Filtering and mapping systems for underwater 3D imaging sonar

Size: px
Start display at page:

Download "Filtering and mapping systems for underwater 3D imaging sonar"

Transcription

1 Filtering and mapping systems for underwater 3D imaging sonar Tomohiro Koshikawa 1, a, Shin Kato 1,b, and Hitoshi Arisumi 1,c 1 Field Robotics Research Group, National Institute of Advanced Industrial Science and Technology (AIST), Umezono, Tsukuba city, Ibaraki, Japan a <koshikawa.tomohiro@aist.go.jp>, b <shin.kato@aist.go.jp >, c <h-arisumi@aist.go.jp> Keywords: sonar, filtering, 3d mapping, point cloud Abstract. In recent years, seawalls and piers are aging in Japan and using of sonar and robot in underwater is expected to check of those parts. It is possible to check condition in bad environment. Otherwise, sonar data has much occlusion and clutter owing to flowing water stream and waves. So automatic filtering is difficult and often do it by human. In this paper, we propose new method of automatic filtering and mapping for such sonar data. To achieve this goal, we apply filtering method and object recognition method of point cloud processing. By using these method, we made new method of robust mapping system for underwater sonar. 1. Introduction 1.1 Related work Sonar and camera are used for measuring underwater. Sonar has bad accuracy and data density relatively to laser range finder. So it is hard to use sonar for mapping and navigation[1]. Particularly, it is hard to extract robust feature points[2]. So by using high performance IMU(Inertial Measuring Unit) and camera with sonar, they improve position accuracy[3,4]. In the case of using only sonar data, surface smoothing and ICP algorithm are used[5] from of old. Also, there are methods of mapping by making use of plane matching in river[6] and data spectrum to know data feature[7]. Recently, Aitor Aldoma made mapping system[8] with IMU and Graph SLAM, one of the most major SLAM (Simultaneous Localization and Mapping) method. As other method, Thomas Guerneve made mapping system[9] matching object to 3D CAD data. 1.2 Target data We measured a triplen and wall in water by moving sonar from right to left (refer with: Fig. 1), and use the data for mapping. We measure two stationary objects by moving sonar. 3D imaging sonar which we used has angle of view about 55 degrees in horizontal direction and about 40 degrees in vertical direction. The accuracy of the data is 26cm. In this measurement, we move sonar in the pool whose horizontal length is about 10m and depth is about 8m. The measured data has 3D coordinate called Point Cloud. In this measurement, the data has about 2000 points when sonar measure environment at one time. Measured data has much noise referring to Fig. 2 left side, wall has 40cm depth at most owing to bad accuracy. In this paper, we made mapping system only using 3D point cloud data.

2 Fig. 1 Measurement environment 2. Filtering Fig. 2 Raw sonar data (left side) and filtered data (right side) We implemented three filters by using PCL (Point Cloud Library)[10] which is 3D point processing library to filter noise from sonar data. 2.1 Statistical outlier removal filter Statistical outlier removal filter removes point which has large point-to-point distance by computing distribution of the distance statistically. In each point, the filter sum up point-to-point distances and compute average value and a standard deviation. To assume distance of points distribute under Gaussian distribution, the filter removes point which has the distance larger than sum of average and n times standard deviation. In this paper, we set n to 0, that is, removing points whose distance is larger than average. 2.2 Time filter Time filter removes unexpected points on time series by comparing former data and current data. It only leaves points of sequence object on time series. It extracts a coordinate of current point cloud

3 and search nearby points in former point cloud. And if the distance of search point to nearby point is larger than set value, filter remove the research point. 2.3 Smoothing filter In this measuring sonar data, sonar measure the surface of objects as thickness points. Thus we have to smooth the surface. In this paper, we smoothed surface using moving least-squares method [11] computing nearby points within set radius. Using above edited three filters, we processed the data of Fig. 2 left side and made filtered data like Fig. 2 right side. 3. Mapping Mapping is the process which concatenate two clouds by comparing current point cloud to former point cloud. It estimates transformation matrix which correspond the best to former cloud. Concatenating clouds sequentially, we expand the map data. To make mapping system, we apply 3D object recognition process (Fig. 3). Object recognition process has many methods in each step, so we think it is easily to compare many methods. Its process called pipeline. And there are two pipeline, local pipeline and global pipeline. 3.1 Local pipeline Fig. 3 Object recognition pipeline Local pipeline suits to noisily data, and use in segmentation which classify objects and 3D SLAM (Simultaneous Localization and Mapping) and so on. Shown in Fig. 3, local pipeline extract keypoints which have characteristic in whole points first. Second, it describes characteristic around the keypoint in each keypoints. As the method of description, there are process which use normal vector and position vector and so on. Next, it compares the description in former cloud (model cloud) and current cloud (scene cloud) and match the similar description to make correspondences. In correspondence Grouping phase, it estimates rough transformation matrix by taking geometric relationship of correspondences into account. After that, ICP (Iterative closest point) algorithm refine point cloud. In keypoint extraction phase, repeatability is the most important point. According to Silvio and Samuele, ISS (Intrinsic Shape Signature)[12] is the most robust method[13,14] against noise and rotation and changing scale. 3.2 Global pipeline Global pipeline needs perfect surface information such as 3D CAD data. However, it has much good point such as low memory consumption and more descriptive even in object with poor

