• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.997-999
• /
• 2003

The effects of JPEG2000 compression on automated DSM extraction by using the area-based matching are evaluated in this paper. The influences on DSM heights obtained via PCI Geomatics OrthoEngine module are investigated using a single stereo model of 1:5,000 scale aerial photography at image resolution of 20 ${\mu}$m. The experiment design of elevation errors are computed for a range of compression rates from 2:1 to about 100:1, and the DSM which generated from uncompressed image is used as ‘ground truth’ data for comparison. The experimental results show that the standard deviation ranged from 0.9m to 2.5m with the compression ratio from 2 to 100. It is also observed that there is no significant degradation on DSM accuracy up to the compression ratio of 33.

• Journal of Broadcast Engineering
• /
• v.19 no.5
• /
• pp.616-626
• /
• 2014

In this paper, we introduce a real-time stereo-to-multiview conversion system using FPGA and GPU. The system is based on two different devices so that it consists of two major blocks. The first block is a disparity estimation block that is implemented on FPGA. In this block, each disparity map of stereoscopic video is estimated by DP(dynamic programming)-based stereo matching. And then the estimated disparity maps are refined by post-processing. The refined disparity map is transferred to the GPU device through USB 3.0 and PCI-express interfaces. Stereoscopic video is also transferred to the GPU device. These data are used to render arbitrary number of virtual views in next block. In the second block, disparity-based view interpolation is performed to generate virtual multi-view video. As a final step, all generated views have to be re-arranged into a single image at full resolution for presenting on the target autostereoscopic 3D display. All these steps of the second block are performed in parallel on the GPU device.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.10 no.4
• /
• pp.318-324
• /
• 2017

We introduced a stereo camera on the aircraft to detect flight objects and to estimate the 3D position of them. The Saliency map algorithm based on PCT was proposed to detect a small object between clouds, and then we processed a stereo matching algorithm to find out the disparity between the left and right camera. In order to extract accurate disparity, cost aggregation region was used as a variable region to adapt to detection object. In this paper, we use the detection result as the cost aggregation region. In order to extract more precise disparity, sub-pixel interpolation is used to extract float type-disparity at sub-pixel level. We also proposed a method to estimate the spatial position of an object by using camera parameters. It is expected that it can be applied to image - based object detection and collision avoidance system of autonomous aircraft in the future.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.4 s.304
• /
• pp.13-20
• /
• 2005

In this paper, we propose new smoothing filters, i.e., occluding patterns that can accurately correct disparities of occluded areas in the estimated disparity map. An image is composed of several layers and each layer presents similar disparity. Furthermore, the distribution of the estimated disparities has a specific direction around the boundary of the occlusion, and this distribution presents the different direction with respect to the left- and the right-based disparity map. However, typical smoothing filters, such as mean filter and median filter, did not take into account those characteristic. So, they can decrease some error, but they cannot guarantee the accuracy of the corrected disparity. On the contrary, occluding patterns can accurately correct disparities of occluded areas because they consider both the characteristic that occlusion occurs and the characteristic that disparities of the occlusion are ranged, from estimated disparity maps with respect to the left and the right images. We made experiments on occluding patterns with some real stereo image set, and as a result, we can correct disparities of occluded areas more accurately than typical smoothing filters did.

• Journal of Korea Multimedia Society
• /
• v.1 no.1
• /
• pp.26-36
• /
• 1998

It is very important study field in computer vision 'How we obtain 3D information from 2D image'. For this purpose, we must know position of camera, direction of light source, and surface reflectance property before we take the image, which are intrinsic information of the object in the scene. Among them, surface reflectance property presents very important clues. Most previous researches assume that objects have only Lambertian reflectance, but many real world objects have Non-Lambertian reflectance property. In this paper the new method for analyzing the properties of surface reflectance and reconstructing the shape of object through estimation of reflectance parameters is proposed. We have interest in Non-Lambertian reflectance surface that has specular reflection and diffuse reflection which can be explained by Torrance-Sparrow model. Photometric matching method proposed in this paper is robust method because it match reference image and object image considering the neighbor brightness distribution. Also in this thesis, the neural network based shaped reconstruction method is proposed, which can be performed in the absence of reflectance information. When brightness obtained by each light is inputted, neural network is trained by surface normal and can determine the surface shape of object.

