• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.31 no.6_2
• /
• pp.611-623
• /
• 2013

Point cloud data (i.e., LiDAR; Light Detection and Ranging) collected by Airborne Laser Scanner (ALS) system is one of the major sources for surface reconstruction including DEM generation, topographic mapping and object modeling. Recently, demand and requirement of the accurate and realistic Digital Building Model (DBM) increase for geospatial platforms and spatial data infrastructure. The main issues in the object modeling such as building and city modeling are efficiency of the methodology and quality of the final products. Efficiency and quality are associated with automation and accuracy, respectively. However, these two factors are often opposite each other. This paper aims to introduce correction scheme of incorrectly determined Model Key Points (MKPs) regardless of the segmentation method. Planimetric and height locations of the MKPs were refined by surface patch fitting based on the Least-Squares Solution (LESS). The proposed methods were applied to the synthetic and real LiDAR data. Finally, the results were analyzed by comparing adjusted MKPs with the true building model data.

• Korean Journal of Remote Sensing
• /
• v.24 no.2
• /
• pp.141-152
• /
• 2008

The 3D building modeling is one of crucial components in constructing 3D geospatial information. The existing methods for 3D building modeling depend mainly on manual photogrammetric processes, which indeed take great amount of time and efforts. In recent years, many researches on 3D building modeling using aerial LiDAR data have been actively performed to aim at overcoming the limitations of existing 3D building modeling methods. Either techniques with interpolated grid data or data fusion with digital map and images have been investigated in most of existing researches on 3D building modeling with aerial LiDAR data. The paper proposed a method of 3D building modeling with LiDAR data only. Firstly, octree-based segmentation is applied recursively to LiDAR data classified as buildings in 3D space until there are no more LiDAR points to be segmented. Once octree-based segmentation is completed, each segmented patch is thereafter merged together based on its geometric spatial characteristics. Secondly, building model components are created with merged patches. Finally, a 3D building model is generated and composed with building model components. The experimental results with real LiDAR data showed that the proposed method was capable of modeling various types of 3D buildings.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.27 no.1
• /
• pp.39-47
• /
• 2017

Nowadays, there have been an increasing number of illegal use cases where people try to fabricate the working hours by using fake fingerprints. So, the fingerprint liveness detection techniques have been actively studied and widely demanded in various applications. This paper proposes a new method to detect fake fingerprints using CNN (Convolutional Neural Ntworks) based on the patches of fingerprint images. Fingerprint image is divided into small square sized patches and each patch is classified as live, fake, or background by the CNN. Finally, the fingerprint image is classified into either live or fake based on the voting result between the numbers of fake and live patches. The proposed method does not need preprocessing steps such as segmentation because it includes the background class in the patch classification. This method shows promising results of 3.06% average classification errors on LivDet2011, LivDet2013 and LivDet2015 dataset.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.34S no.6
• /
• pp.58-66
• /
• 1997

This paper proposes a multiresolution surface parameter estimation method for range images. Based on robust estimation of surface parameters, it approximates a patch to a planar surface in the locally adaptive window. Selection of resolution is made pixelwise by comparing a locally computed homogeneity measure with th eglobal threshold determined by te distribution of the approximation error. The proposed multiresolution surface parameter estimation method is applied to range image reconstruction. Computer simulation results with noisy rnag eimages contaminated by additive gaussian noise and impulse noise show that the proposed multiresolution reconstruction method well preserves step and roof edges compared with the conventional methods. Also the segmentation method based on the estimated surface parameters is shown to be robust to noise.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.11a
• /
• pp.161-163
• /
• 2022

본 논문에서는 패치매치 기법 및 분할 기법의 조밀 깊이지도들의 효율적인 결합을 통해 기존의 패치매치 기반의 방법들이 낮은 깊이값 추정 정확도를 보인 영역들인 텍스처가 부족한 영역과 기존의 분할 기반 방법들이 깊이값 추정에 한계를 보인 세밀한 영역에서의 깊이값 추정 정확도를 동시에 높이고 고품질의 조밀 깊이지도를 얻는 것을 목표로 한다. 이를 위해 제안한 방법에서는 신뢰지도를 바탕으로 패치매치 기법의 조밀 깊이지도, 조밀 노말지도와 분할 기법의 조밀 깊이지도, 조밀 노말지도의 초기 결합 깊이지도 및 초기 결합 노말지도를 생성한다. 이후 각 픽셀에서 원래 픽셀과 주변 픽셀에서의 깊이값, 노말값들로 업데이트를 위한 후보들을 만든다. 이후 각각의 후보들에 대해서 깊이값, 노말값, 컬러값들을 바탕으로 비용을 계산한다. 이후 가장 최적의 비용을 가지는 후보값으로 각 픽셀의 깊이값과 노말값을 업데이트한다. 이를 통해 패치매치 기법 및 분할 기법의 조밀 깊이지도들의 장점을 합친 결합 조밀 깊이지도를 생성한다.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.29-39
• /
• 2022