4 geometric structure[15]. In this pipeline, description is done in each object, so segmentation step is the first step. After that, description is done in each object and matched. And it estimates rough rotation matrix by matching teaching data. Translation matrix is estimated by corresponding the centroids of objects in scene and model point cloud. This pipeline needs prerequisite that detect known object because we have to prepare teaching data. 3.3 ICP algorithm ICP algorithm is the estimation which don t take into account object geometric characteristics. At first, it makes tentative correspondences by computing distances between scene and model cloud and determining a pairs which have the shortest distance. After that, it estimates the transformation matrix to minimize the sum of distances between the correspondences. Repeating this process, estimate matrix which can match two point clouds. ICP algorithm has various method of how to set distance such as using normal vector method and Generalized-ICP[16]. ICP algorithm needs initial alignment like local and global pipeline because it can fail into a local solution. 4. Implementation and result We implemented mapping system using local and global pipeline. We implemented local pipeline first. However, the positions of Model cloud and scene cloud didn t match in keypoint extraction phase. Using ISS and SIFT (Scale Invariant Feature Transform)[17], NARF (Normal Aligned Radial Feature)[18], Uniform Sampling, Harris[19], but all method couldn t extract match keypoints. We thought this problem cause due to noisy and low density sonar data. Such data can t always detect same point of object, so out of position cause the change of local characteristics. We tried to do after pipeline steps with various methods, but estimated transformation matrix has clearly wrong values such as too big translation matrix in measuring field. There were too many times fails of estimation, we concluded that local pipeline can t use in sonar mapping. Global pipeline can t make teaching data beforehand in this mapping system. So we segment objects and match each objects between scene and model cloud. The coordinate of triplen move in time series, but wall position doesn t move. So we extracted triplen as landmark object and match it in time series. We made process pipeline that it extracts landmark object and match it. After that, final alignment is done by ICP algorithm. To extract landmark object, we use Euclidean segmentation method and count the number of each object. In the condition which has many objects, we think other method like region growing is valid, but in this data, it doesn t need. If the point number of cloud and centroid position are similar in time series, we reckon them as same object. As initial alignment of ICP, we use centroid position of landmark object. After that, we use ICP algorithm taking distance as Euclidean distance. In other ICP methods whose distance is like planeto-plane, estimated transformation matrix has big rotation and scale changing value. Fig. 4 is the result of connecting cloud by initial alignment and ICP with Euclidean distance. As a comparison of above map data, we show another map data without initial alignment(fig. 4). In this map data, the portion of wall is accorded, but triplen data is scattered. We thought correspondences of triplen didn t minimize because there were far more than correspondences in wall and the sum of distances were minimized mostly in portion of wall. If it tries to minimize the distances of correspondences in triplen portion, the sum of distances increase because the alignment in wall portion which has many correspondences would be bad condition. In estimation of ICP without initial alignment, alignment of object which has many points has significant impact to whole estimation. We made mapping system without description of objects which uses only coordinate of points. In general, pointwise description is simple and low memory consumption, but not robust to noise,

5 Fig. 4 Map data (left side: aligning landmark object before ICP algorithm with Euclidean distance, right side: no aligning before ICP algorithm) and often not descriptive enough[20]. However, considering failure of keypoint extraction and description in local pipeline and result of ICP algorithm which take distances as plane-to-plane, we shouldn t take method which use normal vector and position vector in low resolution data such as sonar data. We will implement these systems on ROS (Robot Operating System) as independent thread. Each thread is asynchronous process and it can realize real-time process. 5. Conclusion We made robust mapping system assuming that using imaging sonar which search limited range. We remove noise data by statistical outlier removal filter and time filter and smoothing filter. We suggested the method which extract landmark object and correspond it in initial alignment and use ICP algorithm. We use Euclidean segmentation as extraction of landmark object. In initial alignment, we use coordinates of centroid of landmark object and estimate translation matrix. In ICP algorithm, the result of method minimizing point to point distance was the best. If we make mapping system in sonar, method using only pointwise description is good. In this paper, we made filtering and mapping system, but improving processing speed and implementation of localization aren t done. In the future, we want to add these functionalities in the system. References [1] J. Tardós, J. Neira, P. Newman, and J. Leonard, Robust Mapping and Localization in Indoor Environments Using Sonar Data, International Journal of Robotics Research, Vol21, pp , [2] S. Majumder, J. Rosenblatt, S. Scheding, and H. Durrant-Whyte, Map building and localization for underwater navigation, In Lecture notes in control and information sciences, Vol271, pp , [3] S. Majumder, J. Rosenblatt, S. Scheding, and H. Durrant-Whyte, Map building and localization for underwater navigation, In Lecture notes in control and information sciences, Vol271, pp , 2001.

6 [4] M. Hammond, and S. Rock, A SLAM-based Approach for Underwater Mapping using AUVs with Poor Inertial Information, Proceedings of the IEEE/OES AUV Conference, [5] V. Murino, A. Fusiello, N. Iuretigh, and E. Puppo, 3D mosaicing for environment reconstruction Pattern recognition, IEEE Pattern Recognition, Proceedings. 15th International Conference on, Vol3, pp , [6] K. Pathak, A. Birk, and N. Vaskevicius, Plane-based registration of sonar data for underwater 3D mapping, Intelligent Robots and Systems (IROS) 2010 IEEE/RSJ International Conference on, [7] H. Bülow, and A. Birk, Spectral registration of noisy sonar data for underwater 3D mapping, Autonomous Robots, Vol30, pp , [8] A. Aldoma, Z. Marton, F. Tombari, W. Wohlkinger, C. Potthast, B. Zeisl, R. Rusu, S. Gedikli, and M. Vincze, Point Cloud Library Three-Dimensional Object Recognition and 6 DoF Pose Estimation, IEEE Robotics & Automation Magazine, Vol19, pp.80-91, [9] T. Guerneve, K. Subr, and Y. Petillot, Underwater 3D Structures as Semantic Landmarks in SONAR Mapping, Intelligent Robots and Systems, [10] R. Rusu, and S. Cousins, 3D is here: Point Cloud Library (PCL), Robotics and Automation (ICRA), 2011 IEEE International Conference, pp.1-4, 2011 [11] M. Alexa, J. Behr, D. Cohen-Or, S. Fleishman, D. Levin, and C. Silva, Computing and rendering point set surfaces, IEEE Transactions on Visualization and Computer Graphics, Vol9, pp.3-15, [12] Y. Zhong, Intrinsic Shape Signatures: A Shape Descriptor for 3D Object Recognition, Computer Vision Workshops (ICCV Workshops), [13] S. Filipe, and L. Alexandre, A comparative evaluation of 3D keypoint detectors in a RGB-D object dataset, In 9th International Conference on Computer Vision Theory and Applications, pp. 1 8, [14] S. Salti, F. Tombari, and L. Stefano, A Performance Evaluation of 3D Keypoint Detectors, 2011 International Conference on 3D Imaging, Modeling, Processing, Visualization and Transmission, pp , [15] F. Tombari, Keypoints and features, CGLibs Conference in Pisa, pp , [16] A. Segal, D. Haehnel, and S. Thrun, Generalized-ICP, in Robotics: Science and Systems, [17] D. Lowe, Distinctive image features from scale-invariant keypoints, International Journal of Computer Vision, Vol60, pp , [18] B. Steder, R. B. Rusu, and K. Konolige, Point Feature Extraction on 3D Range Scans Taking into Account Object Boundaries, Robotics and Automation (ICRA) IEEE International Conference, [19] C. Harris, and M. Stephens, A combined corner and edge detector, Alvey Vision Conferenfce, pp , [20] F. Tombari, Keypoints and features, CGLibs Conference in Pisa, pp , 2013.

