Ozcan, A.H. | Ünsalan, Cem

Article | 2017 | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing10 ( 1 ) , pp.360 - 371

LiDAR technology is advancing. As a result, researchers can benefit from high-resolution height data from Earth's surface. Digital terrain model (DTM) generation and point classification (filtering) are two important problems for LiDAR data. These are connected problems since solving one helps solving the other. Manual classification of LiDAR point data could be time consuming and prone to errors. Hence, it would not be feasible. Therefore, researchers proposed several methods to solve DTM generation and point classification problems. Although these methods work fairly well in most cases, they may not be effective for all scenarios. . . . To contribute in this research topic, a novel method based on two-dimensional (2-D) empirical mode decomposition (EMD) is proposed in this study. Local, nonlinear, and nonstationary characteristics of EMD allow better DTM generation. The proposed method is tested on two publicly available LiDAR dataset, and promising results are obtained. Besides, the proposed method is compared with other methods in the literature. Comparison results indicate that the proposed method has certain advantages in terms of performance. © 2008-2012 IEEE Daha fazlası Daha az

Benveniste, R. | Ünsalan, Cem

Conference Object | 2010 | Lecture Notes in Electrical Engineering62 LNEE , pp.241 - 244

Three dimensional range data provides useful information for computer vision, computer graphics, and object recognition applications. For these, extracting the range data reliably is utmost important. Therefore, various range scanners based on different operating principles are proposed in the literature. Although these scanners can be used in diverse applications, most of them cannot be used to scan shiny objects under ambient light. This is a severe restriction. We propose color invariant based binary and ternary coded structured light range scanners to solve this problem. We hypothesize that, by using color invariants we can elim . . .inate the effects of highlights and ambient light in the scanning process. Therefore, we can extract the range data of shiny and matte objects in a robust manner. We implemented three different range scanners to test our hypothesis. We performed tests on various objects and provided their range data. © 2011 Springer Science+Business Media B.V Daha fazlası Daha az

Irgan, K. | Ünsalan, Cem | Baydere, Ş.

Conference Object | 2010 | Lecture Notes in Electrical Engineering62 LNEE , pp.191 - 194

25th International Symposium on Computer and Information Sciences, ISCIS 2010 -- 22 September 2010 through 24 September 2010 -- London -- 82255

Sirmaçek, B. | Ünsalan, Cem

Article | 2010 | Pattern Recognition Letters31 ( 10 ) , pp.1155 - 1159

Monitoring urban development in a given region provides valuable information to researchers. Currently available, very high resolution satellite images can be used for this purpose. However, manually monitoring land development using these large and complex images is time consuming and prone to errors. To handle this problem, an automated system is needed to measure development in urban regions. Therefore, in this study we propose such an automated method to measure land development in a given urban region imaged in different times. We benefit from novel land development measures for this purpose. They are based on local features ob . . .tained from sequential images. As a novel contribution, we represent these local features in a spatial voting matrix. Then, we propose five different land development measures on the formed voting matrix. We test our method on 19 sets of sequential panchromatic Ikonos images. Our test results indicate the possible use of our method in measuring land development automatically. © 2009 Elsevier B.V. All rights reserved Daha fazlası Daha az

Ünsalan, Cem

Article | 2006 | IEEE Geoscience and Remote Sensing Letters3 ( 4 ) , pp.546 - 550

Land use classification is one of the major problems in remote sensing. Previous studies focused on multispectral information, texture-based features, and features based on edge detection to classify land usage from satellite images. In a previous study, structural features are introduced to classify land development using high-resolution satellite images. These structural features were based on line support regions (LSRs). LSRs are introduced to detect and represent straight lines in images using a pixel-grouping process. The structural features are calculated on these grouped pixels. It is shown that gradient-magnitude-based pixel . . . grouping may also be used in structural feature calculations. Therefore, the aim of this letter is twofold. First, the previous structural feature calculation method is shown to be more general than the LSR. Second, LSR-based features are shown to require fairly high computation compared to gradient-magnitude-based features with similar classification performance. © 2006 IEEE Daha fazlası Daha az

Özcan, A.H. | Ünsalan, Cem | Reinartz, P.

