• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.10
• /
• pp.983-988
• /
• 2011

This paper presents a robust method for ground plane detection in vision-based applications based on a monocular sequence of images with a non-stationary camera. The proposed method, which is based on the reliable estimation of the homography between two frames taken from the sequence, aims at designing a practical system to detect road surface from traffic scenes. The homography is computed using a feature matching approach, which often gives rise to inaccurate matches or undesirable matches from out of the ground plane. Hence, the proposed homography estimation minimizes the effects from erroneous feature matching by the evaluation of the difference between the predicted and the observed matrices. The method is successfully demonstrated for the detection of road surface performed on experiments to fill an information void area taken place from geometric transformation applied to captured images by an in-vehicle camera system.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.2
• /
• pp.61-66
• /
• 2019

Augmented reality is a technology that makes a virtual object appear as if it exists in reality by composing a virtual object in real time with the image captured by the camera. In order to augment the virtual object on the object existing in reality, the homography of images utilized to estimate the position and orientation of the object. The homography can be estimated by applying the RANSAC algorithm to the feature points of the images. But the homography estimation method using the RANSAC algorithm has a problem that accurate homography can not be estimated when there are many feature points in the background. In this paper, we propose a method to filter feature points of a background when the object is near and the background is relatively far away. First, we classified the depth image into relatively near region and a distant region using the Otsu's method and improve homography estimation performance by filtering feature points on the relatively distant area. As a result of experiment, processing time is shortened 71.7% compared to a conventional homography estimation method, and the number of iterations of the RANSAC algorithm was reduced 69.4%, and Inlier rate was increased 16.9%.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.5 no.4
• /
• pp.297-301
• /
• 2005

In this paper, we propose a method to estimate camera motion parameter based on invariant point features. Typically, feature information of image has drawbacks, it is variable to camera viewpoint, and therefore information quantity increases after time. The LM(Levenberg-Marquardt) method using nonlinear minimum square evaluation for camera extrinsic parameter estimation also has a weak point, which has different iteration number for approaching the minimal point according to the initial values and convergence time increases if the process run into a local minimum. In order to complement these shortfalls, we, first propose constructing feature models using invariant vector of geometry. Secondly, we propose a two-stage calculation method to improve accuracy and convergence by using homography and LM method. In the experiment, we compare and analyze the proposed method with existing method to demonstrate the superiority of the proposed algorithms.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.8
• /
• pp.215-224
• /
• 2013

An efficient homography estimation method for large sized images is proposed. Estimating an accurate homography is one of the most important parts in image stitching processes. Since hardwares have been advanced, it has been passible to take higher resolution images. However, computational cost for estimating homography has been also increased. Specifically, when too many features exist in the images, it requires lots of computations to estimate a correct homography. Furthermore, there is a high probability of obtaining an incorrect homography. Therefore, we propose a numerical method to extract the appropriate correspondences from several down-scaled images to estimate and compensate the homography numerically for restoring an original homography. Also, if there is an unbalance in color tone between the reference and the target images, we make them balanced by using local information of the overlapped regions. Experimental results show that proposed method is three times faster in 3.2 mega pixel images, five times faster in 8mega pixel images than the conventional approach. Therefore, we believe that the proposed method can be a useful tool to efficiently estimate a homography.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.5
• /
• pp.1162-1170
• /
• 2014

In this paper, an accurate remote gaze tracking method with two cameras is presented using a modified Starburst algorithm and honography normalization. Starburst algorithm, which was originally developed for head-mounted systems, often fails in detecting accurate pupil centers in remote tracking systems with a larger field of view due to lots of noises. A region of interest area for pupil is found using template matching, and then only within this area Starburst algorithm is applied to yield pupil boundary candidate points. These are used in improved RANSAC ellipse fitting to produce the pupil center. For gaze estimation robust to head movement, an improved homography normalization using four LEDs and calibration based on high order polynomials is proposed. Finally, it is shown that accuracy and robustness of the system is improved using two cameras rather than one camera.

• Journal of Internet Computing and Services
• /
• v.6 no.6
• /
• pp.187-197
• /
• 2005

In this paper, we propose a method to estimate camera motion parameter based on invariant point features, Typically, feature information of image has drawbacks, it is variable to camera viewpoint, and therefore information quantity increases after time, The LM(Levenberg-Marquardt) method using nonlinear minimum square evaluation for camera extrinsic parameter estimation also has a weak point, which has different iteration number for approaching the minimal point according to the initial values and convergence time increases if the process run into a local minimum, In order to complement these shortfalls, we, first proposed constructing feature models using invariant vector of geometry, Secondly, we proposed a two-stage calculation method to improve accuracy and convergence by using 2D homography and LM method, In the experiment, we compared and analyzed the proposed method with existing method to demonstrate the superiority of the proposed algorithms.