3D object recognition used by team robotto

3D object recognition used by team robotto 3D object recognition used by team robotto Workshop Juliane Hoebel February 1, 2016 Faculty of Computer Science, Otto-von-Guericke University Magdeburg Content 1. Introduction 2. Depth sensor 3. 3D object

More information

Using Augmented Measurements to Improve the Convergence of ICP. Jacopo Serafin and Giorgio Grisetti

Using Augmented Measurements to Improve the Convergence of ICP. Jacopo Serafin and Giorgio Grisetti Jacopo Serafin and Giorgio Grisetti Point Cloud Registration We want to find the rotation and the translation that maximize the overlap between two point clouds Page 2 Point Cloud Registration We want

More information

LOCAL AND GLOBAL DESCRIPTORS FOR PLACE RECOGNITION IN ROBOTICS

LOCAL AND GLOBAL DESCRIPTORS FOR PLACE RECOGNITION IN ROBOTICS 8th International DAAAM Baltic Conference "INDUSTRIAL ENGINEERING - 19-21 April 2012, Tallinn, Estonia LOCAL AND GLOBAL DESCRIPTORS FOR PLACE RECOGNITION IN ROBOTICS Shvarts, D. & Tamre, M. Abstract: The

More information

3D Environment Reconstruction

3D Environment Reconstruction 3D Environment Reconstruction Using Modified Color ICP Algorithm by Fusion of a Camera and a 3D Laser Range Finder The 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems October 11-15,

More information

Robotics Programming Laboratory

Robotics Programming Laboratory Chair of Software Engineering Robotics Programming Laboratory Bertrand Meyer Jiwon Shin Lecture 8: Robot Perception Perception http://pascallin.ecs.soton.ac.uk/challenges/voc/databases.html#caltech car

More information

Structured Light II. Thanks to Ronen Gvili, Szymon Rusinkiewicz and Maks Ovsjanikov

Structured Light II. Thanks to Ronen Gvili, Szymon Rusinkiewicz and Maks Ovsjanikov Structured Light II Johannes Köhler Johannes.koehler@dfki.de Thanks to Ronen Gvili, Szymon Rusinkiewicz and Maks Ovsjanikov Introduction Previous lecture: Structured Light I Active Scanning Camera/emitter

More information

A descriptor and voting scheme for fast 3D self-localization in man-made environments

A descriptor and voting scheme for fast 3D self-localization in man-made environments A descriptor and voting scheme for fast 3D self-localization in man-made environments Marvin Gordon, Marcus Hebel and Michael Arens Fraunhofer Institute of Optronics, System Technologies and Image Exploitation

More information

Accurate Motion Estimation and High-Precision 3D Reconstruction by Sensor Fusion

Accurate Motion Estimation and High-Precision 3D Reconstruction by Sensor Fusion 007 IEEE International Conference on Robotics and Automation Roma, Italy, 0-4 April 007 FrE5. Accurate Motion Estimation and High-Precision D Reconstruction by Sensor Fusion Yunsu Bok, Youngbae Hwang,

More information

A Comparative Evaluation of 3D Keypoint Detectors in a RGB-D Object Dataset

A Comparative Evaluation of 3D Keypoint Detectors in a RGB-D Object Dataset A Comparative Evaluation of 3D Keypoint Detectors in a RGB-D Object Dataset Sílvio Filipe PhD Student, Informatics, IT - Instituto de Telecomunicações / University of Beira Interior, Department of Informatics,

More information

CSC494: Individual Project in Computer Science

CSC494: Individual Project in Computer Science CSC494: Individual Project in Computer Science Seyed Kamyar Seyed Ghasemipour Department of Computer Science University of Toronto kamyar@cs.toronto.edu Abstract One of the most important tasks in computer

More information

CRF Based Point Cloud Segmentation Jonathan Nation

CRF Based Point Cloud Segmentation Jonathan Nation CRF Based Point Cloud Segmentation Jonathan Nation jsnation@stanford.edu 1. INTRODUCTION The goal of the project is to use the recently proposed fully connected conditional random field (CRF) model to

More information

PRECEDING VEHICLE TRACKING IN STEREO IMAGES VIA 3D FEATURE MATCHING

PRECEDING VEHICLE TRACKING IN STEREO IMAGES VIA 3D FEATURE MATCHING PRECEDING VEHICLE TRACKING IN STEREO IMAGES VIA 3D FEATURE MATCHING Daniel Weingerl, Wilfried Kubinger, Corinna Engelhardt-Nowitzki UAS Technikum Wien: Department for Advanced Engineering Technologies,

More information

Real-Time RGB-D Registration and Mapping in Texture-less Environments Using Ranked Order Statistics

Real-Time RGB-D Registration and Mapping in Texture-less Environments Using Ranked Order Statistics Real-Time RGB-D Registration and Mapping in Texture-less Environments Using Ranked Order Statistics Khalid Yousif 1, Alireza Bab-Hadiashar 2, Senior Member, IEEE and Reza Hoseinnezhad 3 Abstract In this

More information

Ensemble of Bayesian Filters for Loop Closure Detection

Ensemble of Bayesian Filters for Loop Closure Detection Ensemble of Bayesian Filters for Loop Closure Detection Mohammad Omar Salameh, Azizi Abdullah, Shahnorbanun Sahran Pattern Recognition Research Group Center for Artificial Intelligence Faculty of Information

More information

3D Computer Vision. Structured Light II. Prof. Didier Stricker. Kaiserlautern University.

