• Proceedings of the IEEK Conference
• /
• 2002.07c
• /
• pp.2004-2007
• /
• 2002

In the linear combination type stereophonic echo canceller, it is known not to converge the coefficient vector of the adaptive filter to a correct echo path. In this report, we analyze the convergence value of the filter coefficient vector of the stereo echo canceling algorithm using input signals of all channels in relation to this problem. In this analysis, one of the two inputs to the un-known system and adaptive one are assumed to be a delayed and attenuated version of the other signal as a model of the input signal with a strong cross-correlation. As a result, it is shown for the coefficient vectors not to converge to echo paths, and nor to converge to the value which depends on the time delay and the attenuation of the input signal. We show that the computer simulation result are corresponding to our analytical results.

• Journal of KIISE:Software and Applications
• /
• v.33 no.1
• /
• pp.69-83
• /
• 2006

This paper describes an image disparity estimation method using a segmented-region based stereo matching. Segmented-region based disparity estimation yields a disparity map as the unit of segmented region. However, there is a problem that it estimates disparity imprecisely. The reason is that because it not only have matching errors but also apply an identical way to disparity estimation, which is not considered each type of matched regions. To solve this problem, we proposes a disparity estimation method which is considered the type of matched regions. That is, the proposed method classifies whole matched regions into similar-matched region, dissimilar-matched region, false-matched region and miss-matched region by analyzing the type of matched regions. We then performs proper disparity estimation for each type of matched regions. This method minimizes the error in estimating disparity which is caused by inaccurate matching and also improves the accuracy of disparity of the well-matched regions. For the purpose of performance evaluations, we perform tests on a variety of scenes for synthetic, indoor and outdoor images. As a result of tests, we can obtain a dense disparity map which has the improved accuracy. The remarkable result is that the accuracy of disparity is also improved considerably for complex outdoor images which are barely treatable in the previous methods.

• Korean Journal of Remote Sensing
• /
• v.33 no.6_2
• /
• pp.1151-1157
• /
• 2017

Street Trees is an important object for urban environment improvement. Especially the height of the trees needs to be precisely measured as a factor that greatly influences the removal of air pollutants in the Urban Street Canyons. In this study, we extracted the height of the tree based on the stereo image using the precisely adjusted UAV Images of the target area. The adjustment of UAV image was applied photogrammetric SfM (Structure from motion) based on the collinear condition. We measured the height of the trees on the Street Canyon using stereoscopic vision on stereo plotting system. We also acquired the height of the building adjacent to the street trees and the average height of the road surface was calculated for accurate measurement of the height of each object. Through the visual analysis with the plotting operation system, it was possible to measure height of the tree and to calculate the relative height difference value with building quickly. This means that the height of buildings and trees can be calculated without making a 3D point cloud of UAV and it has the advantage of being able to utilize non-experts. In the future, further studies for semiautomatic/automation of this technique should be performed. The development and research of these technologies is expected to help to understand the current status of environmental policies and roadside trees in urban areas.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.41 no.6
• /
• pp.81-89
• /
• 2004

In this paper, we compare face recognition rate by PCA algorithm using distance change and embedded data being input left side and right side image in stereo images. The proposed method detects face region from RGB color space to YCbCr color space. Also, The extracted face image's scale up/down according to distance change and extracts more robust face region. The proposed method through an experiment could establish standard distance (100cm) in distance about 30∼200cm, and get 99.05% (100cm) as an average recognition result by scale change. The definition of super state is specification region in normalized size (92${\times}$112), and the embedded data extracts the inner factor of defined super state, achieved face recognition through PCA algorithm. The orignal images can receive specification data in limited image's size (92${\times}$112) because embedded data to do learning not that do all learning, in image of 92${\times}$112 size averagely 99.05%, shows face recognition rate of test 1 99.05%, test 2 98.93%, test 3 98.54%, test 4 97.85%. Therefore, the proposed method through an experiment showed that if apply distance change rate could get high recognition rate, and the processing speed improved as well as reduce face information.

• Convergence Security Journal
• /
• v.13 no.3
• /
• pp.9-15
• /
• 2013

Calculating of matching cost is an important for efficient stereo matching. To investigate the performance of matching process, the concepts of the existing methods are introduced. Also we analyze the performance and merits of them. The simplest matching costs assume constant intensities at matching image locations. We consider matching cost functions which can be distinguished between pixel-based and window-based approaches. The Pixel-based approach includes absolute differences (AD) and sampling-intensitive absolute differences (BT). The window-based approach includes the sum of the absolute differences, the sum of squared differences, the normalized cross-correlation, zero-mean normalized cross-correlation, census transform, and the absolute differences census transform (AD-Census). We evaluate matching cost functions in terms of accuracy and time complexity. In terms of the accuracy, AD-Census method shows the lowest matching error ratio (the best solution). The ZNCC method shows the lowest matching error ratio in non-occlusion and all evaluation part. But it performs high matching error ratio at the discontinuities evaluation part due to blurring effect in the boundary. The pixel-based AD method shows a low complexity in terms of time complexity.