• Journal of Korea Society of Industrial Information Systems
• /
• v.27 no.1
• /
• pp.19-26
• /
• 2022

This paper presents a pre-processing method to facilitate the container's BIC-code recognition. We propose a network that can find ROI(Region Of Interests) containing a BIC-code region and estimate a homography matrix for warping. Taking the structure of STN(Spatial Transformer Networks), the proposed network consists of next 3 steps, ROI detection, homography matrix estimation, and warping using the homography estimated in the previous step. It contributes to improving the accuracy of BIC-code recognition by estimating ROI and matrix using the proposed network and correcting perspective distortion of ROI using the estimated matrix. For performance evaluation, five evaluators evaluated the output image as a perfect score of 5 and received an average of 4.25 points, and when visually checked, 224 out of 312 photos are accurately and perfectly corrected, containing ROI.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.6
• /
• pp.499-508
• /
• 2023

In this paper, we propose a technique to estimate the coordinates of feature points existing on a 2D planar object in the three dimensional space. The proposed method detects multiple 3D features from the image, and excludes those which are not located on the plane. The proposed technique estimates the planar homography between the planar object in the 3D space and the camera image plane, and computes back-projection error of each feature point on the planar object. Then any feature points which have large error is considered as off-plane points and are excluded from the feature estimation phase. The proposed method is archived on the basis of the planar homography without any additional sensors or optimization algorithms. In the expretiments, it was confirmed that the speed of the proposed method is more than 40 frames per second. In addition, compared to the RGB-D camera, there was no significant difference in processing speed, and it was verified that the frame rate was unaffected even in the situation that the number of detected feature points continuously increased.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.43 no.2 s.308
• /
• pp.11-19
• /
• 2006

In this paper, we propose a method to estimate camera extrinsic parameter based on projective and permutation invariance point features. Because feature informations in previous research is variant to c.:men viewpoint, extraction of correspondent point is difficult. Therefore, in this paper, we propose the extracting method of invariant point features, and new matching method using similarity evaluation function and Graham search method for reducing time complexity and finding correspondent points accurately. In the calculation of camera extrinsic parameter stage, we also propose two-stage motion parameter estimation method for enhancing convergent degree of LM algorithm. In the experiment, we compare and analyse the proposed method with existing method by using various indoor images to demonstrate the superiority of the proposed algorithms.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.401-403
• /
• 2021

본 시스템은 같은 환자로부터 촬영한 기존 안저영상과 초광각 안저영상을 Paired Dataset으로 지니고 있으며, 영상의 크기 및 해상도를 똑같이 맞추고, 황반부와 신경유두 및 혈관의 위치를 미세조정하는 과정을 자동화하는 것을 목표로 하고 있다. 이 과정은 황반부를 중심으로 하여 영상을 잘라내어 이미지의 크기를 맞추는 과정(Scaling)과, 황반부를 중심으로 잘라낸 한 쌍의 영상을 포개었을 때 황반부, 신경 유두, 혈관 등의 위치가 동일하도록 미세조정하는 과정(Warping)이 있다. Scaling Stage에선 기존 안저영상과 초광각 안저영상의 촬영범위가 현저하게 차이나기 때문에, 황반변성 부위를 잘 나타내도록 사전에 잘라낼 필요가 있으며, 이를 신경유두의 Object Detection을 활용할 예정이다. Warping Stage에선 동일한 위치에 같은 황반변성 정보가 내포되어야 하므로 규격조정 및 위치조정 과정이 필수적이며, 이후 안저영상 내의 특징들을 매칭하는 작업을 하기 위해 회전, 회절, 변환 작업 등이 이루어지며, 이는 Homography Estimation을 통하여 이미지 변환 matrix를 구하는 방법으로 진행된다. 자동조정된 안저영상 데이터는 추후에 GAN을 이용한 안저영상 생성모델을 위한 학습데이터로 이용할 예정이며, 현재로선 2500쌍의 데이터를 대상으로 실험을 진행중이지만, 최종적으로 3만 쌍의 안저영상 데이터를 목표로 하고 있다.