3D Computer Vision. Structured Light II. Prof. Didier Stricker. Kaiserlautern University. 3D Computer Vision Structured Light II Prof. Didier Stricker Kaiserlautern University http://ags.cs.uni-kl.de/ DFKI Deutsches Forschungszentrum für Künstliche Intelligenz http://av.dfki.de 1 Introduction

More information

Robotics. Lecture 7: Simultaneous Localisation and Mapping (SLAM)

Robotics. Lecture 7: Simultaneous Localisation and Mapping (SLAM) Robotics Lecture 7: Simultaneous Localisation and Mapping (SLAM) See course website http://www.doc.ic.ac.uk/~ajd/robotics/ for up to date information. Andrew Davison Department of Computing Imperial College

More information

Vision-based Mobile Robot Localization and Mapping using Scale-Invariant Features

Vision-based Mobile Robot Localization and Mapping using Scale-Invariant Features Vision-based Mobile Robot Localization and Mapping using Scale-Invariant Features Stephen Se, David Lowe, Jim Little Department of Computer Science University of British Columbia Presented by Adam Bickett

More information

arxiv: v2 [cs.ro] 19 Sep 2016

arxiv: v2 [cs.ro] 19 Sep 2016 : A Novel Rotation, Illumination, Scale Invariant Appearance and Shape Feature Kanzhi Wu, Xiaoyang Li 2, Ravindra Ranasinghe, Gamini Dissanayake and Yong Liu 2 arxiv:6.44v2 [cs.ro] 9 Sep 26 Abstract This

More information

Estimating Geospatial Trajectory of a Moving Camera

Estimating Geospatial Trajectory of a Moving Camera Estimating Geospatial Trajectory of a Moving Camera Asaad Hakeem 1, Roberto Vezzani 2, Mubarak Shah 1, Rita Cucchiara 2 1 School of Electrical Engineering and Computer Science, University of Central Florida,

More information

Structured light 3D reconstruction

Structured light 3D reconstruction Structured light 3D reconstruction Reconstruction pipeline and industrial applications rodola@dsi.unive.it 11/05/2010 3D Reconstruction 3D reconstruction is the process of capturing the shape and appearance

More information

A NEW FEATURE BASED IMAGE REGISTRATION ALGORITHM INTRODUCTION

A NEW FEATURE BASED IMAGE REGISTRATION ALGORITHM INTRODUCTION A NEW FEATURE BASED IMAGE REGISTRATION ALGORITHM Karthik Krish Stuart Heinrich Wesley E. Snyder Halil Cakir Siamak Khorram North Carolina State University Raleigh, 27695 kkrish@ncsu.edu sbheinri@ncsu.edu

More information

Fast and Robust 3D Feature Extraction from Sparse Point Clouds

Fast and Robust 3D Feature Extraction from Sparse Point Clouds Fast and Robust 3D Feature Extraction from Sparse Point Clouds Jacopo Serafin 1, Edwin Olson 2 and Giorgio Grisetti 1 Abstract Matching 3D point clouds, a critical operation in map building and localization,

More information

A Rapid Automatic Image Registration Method Based on Improved SIFT

A Rapid Automatic Image Registration Method Based on Improved SIFT Available online at www.sciencedirect.com Procedia Environmental Sciences 11 (2011) 85 91 A Rapid Automatic Image Registration Method Based on Improved SIFT Zhu Hongbo, Xu Xuejun, Wang Jing, Chen Xuesong,

More information

CS4670: Computer Vision

CS4670: Computer Vision CS4670: Computer Vision Noah Snavely Lecture 6: Feature matching and alignment Szeliski: Chapter 6.1 Reading Last time: Corners and blobs Scale-space blob detector: Example Feature descriptors We know

More information

Plane-Based Registration of Sonar Data for Underwater 3D Mapping

Plane-Based Registration of Sonar Data for Underwater 3D Mapping The 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems October 18-22, 2010, Taipei, Taiwan Plane-Based Registration of Sonar Data for Underwater 3D Mapping Kaustubh Pathak, Andreas

More information

Removing Moving Objects from Point Cloud Scenes

Removing Moving Objects from Point Cloud Scenes Removing Moving Objects from Point Cloud Scenes Krystof Litomisky and Bir Bhanu University of California, Riverside krystof@litomisky.com, bhanu@ee.ucr.edu Abstract. Three-dimensional simultaneous localization

More information

AR Cultural Heritage Reconstruction Based on Feature Landmark Database Constructed by Using Omnidirectional Range Sensor

AR Cultural Heritage Reconstruction Based on Feature Landmark Database Constructed by Using Omnidirectional Range Sensor AR Cultural Heritage Reconstruction Based on Feature Landmark Database Constructed by Using Omnidirectional Range Sensor Takafumi Taketomi, Tomokazu Sato, and Naokazu Yokoya Graduate School of Information

More information

Lecture 10 Detectors and descriptors

Lecture 10 Detectors and descriptors Lecture 10 Detectors and descriptors Properties of detectors Edge detectors Harris DoG Properties of detectors SIFT Shape context Silvio Savarese Lecture 10-26-Feb-14 From the 3D to 2D & vice versa P =

More information

Efficient Surface and Feature Estimation in RGBD

Efficient Surface and Feature Estimation in RGBD Efficient Surface and Feature Estimation in RGBD Zoltan-Csaba Marton, Dejan Pangercic, Michael Beetz Intelligent Autonomous Systems Group Technische Universität München RGB-D Workshop on 3D Perception

More information

Structured Light II. Thanks to Ronen Gvili, Szymon Rusinkiewicz and Maks Ovsjanikov

Structured Light II. Thanks to Ronen Gvili, Szymon Rusinkiewicz and Maks Ovsjanikov Structured Light II Johannes Köhler Johannes.koehler@dfki.de Thanks to Ronen Gvili, Szymon Rusinkiewicz and Maks Ovsjanikov Introduction Previous lecture: Structured Light I Active Scanning Camera/emitter

More information

A Real-Time RGB-D Registration and Mapping Approach by Heuristically Switching Between Photometric And Geometric Information

A Real-Time RGB-D Registration and Mapping Approach by Heuristically Switching Between Photometric And Geometric Information A Real-Time RGB-D Registration and Mapping Approach by Heuristically Switching Between Photometric And Geometric Information The 17th International Conference on Information Fusion (Fusion 2014) Khalid

More information

CS 378: Autonomous Intelligent Robotics. Instructor: Jivko Sinapov

CS 378: Autonomous Intelligent Robotics. Instructor: Jivko Sinapov CS 378: Autonomous Intelligent Robotics Instructor: Jivko Sinapov http://www.cs.utexas.edu/~jsinapov/teaching/cs378/ Visual Registration and Recognition Announcements Homework 6 is out, due 4/5 4/7 Installing