• Journal of Broadcast Engineering
• /
• v.7 no.3
• /
• pp.238-247
• /
• 2002

In this paper, we propose an object-based stereo sequence compression technique using disparity-motion vector relationship. The proposed method uses the coherence of motion vectors and disparity vectors in the left and right Image sequences. After two motion vectors and one disparity vector ate computed using FBMA(Fixed Block Matching Algorithm), the disparity vector of the current stereoscopic pall is computed by disparity-motion vector relationship with vectors which are previously estimated. Moreover, a vector regularization technique is applied in order to obtain reliable vectors. For an object-based coding. the object is defined and coded in terms of layers of VOP such as in MPEG-4. we present a method using disparity and motion vector relationship for extending two-frame compensation into three-frame compensation method for prediction coding of B-VOP. The proposed algorithm shows a high performance when comparing with a conventional method.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.36 no.4
• /
• pp.287-293
• /
• 2018

It is prerequisite to generate epipolar resampled images by reducing the y-parallax for accurate and efficient processing of satellite stereo images. Minimizing y-parallax requires the accurate sensor modeling that is carried out with ground control points. However, the approach is not feasible over inaccessible areas where control points cannot be easily acquired. For the case, a relative orientation can be utilized only with conjugate points, but its accuracy for satellite sensor should be studied because the sensor has different geometry compared to well-known frame type cameras. Therefore, we carried out the bias-compensation of RPCs (Rational Polynomial Coefficients) without any ground control points to study its precision and effects on the y-parallax in epipolar resampled images. The conjugate points were generated with stereo image matching with outlier removals. RPCs compensation was performed based on the affine and polynomial models. We analyzed the reprojection error of the compensated RPCs and the y-parallax in the resampled images. Experimental result showed one-pixel level of y-parallax for Kompsat-3 stereo data.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.2
• /
• pp.92-98
• /
• 2017

Generally, image contains geometrical and radiometric errors. Census transform can solve the stereo mismatching problem caused by the radiometric distortion. Since the general census transform compares center of window pixel value with neighbor pixel value, it is hard to obtain an accurate matching result when the difference of pixel value is not large. To solve that problem, we propose a census transform method that applies different 4-step weight for each pixel value difference by applying an assistance window inside the window kernel. If the current pixel value is larger than the average of assistance window pixel value, a high weight value is given. Otherwise, a low weight value is assigned to perform a differential census transform. After generating an initial disparity map using a weighted census transform and input images, the gradient information is additionally used to model a cost function for generating a final disparity map. In order to find an optimal cost value, we use guided filtering. Since the filtering is performed using the input image and the disparity image, the object boundary region can be preserved. From the experimental results, we confirm that the performance of the proposed stereo matching method is improved compare to the conventional method.

• Journal of Broadcast Engineering
• /
• v.17 no.6
• /
• pp.911-923
• /
• 2012

Recently, 3DTV and 3D displays have been released in the market. Accordingly, the production of stereoscopic images has gained much interest. Stereoscopic image being composed of left and right images are currently delivered to viewers without any modifications. The researches on the enhancement of depth perception using high-frequency components and the re-production of natural color by color compensation have been carried out for 2D images. The application of such 2D technologies to 3D stereoscopic images is an aim of this paper. This paper proposes the enhancement of 3D perception by color transformation. For this, we propose a stereo matching method for obtaining a depth map and two color transformation methods such as contrast transformation and background darkening. The effectiveness of the proposed method was verified through experiments.

• Korean Journal of Remote Sensing
• /
• v.31 no.2
• /
• pp.183-191
• /
• 2015

If the damage analysis on river-side facilities such as dam, river bank structures and bridges caused by disasters such as typhoon, flood, etc. becomes available, it can be a great help for disaster recovery and decision-making. In this research, We tried to extract a Digital Surface Model (DSM) and analyze the accuracy from Unmanned Air Vehicle (UAV) images on river-side facilities. We tried to apply stereo image-based matching technique, then extracted match results were united with one mosaic DSM. The accuracy was verified compared with a DSM derived from LIDAR data. Overall accuracy was around 3m of absolute and root mean square error. As an analysis result, we confirmed that exterior orientation parameters exerted an influence to DSM accuracy. For more accurate DSM generation, accurate EO parameters are necessary and effective interpolation and post process technique needs to be developed. And the damage analysis simulation with DSM has to be performed in the future.