• Journal of the Korean Institute of Telematics and Electronics
• /
• v.27 no.8
• /
• pp.1275-1283
• /
• 1990

Perception of surfaces from range images plays a key role in 3-D object recognition. Recognition of 3-D objects from range images is performed by matching the perceived surface descriptions with stored object models. The first step of the 3-d object recognition from range images is image segmentation. In this paper, an approach for segmenting 3-D range images into symbolic surface descriptions using spatial Gabor filter is proposed. Since the phase of data has a lot of important information, the phase information with magnitude information can effectively segment the range imagery into regions satisfying a common homogeneity criterion. The phase and magnitude of Gabor filter can represent a unique featur vector at a point of range data. As a result, range images are trnasformed into feature vectors in 3-parameter representation. The methods not only to extract meaningful features but also to classify a patch information from range images is presented.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.3
• /
• pp.459-470
• /
• 2013

The discovery of nearest neighbors, without training in advance, has many applications, such as the formation of mosaic images, image matching, image retrieval and image stitching. When the quantity of data is huge and the number of dimensions is high, the efficient identification of a nearest neighbor (NN) is very important. This study proposes a variation of the KD-tree - the arbitrary KD-tree (KDA) - which is constructed without the need to evaluate variances. Multiple KDAs can be constructed efficiently and possess independent tree structures, when the amount of data is large. Upon testing, using extended synthetic databases and real-world SIFT data, this study concludes that the KDA method increases computational efficiency and produces satisfactory accuracy, when solving NN problems.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2015.11a
• /
• pp.185-188
• /
• 2015

미리 저장된 객체의 3차원 특징점(Feature point) 좌표와 카메라 영상의 2차원 특징점 좌표를 매칭(Matching)하여 객체를 추적하는 방식의 경우, 카메라의 시점이 변할 때 특징점에서 발생되는 원근 효과(Perspective effect)가 반영되지 못하여 특징점 매칭 오류가 발생한다. 따라서 특징점에서 발생하는 원근 효과를 반영하여 정확한 카메라 포즈를 추정하기 위해 이전 프레임(Frame)의 카메라 포즈(Camera Pose)에 맞추어 텍스쳐가 포함 된 3차원 객체의 모델을 렌더링 하여 원근 효과를 적용한 후, 현재 카메라 영상과 특징점 매칭하여 프레임 사이의 카메라 움직임을 구하여 객체를 추적한다. 더 나아가 본 논문에서는 특징점 매칭에서 발생하는 작은 오류들로 인한 미세한 카메라 움직임은 2단계의 임계치(Threshold)를 적용하여 떨림 현상으로 간주하여 떨림 현상이 제거된 객체 추적을 수행한다. 매 프레임마다 카메라 포즈에 맞춘 추적 객체를 렌더링 하기 때문에 떨림 현상으로 간주되어 제거된 카메라 움직임은 누적되지 않고, 추적 오류도 발생시키지 않는다.

• Proceedings of the KIEE Conference
• /
• 1999.07g
• /
• pp.2950-2952
• /
• 1999

This paper presents the method of 3D reconstruction of the depth information from the endoscopic stereo scopic images. After camera modeling to find camera parameters, we performed feature-point based stereo matching to find depth information. Acquired some depth information is finally 3D reconstructed using the NURBS(Non Uniform Rational B-Spline) method and OpenGL. The final result image is helpful for the understanding of depth information visually.

• Journal of KIISE:Software and Applications
• /
• v.28 no.5
• /
• pp.440-447
• /
• 2001

본 논문은 영상과 연관된 3차원 정보로부터 초기 3차원 변환을 유추, 상관윈도우를 변환시켜 정합에 이용하는 새로운 정합기법을 제안한다. 즉, 초기 스테레오 정합 등을 통한 3차원 정보를 추출하고, 인위적인 초기 특징점의 대응을 통해 3차원 변환을 얻으며, 이를 이용해 상관 윈도우의 3차원 변환을 가능하게 한다. 상관 윈도우의 3차원 변환은 기존의 방법이 가지는 영상 흐름의 2차원적인 제한을 이용한 정합방법에 비해 실제 카메라의 변환 유추에 합당하다. 또한 3차원 변환을 통해 정합 대상 점의 탐색범위를 최소화하고 정합의 결과에 신뢰성을 더한다. 실험에서는 다양한 영상의 정합 결과와 기존 방법과의 상관 계수 비교를 통해 본 논문이 제안하는 정합방법의 우월성을 보인다.