More information

Local features: detection and description. Local invariant features

Local features: detection and description. Local invariant features Local features: detection and description Local invariant features Detection of interest points Harris corner detection Scale invariant blob detection: LoG Description of local patches SIFT : Histograms

More information

Fast Sampling Plane Filtering, Polygon Construction and Merging from Depth Images

Fast Sampling Plane Filtering, Polygon Construction and Merging from Depth Images Fast Sampling Plane Filtering, Polygon Construction and Merging from Depth Images Joydeep Biswas Robotics Institute Carnegie Mellon University Pittsburgh, PA 523, USA joydeepb@ri.cmu.edu Manuela Veloso

More information

A New Approach For 3D Image Reconstruction From Multiple Images

A New Approach For 3D Image Reconstruction From Multiple Images International Journal of Electronics Engineering Research. ISSN 0975-6450 Volume 9, Number 4 (2017) pp. 569-574 Research India Publications http://www.ripublication.com A New Approach For 3D Image Reconstruction

More information

EE368 Project Report CD Cover Recognition Using Modified SIFT Algorithm

EE368 Project Report CD Cover Recognition Using Modified SIFT Algorithm EE368 Project Report CD Cover Recognition Using Modified SIFT Algorithm Group 1: Mina A. Makar Stanford University mamakar@stanford.edu Abstract In this report, we investigate the application of the Scale-Invariant

More information

Monocular SLAM for a Small-Size Humanoid Robot

Monocular SLAM for a Small-Size Humanoid Robot Tamkang Journal of Science and Engineering, Vol. 14, No. 2, pp. 123 129 (2011) 123 Monocular SLAM for a Small-Size Humanoid Robot Yin-Tien Wang*, Duen-Yan Hung and Sheng-Hsien Cheng Department of Mechanical

More information

CS 231A Computer Vision (Winter 2014) Problem Set 3

CS 231A Computer Vision (Winter 2014) Problem Set 3 CS 231A Computer Vision (Winter 2014) Problem Set 3 Due: Feb. 18 th, 2015 (11:59pm) 1 Single Object Recognition Via SIFT (45 points) In his 2004 SIFT paper, David Lowe demonstrates impressive object recognition

More information

Registration of Noisy Point Clouds using Virtual Interest Points

Registration of Noisy Point Clouds using Virtual Interest Points Registration of Noisy Point Clouds using Virtual Interest Points Mirza Tahir Ahmed, Mustafa Mohamad, Joshua A. Marshall, and Michael Greenspan Department of Electrical and Computer Engineering School of

More information

Efficient SLAM Scheme Based ICP Matching Algorithm Using Image and Laser Scan Information

Efficient SLAM Scheme Based ICP Matching Algorithm Using Image and Laser Scan Information Proceedings of the World Congress on Electrical Engineering and Computer Systems and Science (EECSS 2015) Barcelona, Spain July 13-14, 2015 Paper No. 335 Efficient SLAM Scheme Based ICP Matching Algorithm

More information

3D Perception. CS 4495 Computer Vision K. Hawkins. CS 4495 Computer Vision. 3D Perception. Kelsey Hawkins Robotics

3D Perception. CS 4495 Computer Vision K. Hawkins. CS 4495 Computer Vision. 3D Perception. Kelsey Hawkins Robotics CS 4495 Computer Vision Kelsey Hawkins Robotics Motivation What do animals, people, and robots want to do with vision? Detect and recognize objects/landmarks Find location of objects with respect to themselves

More information

Feature Matching and Robust Fitting

Feature Matching and Robust Fitting Feature Matching and Robust Fitting Computer Vision CS 143, Brown Read Szeliski 4.1 James Hays Acknowledgment: Many slides from Derek Hoiem and Grauman&Leibe 2008 AAAI Tutorial Project 2 questions? This

More information

Basics of Localization, Mapping and SLAM. Jari Saarinen Aalto University Department of Automation and systems Technology

Basics of Localization, Mapping and SLAM. Jari Saarinen Aalto University Department of Automation and systems Technology Basics of Localization, Mapping and SLAM Jari Saarinen Aalto University Department of Automation and systems Technology Content Introduction to Problem (s) Localization A few basic equations Dead Reckoning

More information

Image Processing. Image Features

Image Processing. Image Features Image Processing Image Features Preliminaries 2 What are Image Features? Anything. What they are used for? Some statements about image fragments (patches) recognition Search for similar patches matching

More information

Ensemble Learning for Object Recognition and Pose Estimation

Ensemble Learning for Object Recognition and Pose Estimation Ensemble Learning for Object Recognition and Pose Estimation German Aerospace Center (DLR) May 10, 2013 Outline 1. Introduction 2. Overview of Available Methods 3. Multi-cue Integration Complementing Modalities

More information

Robust Moving Least-squares Fitting with Sharp Features

Robust Moving Least-squares Fitting with Sharp Features Survey of Methods in Computer Graphics: Robust Moving Least-squares Fitting with Sharp Features S. Fleishman, D. Cohen-Or, C. T. Silva SIGGRAPH 2005. Method Pipeline Input: Point to project on the surface.

More information

Temporal Integration of Feature Correspondences For Enhanced Recognition in Cluttered And Dynamic Environments

Temporal Integration of Feature Correspondences For Enhanced Recognition in Cluttered And Dynamic Environments Temporal Integration of Feature Correspondences For Enhanced Recognition in Cluttered And Dynamic Environments Thomas Fäulhammer, Aitor Aldoma, Michael Zillich and Markus Vincze Abstract We propose a method

More information

Coarse-to-fine image registration

Coarse-to-fine image registration Today we will look at a few important topics in scale space in computer vision, in particular, coarseto-fine approaches, and the SIFT feature descriptor. I will present only the main ideas here to give

More information

SIFT: SCALE INVARIANT FEATURE TRANSFORM SURF: SPEEDED UP ROBUST FEATURES BASHAR ALSADIK EOS DEPT. TOPMAP M13 3D GEOINFORMATION FROM IMAGES 2014

SIFT: SCALE INVARIANT FEATURE TRANSFORM SURF: SPEEDED UP ROBUST FEATURES BASHAR ALSADIK EOS DEPT. TOPMAP M13 3D GEOINFORMATION FROM IMAGES 2014 SIFT: SCALE INVARIANT FEATURE TRANSFORM SURF: SPEEDED UP ROBUST FEATURES BASHAR ALSADIK EOS DEPT. TOPMAP M13 3D GEOINFORMATION FROM IMAGES 2014 SIFT SIFT: Scale Invariant Feature Transform; transform image