• Korean Journal of Remote Sensing
• /
• v.35 no.6_1
• /
• pp.895-905
• /
• 2019

Commercial Unmanned Aerial Vehicle (UAV) image processing software products currently used in the industry provides camera calibration information and block bundle adjustment accuracy. However, they provide mapping accuracy achievable out of input UAV images. In this paper, the quality of mapping is calculated by using orientation parameters from UAV image processing software. We apply the orientation parameters to the digital photogrammetric workstation (DPW) for verifying the reliability of the mapping quality calculated. The quality of mapping accuracy was defined as three types of accuracy: Y-parallax, relative model and absolute model accuracy. The Y-parallax is an accuracy capable of determining stereo viewing between stereo pairs. The Relative model accuracy is the relative bundle adjustment accuracy between stereo pairs on the model coordinates system. The absolute model accuracy is the bundle adjustment accuracy on the absolute coordinate system. For the experimental data, we used 723 images of GSD 5 cm obtained from the rotary wing UAV over an urban area and analyzed the accuracy of mapping quality. The quality of the relative model accuracy predicted by the proposed technique and the maximum error observed from the DPW showed precise results with less than 0.11 m. Similarly, the maximum error of the absolute model accuracy predicted by the proposed technique was less than 0.16 m.

• Journal of Korean Society for Geospatial Information Science
• /
• v.23 no.1
• /
• pp.129-135
• /
• 2015

The Ubiquitous City (U-City) spatial information technology aimed to provide services freely anytime and anywhere by converging high-tech information & communication technology in urban infrastructure has been available in diverse patterns. In particular, there have been studies on the development of 3D spatial information after selecting and matching key points with stereo images from the many Closed Circuit TV (CCTV) in the U-City. However, the data mostly used in extracting matching points haven't considered external environmental impacts such as illuminance. This study tested how much the matching points needed to construct 3D spatial information with the CCTV whose image quality is dependent upon changes in illuminance fluctuate under the same hardware performances. According to analysis on the number of matching points by illuminance, the number of matching points increased up to 3,000Lux in proportion to the illuminance when IRIS, shutter speed and ISO were fixed. In addition, a border between an object and background became more distinctive. When there was too much light, however, the page became brighter, and noise occurred. Furthermore, it was difficult to name key points because of the collapse of an inter-object border. It appears that if filmed with the study results, the number of matching points would increase.

• Journal of Korean Society for Geospatial Information Science
• /
• v.23 no.2
• /
• pp.27-38
• /
• 2015

In this study, the 3D position coordinates were calculated for the targets using DLT and self-calibration bundle adjustment with additional parameters in close-range photogrammetry. And then, the accuracy of the results were analysed. For this purpose, the results of camera calibration and orientation parameters were calculated for each images by performing reference surveying using total station though the composition of experimental conditions attached numerous targets. To analyze the accuracy, 3D position coordinates were calculated for targets that has been identically selected and compared with the reference coordinates obtained from a total station. For the image coordinate measurement of the stereo images, we performed the ellipse fitting procedure for measuring the center point of the circular target. And then, the results were utilized for the image coordinate for targets. As a results from experiments, position coordinates calculated by the stereo images-based photogrammetry have resulted out the deviation of less than an average 4mm within the maximum error range of less than about 1cm. From this result, it is expected that the stereo images-based photogrammetry would be used to field of various close-range photogrammetry required for precise accuracy.

• Korean Journal of Remote Sensing
• /
• v.23 no.2
• /
• pp.113-117
• /
• 2007

In this paper, we propose a new method for determining automatically the size of the matching window using histogram of the gradient in order to improve the performance of stereo matching using one-meter resolution satellite imagery. For each pixel, we generate Flatness Index Image by calculating the mean value of the vertical or horizontal intensity gradients of the 4-neighbors of every pixel in the entire image. The edge pixel has high flatness index value, while the non-edge pixel has low flatness index value. By using the histogram of the Flatness Index Image, we find a flatness threshold value to determine whether a pixel is edge pixel or non-edge pixel. If a pixel has higher flatness index value than the flatness threshold value, we classify the pixel into edge pixel, otherwise we classify the pixel into non-edge pixel. If the ratio of the number of non-edge pixels in initial matching window is low, then we consider the pixel to be in homogeneous region and enlarge the size of the matching window We repeat this process until the size of matching window reaches to a maximum size. In the experiment, we used IKONOS satellite stereo imagery and obtained more improved matching results than the matching method using fixed matching window size.

• Korean Journal of Remote Sensing
• /
• v.30 no.1
• /
• pp.37-45
• /
• 2014

This paper reports the geolocation accuracy of KOMPSAT-3 imagery. KOMPSAT-3 was launched successfully on May 18, 2012 and has been released last March. In this paper, we have checked the geolocation accuracy of initial sensor model, precise sensor model and stereo-and multi-image model using four KOMPSAT-3 images covering the same area. The KOMPSAT-3 images without GCPs provided the geolocation accuracy of about 30m and the geocorrected KOMPSAT-3 images provided the geolocation accuracy of about 1m or less. KOMPSAT-3 stereo- and multi-images models yield threedimensional points with sub-meter accuracy in horizontal and vertical direction. Overall, KOMPSAT-3 showed much improved performance in terms of the geolocation accuracy over KOMPSAT-2. KOMPSAT-3 is expected to be able to replace foreign satellite data with sub-meter accuracy level for achieving accurate geometric information.