해 제방근처의 표정점 4개와 카메라의 좌표와 카메라와 수표면까지의 연직거리를 이용한 영상왜곡 보정식을 개발하였다. 그리고 표정점과 수표면의 높이를 다양하게 변화시키면서 개발한 보정식을 적용하였다. 표정점이 수위에 맞게 설정된 경우를 기준으로 수위보다 높게 설정된 표정점에 대하여 보정식을 적용한 경우의 유속은 표정점이 수위보다 높게 설정된 경우의 유속과 비교한 결과 오차가 크게 개선되었음을 확인하였다. 따라서 하천에 CCTV를 고정적으로 설치하여 유량을 산정할 경우 본 연구에서 제시한 표정점 보정식을 활용한다면 수위가 급변하는 상황에서도 정확한 표면유속을 산정할 수 있을 것으로 기대한다.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.35 no.5
• /
• pp.332-340
• /
• 2015

Acoustic emission (AE) is an effective nondestructive test that uses transient elastic wave generated by the rapid release of energy within a material to detect any further growth or expansion of existing defects. Over the past decades, because of environmental issues, the use of compressed natural gas (CNG) as an alternative fuel for vehicles is increasing because of environmental issues. For this reason, the importance and necessity of detecting defects on a CNG fuel tank has also come to the fore. The conventional AE method used for source location is highly affected by the wave speed on the structure, and this creates problems in inspecting a composite CNG fuel tank. Because the speed and dispersion characteristics of the wave are different according to direction of structure and laminated layers. In this study, both the conventional AE method and the energy based contour map method were used for source location. This new method based on pre-acquired D/B was used for overcoming the limitation of damage localization in a composite CNG fuel tank specimen which consists of a steel liner cylinder overwrapped by GFRP. From the experimental results, it is observed that the damage localization is determined with a small error at all tested points by using the energy based contour map method, while there were a number of mis-locations or large errors at many tested points by using the conventional AE method. Therefore, the energy based contour map method used in this work is more suitable technology for inspecting composite structures.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.8 no.1
• /
• pp.88-95
• /
• 2005

Direct Georeferencing by combined multi-sensor based on the direct measurement of the projection centers and rotation angle of sensor through loading the GPS and INS in aircraft. The method of combined multi-sensor can offer us to acquire the exterior orientation parameters with only minimum GCPs, even the ground control process could be completely skipped. Consequently, It is possible extreme to reduce the time and expense for the mapping process. In this study, a CCD camera is simultaneously used in combined multi-sensor surveying, and acquired CCD image through Direct Georeferencing produce digital orthoimage. In this process, methods of combining sensor and digital orthoimage are examined and estimated. For the comparison of the positioning accuracy digital orthoimage through Direct Georeferencing, GCPs determined by GPS surveying are used. Two digital orthoimage are produced; one with a few GCP and the other without them. The accuracy of orthoimages produced through combined multi-sensor with GCPs meets that of 1:1,000 maps. Without GCPs, it meets that of 1:5,000 maps.

• Convergence Security Journal
• /
• v.19 no.1
• /
• pp.67-78
• /
• 2019

Most existing facial expression recognition methods use a uniform grid method that divides the entire facial image into uniform blocks when describing facial features. The problem of this method may include non-face backgrounds, which interferes with discrimination of facial expressions, and the feature of a face included in each block may vary depending on the position, size, and orientation of the face in the input image. In this paper, we propose a variable-size block method which determines the size and position of a block that best represents meaningful facial expression change. As a part of the effort, we propose the way to determine the optimal number, position and size of each block based on the facial feature points. For the evaluation of the proposed method, we generate the facial feature vectors using LDTP and construct a facial expression recognition system based on SVM. Experimental results show that the proposed method is superior to conventional uniform grid based method. Especially, it shows that the proposed method can adapt to the change of the input environment more effectively by showing relatively better performance than exiting methods in the images with large shape and orientation changes.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.33 no.6
• /
• pp.524-530
• /
• 2013