More information

10/03/11. Model Fitting. Computer Vision CS 143, Brown. James Hays. Slides from Silvio Savarese, Svetlana Lazebnik, and Derek Hoiem

10/03/11. Model Fitting. Computer Vision CS 143, Brown. James Hays. Slides from Silvio Savarese, Svetlana Lazebnik, and Derek Hoiem 10/03/11 Model Fitting Computer Vision CS 143, Brown James Hays Slides from Silvio Savarese, Svetlana Lazebnik, and Derek Hoiem Fitting: find the parameters of a model that best fit the data Alignment:

More information

Feature Based Registration - Image Alignment

Feature Based Registration - Image Alignment Feature Based Registration - Image Alignment Image Registration Image registration is the process of estimating an optimal transformation between two or more images. Many slides from Alexei Efros http://graphics.cs.cmu.edu/courses/15-463/2007_fall/463.html

More information

Object Reconstruction

Object Reconstruction B. Scholz Object Reconstruction 1 / 39 MIN-Fakultät Fachbereich Informatik Object Reconstruction Benjamin Scholz Universität Hamburg Fakultät für Mathematik, Informatik und Naturwissenschaften Fachbereich

More information

Features Points. Andrea Torsello DAIS Università Ca Foscari via Torino 155, Mestre (VE)

Features Points. Andrea Torsello DAIS Università Ca Foscari via Torino 155, Mestre (VE) Features Points Andrea Torsello DAIS Università Ca Foscari via Torino 155, 30172 Mestre (VE) Finding Corners Edge detectors perform poorly at corners. Corners provide repeatable points for matching, so

More information

Robot localization method based on visual features and their geometric relationship

Robot localization method based on visual features and their geometric relationship , pp.46-50 http://dx.doi.org/10.14257/astl.2015.85.11 Robot localization method based on visual features and their geometric relationship Sangyun Lee 1, Changkyung Eem 2, and Hyunki Hong 3 1 Department

More information

Personal Navigation and Indoor Mapping: Performance Characterization of Kinect Sensor-based Trajectory Recovery

Personal Navigation and Indoor Mapping: Performance Characterization of Kinect Sensor-based Trajectory Recovery Personal Navigation and Indoor Mapping: Performance Characterization of Kinect Sensor-based Trajectory Recovery 1 Charles TOTH, 1 Dorota BRZEZINSKA, USA 2 Allison KEALY, Australia, 3 Guenther RETSCHER,

More information

Building a Panorama. Matching features. Matching with Features. How do we build a panorama? Computational Photography, 6.882

Building a Panorama. Matching features. Matching with Features. How do we build a panorama? Computational Photography, 6.882 Matching features Building a Panorama Computational Photography, 6.88 Prof. Bill Freeman April 11, 006 Image and shape descriptors: Harris corner detectors and SIFT features. Suggested readings: Mikolajczyk

More information

COMPARISON OF 2D AND 3D APPROACHES FOR THE ALIGNMENT OF UAV AND LIDAR POINT CLOUDS

COMPARISON OF 2D AND 3D APPROACHES FOR THE ALIGNMENT OF UAV AND LIDAR POINT CLOUDS COMPARISON OF 2D AND 3D APPROACHES FOR THE ALIGNMENT OF UAV AND LIDAR POINT CLOUDS Ravi Ancil Persad*, Costas Armenakis Geomatics Engineering, GeoICT Lab Department of Earth and Space Science and Engineering

More information

SLAM with Object Discovery, Modeling and Mapping

SLAM with Object Discovery, Modeling and Mapping SLAM with Object Discovery, Modeling and Mapping Siddharth Choudhary, Alexander J. B. Trevor, Henrik I. Christensen, Frank Dellaert Abstract Object discovery and modeling have been widely studied in the

More information

Chapter 3 Image Registration. Chapter 3 Image Registration

Chapter 3 Image Registration. Chapter 3 Image Registration Chapter 3 Image Registration Distributed Algorithms for Introduction (1) Definition: Image Registration Input: 2 images of the same scene but taken from different perspectives Goal: Identify transformation

More information

3D Photography: Active Ranging, Structured Light, ICP

3D Photography: Active Ranging, Structured Light, ICP 3D Photography: Active Ranging, Structured Light, ICP Kalin Kolev, Marc Pollefeys Spring 2013 http://cvg.ethz.ch/teaching/2013spring/3dphoto/ Schedule (tentative) Feb 18 Feb 25 Mar 4 Mar 11 Mar 18 Mar

More information

Local features: detection and description May 12 th, 2015

Local features: detection and description May 12 th, 2015 Local features: detection and description May 12 th, 2015 Yong Jae Lee UC Davis Announcements PS1 grades up on SmartSite PS1 stats: Mean: 83.26 Standard Dev: 28.51 PS2 deadline extended to Saturday, 11:59

More information

L17. OCCUPANCY MAPS. NA568 Mobile Robotics: Methods & Algorithms

L17. OCCUPANCY MAPS. NA568 Mobile Robotics: Methods & Algorithms L17. OCCUPANCY MAPS NA568 Mobile Robotics: Methods & Algorithms Today s Topic Why Occupancy Maps? Bayes Binary Filters Log-odds Occupancy Maps Inverse sensor model Learning inverse sensor model ML map

More information

3D Reconstruction of a Hopkins Landmark

3D Reconstruction of a Hopkins Landmark 3D Reconstruction of a Hopkins Landmark Ayushi Sinha (461), Hau Sze (461), Diane Duros (361) Abstract - This paper outlines a method for 3D reconstruction from two images. Our procedure is based on known

More information

Visual Recognition and Search April 18, 2008 Joo Hyun Kim

Visual Recognition and Search April 18, 2008 Joo Hyun Kim Visual Recognition and Search April 18, 2008 Joo Hyun Kim Introduction Suppose a stranger in downtown with a tour guide book?? Austin, TX 2 Introduction Look at guide What s this? Found Name of place Where

More information

Automatic Pipeline Generation by the Sequential Segmentation and Skelton Construction of Point Cloud