• Proceedings of the KIEE Conference
• /
• 2001.07d
• /
• pp.2818-2820
• /
• 2001

This paper presents the method of 3D reconstruction of the depth information from the endoscopic stereo scopic images. After camera modeling to find camera parameters, we performed feature-point based stereo matching to find depth information. Acquired some depth information is finally 3D reconstructed using the NURBS(Non Uniform Rational B-Spline) algorithm. The final result image is helpful for the understanding of depth information visually.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.847-850
• /
• 2022

최근 가상 현실(Virtual Reality), 파노라마(Panorama) 영상 등에 관한 관심과 수요가 증가함에 따라 고해상도 영상을 얻기 위한 영상 스티칭(Image Stitching)에 관한 연구가 활발하게 진행되고 있다. 영상 스티칭은 다수의 영상을 하나의 영상으로 합성해 카메라의 좁은 시야각 문제를 해결함으로써 사용자에게 몰입감과 현장감을 제공할 수 있는 기술이다. 영상 스티칭에 있어 특징점 추출 및 매칭 과정의 정확도는 스티칭 영상의 품질을 결정짓는 핵심적인 요소이지만, 기존의 특징점 추출 및 매칭 방법은 밝기가 어둡고 선명도가 낮은 영상의 스티칭에서 정확도가 저하될 수 있고 생성된 스티칭 영상의 품질 또한 저하될 수 있다는 제한 사항이 있다. 이에 본 논문에서는 앞선 제한적 영상에 대하여 특징점 추출 및 매칭의 정확도를 높여 스티칭 영상의 품질을 높이기 위하여 SuperPoint와 SuperGLUE를 활용한 입력 영상의 밝기 적응형 영상 스티칭 방법을 제안하고자 한다.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.31 no.2
• /
• pp.165-172
• /
• 2013

In this paper, high-quality mosaic image is generated by high-resolution hyperspectral strip images using scale-invariant feature transform (SIFT) algorithm, which is one of the representative image matching methods. The experiments are applied to AISA Eagle images geo-referenced by using GPS/INS information acquired when it was taken on flight. The matching points between three strips of hyperspectral images are extracted using SIFT method, and the transformation models between images are constructed from the points. Mosaic image is, then, generated using the transformation models constructed from corresponding images. Optimal band appropriate for the matching point extraction is determined by selecting representative bands of hyperspectral data and analyzing the matched results based on each band. Mosaic image generated by proposed method is visually compared with the mosaic image generated from initial geo-referenced AISA hyperspectral images. From the comparison, we could estimate geometrical accuracy of generated mosaic image and analyze the efficiency of our methodology.

• ETRI Journal
• /
• v.25 no.2
• /
• pp.109-120
• /
• 2003

Though the airborne laser scanning (ALS) technique is becoming more popular in many applications, horizontal accuracy of points scanned by the ALS is not yet satisfactory when compared with the accuracy achieved for vertical positions. One of the major reasons is the drift that occurs in the inertial measurement unit (IMU) during the scanning. This paper presents an algorithm that adjusts for the error that is introduced mainly by the drift of the IMU that renders systematic differences between strips on the same area. For this, we set up an observation equation for strip-wise adjustments and completed it with tie point and control point coordinates derived from the scanned strips and information from aerial photos. To effectively capture the tie points, we developed a set of procedures that constructs a digital surface model (DSM) with breaklines and then performed feature-based matching on strips resulting in a set of reliable tie points. Solving the observation equations by the least squares method produced a set of affine transformation equations with 6 parameters that we used to transform the strips for adjusting the horizontal error. Experimental results after evaluation of the accuracy showed a root mean squared error (RMSE) of the adjusted strip points of 0.27 m, which is significant considering the RMSE before adjustment was 0.77 m.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.21 no.10
• /
• pp.1551-1560
• /
• 1997

This study is devoted to the detection of the 3-dimensional point obstacles on the plane by using accumulated scan line images. The proposed accumulating only one scan line allow to process image at real time. And the change of motion of the feature in image is small because of the short time between image frames, so it does not take much time to track features. To obtain recursive optimal obstacles position and robot motion along to the motion of camera, Kalman filter algorithm is used. After using Kalman filter in case of the fixed environment, 3-dimensional obstacles point map is obtained. The position and motion of moving obstacles can also be obtained by pre-segmentation. Finally, to solve the stereo ambiguity problem from multiple matches, the camera motion is actively used to discard mis-matched features. To get relative distance of obstacles from camera, parallel stereo camera setup is used. In order to evaluate the proposed algorithm, experiments are carried out by a small test vehicle.