Article | 2018 | International Journal of Remote Sensing39 ( 9 ) , pp.2860 - 2883

Automated digital terrain model (DTM) generation from remotely sensed data has gained wide application areas due to increased sensor resolution. In this study, a novel ground filtering and segmentation method is proposed for digital surface model (DSM) data. The proposed method starts with extracting DSM feature points. These are used in a probabilistic framework to generate a non-ground object probability map in spatial domain. Modes of this map are used as seed points in a novel segmentation method based on morphological operations. This leads to ground filtering and DTM generation. The method is tested on three different data set . . .s. Two of these originate from light detection and ranging (lidar) sensors, where resulting kappa coefficient (?) range mostly higher than 95% for differently structured urban areas. Also, the visual appearance of the generated DTM exhibits obvious improvements over all other investigated methods. The third data set is a DSM obtained from WorldView-2 stereo image pairs. Also here, we compare our results with three different methods in the literature. Although the DSM quality is much lower, more than 85% of ? can be reached by the proposed method, showing its superiority over other methods. Overall experimental results show that the proposed method can be used reliably for DTM generation. The results also indicate that the method has prominent advantages in comparison to established methodologies in terms of robustness in handling urban areas of different properties. Moreover, there are only few parameters to adjust in the proposed method, and these are independent of the object size in DSM data. © 2018 Informa UK Limited, trading as Taylor & Francis Group Daha fazlası Daha az

Ünsalan, Cem

Article | 2007 | Pattern Recognition Letters28 ( 1 ) , pp.49 - 57

Pose estimation has been considered to be an important component in many pattern recognition and computer vision systems. In this paper, we introduce a pose estimation method based on implicit polynomials. We also introduce a set of rotation invariant measures for object matching. We test both methods under colored noise, missing points, and affine transformation. Extensive testings indicate that our methods work fairly well under various disturbances. We also compare our methods with existing ones in the literature. © 2006 Elsevier B.V. All rights reserved.

İlsever, M. | Ünsalan, Cem

Book Part | 2012 | SpringerBriefs in Computer Science ( 9781447142546 ) , pp.57 - 70

In this chapter, we provide experimental results obtained from all change detection methods considered in this study. We first explain the data set used in the experiments. Then, we provide performance results in tabular form for each change detection method in detail. © 2012, Cem Ünsalan.

Altintaş, A. | Ünsalan, Cem | Keskin, A.U. | Yencilek, F.

Conference Object | 2010 | 2010 15th National Biomedical Engineering Meeting, BIYOMUT2010 , pp.57 - 70

Extracorporeal Shock Wave Lithotrispy (ESWL) is a procedure based on sound waves to crash kidney stones on the focus. The sound waves are sent to the body of patient when the kidney stone is not even on the focus. When the stone is not on the focus, the sound waves can damage the soft tissue of the kidney. This damage can be prevented by a feedback mechanism that determines the place of kidney stones depending on the images taken from ESWL device. In this study, an automated system is developed to detect kidney stones from X ray images. ©2010 IEEE.

Sertel, O. | Ünsalan, Cem

Conference Object | 2006 | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)4105 LNCS , pp.745 - 752

In this study, we focus on model reconstruction for 3D objects using range images. We propose a crude range image alignment method to overcome the initial estimation problem of the iterative closest point (ICP) algorithm using edge points of range images. Different from previous edge detection methods, we first obtain a function representation of the range image in spherical coordinates. This representation allows detecting smooth edges on the object surface easily by a zero crossing edge detector. We use ICP on these edges to align patches in a crude manner. Then, we apply ICP to the whole point set and obtain the final alignment. . . .This dual operation is performed extremely fast compared to directly aligning the point sets. We also obtain the edges of the 3D object model while registering it. These edge points may be of use in 3D object recognition and classification. © Springer-Verlag Berlin Heidelberg 2006 Daha fazlası Daha az

Ilsever, M. | Ünsalan, Cem

Conference Object | 2013 | RAST 2013 - Proceedings of 6th International Conference on Recent Advances in Space Technologies , pp.115 - 119

One of the most important applications of remote sensing is detecting and analyzing man-made structures in aerial and satellite images. Buildings deserve specific consideration among man-made structures. Manual detection is the one way of labeling the buildings. Recent satellite image resolutions allow researchers manual detection. However, this approach is cumbersome and prone to errors. Therefore, an automated method is needed. Unfortunately, the solution is not straightforward due to various factors. In this study, we propose an adaptation of an earlier successful method, Histogram of Oriented Gradients, to building detection. We . . . test our method on several Ikonos satellite images and provide the results. © 2013 IEEE Daha fazlası Daha az

Özcan, A.H. | Ünsalan, Cem

Conference Object | 2016 | 2016 24th Signal Processing and Communication Application Conference, SIU 2016 - Proceedings , pp.429 - 432

Automated object detection in remotely sensed data has gained wide application areas due to increased sensor resolution. In this study, we propose a novel building detection method using high resolution DSM data and true orthophoto image. In the proposed method, DSM feature points and NDVI are obtained. Then, they are used for spatial voting to generate a building probability map. Local maxima of this map are used as seed points for segmentation. For this purpose, a morphology based segmentation method is proposed. This way, buildings are detected from DSM data. We tested our method on ISPRS semantic labeling dataset and obtained pr . . .omising results. © 2016 IEEE Daha fazlası Daha az