Automatic Pipeline Generation by the Sequential Segmentation and Skelton Construction of Point Cloud , pp.43-47 http://dx.doi.org/10.14257/astl.2014.67.11 Automatic Pipeline Generation by the Sequential Segmentation and Skelton Construction of Point Cloud Ashok Kumar Patil, Seong Sill Park, Pavitra Holi,

More information

Surface Registration. Gianpaolo Palma

Surface Registration. Gianpaolo Palma Surface Registration Gianpaolo Palma The problem 3D scanning generates multiple range images Each contain 3D points for different parts of the model in the local coordinates of the scanner Find a rigid

More information

3D Photography: Stereo

3D Photography: Stereo 3D Photography: Stereo Marc Pollefeys, Torsten Sattler Spring 2016 http://www.cvg.ethz.ch/teaching/3dvision/ 3D Modeling with Depth Sensors Today s class Obtaining depth maps / range images unstructured

More information

Point Clouds Registration with Probabilistic Data Association

Point Clouds Registration with Probabilistic Data Association 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) Daejeon Convention Center October 9-14, 2016, Daejeon, Korea Point Clouds Registration with Probabilistic Data Association

More information

AUTOMATIC RECTIFICATION OF SIDE-SCAN SONAR IMAGES

AUTOMATIC RECTIFICATION OF SIDE-SCAN SONAR IMAGES Proceedings of the International Conference Underwater Acoustic Measurements: Technologies &Results Heraklion, Crete, Greece, 28 th June 1 st July 2005 AUTOMATIC RECTIFICATION OF SIDE-SCAN SONAR IMAGES

More information

The SIFT (Scale Invariant Feature

The SIFT (Scale Invariant Feature The SIFT (Scale Invariant Feature Transform) Detector and Descriptor developed by David Lowe University of British Columbia Initial paper ICCV 1999 Newer journal paper IJCV 2004 Review: Matt Brown s Canonical

More information

Incremental Structured ICP Algorithm

Incremental Structured ICP Algorithm Incremental Structured ICP Algorithm Haokun Geng, Johnny Chien, Radu Nicolescu, and Reinhard Klette The.enpeda.. Project, Tamaki Campus The University of Auckland, New Zealand Abstract. Variants of the

More information

3D Models from Range Sensors. Gianpaolo Palma

3D Models from Range Sensors. Gianpaolo Palma 3D Models from Range Sensors Gianpaolo Palma Who Gianpaolo Palma Researcher at Visual Computing Laboratory (ISTI-CNR) Expertise: 3D scanning, Mesh Processing, Computer Graphics E-mail: gianpaolo.palma@isti.cnr.it

More information

Rigid ICP registration with Kinect

Rigid ICP registration with Kinect Rigid ICP registration with Kinect Students: Yoni Choukroun, Elie Semmel Advisor: Yonathan Aflalo 1 Overview.p.3 Development of the project..p.3 Papers p.4 Project algorithm..p.6 Result of the whole body.p.7

More information

Panoramic Image Stitching

Panoramic Image Stitching Mcgill University Panoramic Image Stitching by Kai Wang Pengbo Li A report submitted in fulfillment for the COMP 558 Final project in the Faculty of Computer Science April 2013 Mcgill University Abstract

More information

Robot Localization based on Geo-referenced Images and G raphic Methods

Robot Localization based on Geo-referenced Images and G raphic Methods Robot Localization based on Geo-referenced Images and G raphic Methods Sid Ahmed Berrabah Mechanical Department, Royal Military School, Belgium, sidahmed.berrabah@rma.ac.be Janusz Bedkowski, Łukasz Lubasiński,

More information

IRON: A Fast Interest Point Descriptor for Robust NDT-Map Matching and its Application to Robot Localization*

IRON: A Fast Interest Point Descriptor for Robust NDT-Map Matching and its Application to Robot Localization* IRON: A Fast Interest Point Descriptor for Robust NDT-Map Matching and its Application to Robot Localization* Thomas Schmiedel, Erik Einhorn and Horst-Michael Gross 1 Abstract This work introduces the

More information

Designing Applications that See Lecture 7: Object Recognition

Designing Applications that See Lecture 7: Object Recognition stanford hci group / cs377s Designing Applications that See Lecture 7: Object Recognition Dan Maynes-Aminzade 29 January 2008 Designing Applications that See http://cs377s.stanford.edu Reminders Pick up

More information

Intensity Augmented ICP for Registration of Laser Scanner Point Clouds

Intensity Augmented ICP for Registration of Laser Scanner Point Clouds Intensity Augmented ICP for Registration of Laser Scanner Point Clouds Bharat Lohani* and Sandeep Sashidharan *Department of Civil Engineering, IIT Kanpur Email: blohani@iitk.ac.in. Abstract While using

More information

A Method of Annotation Extraction from Paper Documents Using Alignment Based on Local Arrangements of Feature Points

A Method of Annotation Extraction from Paper Documents Using Alignment Based on Local Arrangements of Feature Points A Method of Annotation Extraction from Paper Documents Using Alignment Based on Local Arrangements of Feature Points Tomohiro Nakai, Koichi Kise, Masakazu Iwamura Graduate School of Engineering, Osaka

More information

Srikumar Ramalingam. Review. 3D Reconstruction. Pose Estimation Revisited. School of Computing University of Utah

Srikumar Ramalingam. Review. 3D Reconstruction. Pose Estimation Revisited. School of Computing University of Utah School of Computing University of Utah Presentation Outline 1 2 3 Forward Projection (Reminder) u v 1 KR ( I t ) X m Y m Z m 1 Backward Projection (Reminder) Q K 1 q Presentation Outline 1 2 3 Sample Problem

More information

Recognition of Objects and Places in 3D Point Clouds for Robotic Applications

Recognition of Objects and Places in 3D Point Clouds for Robotic Applications Recognition of Objects and Places in 3D Point Clouds for Robotic Applications Robert Cupec, Emmanuel Karlo Nyarko, Damir Filko, Luka Markasović Faculty of Electrical Engineering Osijek Josip Juraj Strossmayer

More information

Octree-Based Obstacle Representation and Registration for Real-Time

Octree-Based Obstacle Representation and Registration for Real-Time Octree-Based Obstacle Representation and Registration for Real-Time Jaewoong Kim, Daesik Kim, Junghyun Seo, Sukhan Lee and Yeonchool Park* Intelligent System Research Center (ISRC) & Nano and Intelligent