Crack detection plays an important role in the maintenance and protection of steel box girder of bridges. However, since the cracks only occupy an extremely small region of the high-resolution images captured from actual conditions, the existing methods cannot deal with this kind of image effectively. To solve this problem, this paper proposed a novel three-stage method based on deep learning technology and morphology operations. The training set and test set used in this paper are composed of 360 images (4928 × 3264 pixels) in steel girder box. The first stage of the proposed model converted high-resolution images into sub-images by using patch-based method and located the region of cracks by CBAM ResNet-50 model. The Recall reaches 0.95 on the test set. The second stage of our method uses the Attention U-Net model to get the accurate geometric edges of cracks based on results in the first stage. The IoU of the segmentation model implemented in this stage attains 0.48. In the third stage of the model, we remove the wrong-predicted isolated points in the predicted results through dilate operation and outlier elimination algorithm. The IoU of test set ascends to 0.70 after this stage. Ablation experiments are conducted to optimize the parameters and further promote the accuracy of the proposed method. The result shows that: (1) the best patch size of sub-images is 1024 × 1024. (2) the CBAM ResNet-50 and the Attention U-Net achieved the best results in the first and the second stage, respectively. (3) Pre-training the model of the first two stages can improve the IoU by 2.9%. In general, our method is of great significance for crack detection.

• Journal of Korean Society for Geospatial Information Science
• /
• v.19 no.1
• /
• pp.97-103
• /
• 2011

LiDAR system has become a popular tool for generating 3D surface data such as Digital Surface Model. Extraction of valuable information, such as digital building models, from LiDAR data has been an attractive research subject. This research addresses to extract planar patches from LiDAR data. Planar patches are important primitives consisting of man-made objects such as buildings. In order to determine the best fitted planes, this research proposed a method to reduce/eliminate the impact of the outliers and the intersection areas of two planes. After finishing plane fitting, planar patches are segmented by pseudo color values which are calculated by determined three plane parameters for each LiDAR point. In addition, a segmentation procedure is conducted using the pseudo color values to find planar patches. This paper evaluates the feasibility of the proposed method using both airborne and terrestrial LiDAR data.

• The KIPS Transactions:PartB
• /
• v.16B no.1
• /
• pp.47-54
• /
• 2009

Automatic object recognition in 3D measuring data is of great interest in many application fields e.g. computer vision, reverse engineering and digital factory. In this paper we present a software tool for a fully automatic object detection and parameter estimation in unordered and noisy point clouds with a large number of data points. The software consists of three interactive modules each for model selection, point segmentation and model fitting, in which the orthogonal distance fitting (ODF) plays an important role. The ODF algorithms estimate model parameters by minimizing the square sum of the shortest distances between model feature and measurement points. The local quadric surface fitted through ODF to a randomly touched small initial patch of the point cloud provides the necessary initial information for the overall procedures of model selection, point segmentation and model fitting. The performance of the presented software tool will be demonstrated by applying to point clouds.

• Journal of Broadcast Engineering
• /
• v.26 no.1
• /
• pp.26-38
• /
• 2021

In this paper, DLKF (Deep Learning Keypoint Filtering), the deep learning-based keypoint filtering method for the rapidization of the image registration method for remote sensing images is proposed. The complexity of the conventional feature-based image registration method arises during the feature matching step. To reduce this complexity, this paper proposes to filter only the keypoints detected in the artificial structure among the keypoints detected in the keypoint detector by ensuring that the feature matching is matched with the keypoints detected in the artificial structure of the image. For reducing the number of keypoints points as preserving essential keypoints, we preserve keypoints adjacent to the boundaries of the artificial structure, and use reduced images, and crop image patches overlapping to eliminate noise from the patch boundary as a result of the image segmentation method. the proposed method improves the speed and accuracy of registration. To verify the performance of DLKF, the speed and accuracy of the conventional keypoints extraction method were compared using the remote sensing image of KOMPSAT-3 satellite. Based on the SIFT-based registration method, which is commonly used in households, the SURF-based registration method, which improved the speed of the SIFT method, improved the speed by 2.6 times while reducing the number of keypoints by about 18%, but the accuracy decreased from 3.42 to 5.43. Became. However, when the proposed method, DLKF, was used, the number of keypoints was reduced by about 82%, improving the speed by about 20.5 times, while reducing the accuracy to 4.51.

• Korean Journal of Remote Sensing
• /
• v.10 no.2
• /
• pp.133-160
• /
• 1994

In this paper, we propose a comprehensive geometric correction algorithm of Along Track Scanning Radiometer(ATSR) images. The procedure consists of two cascaded modules; precorrection and fine co-location. The pre-correction algorithm is based on the navigation model which was derived in mathematical forms. This model was applied for correction raw(un-geolocated) ATSR images. The non-systematic geometric errors are also introduced as the limitation of the geometric correction by this analytical method. A fast and automatic algorithm is also presented in the paper for co-locating nadir and forward views of the ATSR images by using a binary cross-correlation matching technique. It removes small non-systematic errors which cannot be corrected by the analytic method. The proposed algorithm does not require any auxiliary informations, or a priori processing and avoiding the imperfect co-registratio problem observed with multiple channels. Coastlines in images are detected by a ragion segmentation and an automatic thresholding technique. The matching procedure is carried out with binaty coastline images (nadir and forward), and it gives comparable accuracy and faster processing than a patch based matching technique. This technique automatically reduces non-systematic errors between two views to .$\pm$ 1 pixel.