More information

Feature extraction for underwater visual SLAM

Feature extraction for underwater visual SLAM Feature extraction for underwater visual SLAM Josep Aulinas, Marc Carreras, Xavier Llado Joaquim Salvi, Rafael Garcia and Ricard Prados Computer Vision and Robotics Group Institute of Informatics and Applications

More information

Global localization from a single feature correspondence

Global localization from a single feature correspondence Global localization from a single feature correspondence Friedrich Fraundorfer and Horst Bischof Institute for Computer Graphics and Vision Graz University of Technology {fraunfri,bischof}@icg.tu-graz.ac.at

More information

Patch-based Object Recognition. Basic Idea

Patch-based Object Recognition. Basic Idea Patch-based Object Recognition 1! Basic Idea Determine interest points in image Determine local image properties around interest points Use local image properties for object classification Example: Interest

More information

arxiv: v1 [cs.ro] 14 Mar 2016

arxiv: v1 [cs.ro] 14 Mar 2016 : A Novel Rotation, Illumination, Scale Invariant Appearance and Shape Feature Xiaoyang Li, Kanzhi Wu 2, Yong Liu, Ravindra Ranasinghe 2, Gamini Dissanayake 2 and Rong Xiong arxiv:6.44v [cs.ro] 4 Mar 26

More information

Geometrical Feature Extraction Using 2D Range Scanner

Geometrical Feature Extraction Using 2D Range Scanner Geometrical Feature Extraction Using 2D Range Scanner Sen Zhang Lihua Xie Martin Adams Fan Tang BLK S2, School of Electrical and Electronic Engineering Nanyang Technological University, Singapore 639798

More information

Motion Estimation and Optical Flow Tracking

Motion Estimation and Optical Flow Tracking Image Matching Image Retrieval Object Recognition Motion Estimation and Optical Flow Tracking Example: Mosiacing (Panorama) M. Brown and D. G. Lowe. Recognising Panoramas. ICCV 2003 Example 3D Reconstruction

More information

Automatic Image Alignment (feature-based)

Automatic Image Alignment (feature-based) Automatic Image Alignment (feature-based) Mike Nese with a lot of slides stolen from Steve Seitz and Rick Szeliski 15-463: Computational Photography Alexei Efros, CMU, Fall 2006 Today s lecture Feature

More information

Corner Detection. GV12/3072 Image Processing.

Corner Detection. GV12/3072 Image Processing. Corner Detection 1 Last Week 2 Outline Corners and point features Moravec operator Image structure tensor Harris corner detector Sub-pixel accuracy SUSAN FAST Example descriptor: SIFT 3 Point Features

More information

Matching Evaluation of 2D Laser Scan Points using Observed Probability in Unstable Measurement Environment

Matching Evaluation of 2D Laser Scan Points using Observed Probability in Unstable Measurement Environment Matching Evaluation of D Laser Scan Points using Observed Probability in Unstable Measurement Environment Taichi Yamada, and Akihisa Ohya Abstract In the real environment such as urban areas sidewalk,

More information

URBAN STRUCTURE ESTIMATION USING PARALLEL AND ORTHOGONAL LINES

URBAN STRUCTURE ESTIMATION USING PARALLEL AND ORTHOGONAL LINES URBAN STRUCTURE ESTIMATION USING PARALLEL AND ORTHOGONAL LINES An Undergraduate Research Scholars Thesis by RUI LIU Submitted to Honors and Undergraduate Research Texas A&M University in partial fulfillment

More information

Alignment of Centimeter Scale Bathymetry using Six Degrees of Freedom

Alignment of Centimeter Scale Bathymetry using Six Degrees of Freedom Alignment of Centimeter Scale Bathymetry using Six Degrees of Freedom Ethan Slattery, University of California Santa Cruz Mentors: David Caress Summer 2018 Keywords: point-clouds, iterative closest point,

More information

Implicit Shape Models For 3D Shape Classification With a Continuous Voting Space

Implicit Shape Models For 3D Shape Classification With a Continuous Voting Space Implicit Shape Models For 3D Shape Classification With a Continuous Voting Space Viktor Seib, Norman Link and Dietrich Paulus Active Vision Group (AGAS), University of Koblenz-Landau, Universitätsstr.

More information

Image Features: Detection, Description, and Matching and their Applications

Image Features: Detection, Description, and Matching and their Applications Image Features: Detection, Description, and Matching and their Applications Image Representation: Global Versus Local Features Features/ keypoints/ interset points are interesting locations in the image.

More information

Registration of Low Cost Maps within Large Scale MMS Maps

Registration of Low Cost Maps within Large Scale MMS Maps 2016 IEEE 18th International Conference on High Performance Computing and Communications; IEEE 14th International Conference on Smart City; IEEE 2nd International Conference on Data Science and Systems

More information

CS5670: Computer Vision

CS5670: Computer Vision CS5670: Computer Vision Noah Snavely Lecture 4: Harris corner detection Szeliski: 4.1 Reading Announcements Project 1 (Hybrid Images) code due next Wednesday, Feb 14, by 11:59pm Artifacts due Friday, Feb

More information

Visual Tracking (1) Tracking of Feature Points and Planar Rigid Objects

Visual Tracking (1) Tracking of Feature Points and Planar Rigid Objects Intelligent Control Systems Visual Tracking (1) Tracking of Feature Points and Planar Rigid Objects Shingo Kagami Graduate School of Information Sciences, Tohoku University swk(at)ic.is.tohoku.ac.jp http://www.ic.is.tohoku.ac.jp/ja/swk/

More information

Robot Mapping for Rescue Robots

Robot Mapping for Rescue Robots Robot Mapping for Rescue Robots Nagesh Adluru Temple University Philadelphia, PA, USA nagesh@temple.edu Longin Jan Latecki Temple University Philadelphia, PA, USA latecki@temple.edu Rolf Lakaemper Temple

More information

Robot Autonomy Final Report : Team Husky

Robot Autonomy Final Report : Team Husky Robot Autonomy Final Report : Team Husky 1 Amit Bansal Master student in Robotics Institute, CMU ab1@andrew.cmu.edu Akshay Hinduja Master student in Mechanical Engineering, CMU ahinduja@andrew.cmu.edu

More information