Intraoral Laser Scan Data와 Dental CT 영상 접합 및 개별 치아 분할

Slides:



Advertisements
Similar presentations
Image Registration  Mapping of Evolution. Registration Goals Assume the correspondences are known Find such f() and g() such that the images are best.
Advertisements

The fundamental matrix F
CSE554Extrinsic DeformationsSlide 1 CSE 554 Lecture 9: Extrinsic Deformations Fall 2012.
MASKS © 2004 Invitation to 3D vision Lecture 7 Step-by-Step Model Buidling.
Automatic 3D Face Recognition System 邱彥霖 龔士傑 Biometric Authentication.
Automatic Feature Extraction for Multi-view 3D Face Recognition
Automatic in vivo Microscopy Video Mining for Leukocytes * Chengcui Zhang, Wei-Bang Chen, Lin Yang, Xin Chen, John K. Johnstone.
Seoul National University Movement Research Lab. Seoul National University Bundang Hospital Jiwon Jeong.
Special Topic on Image Retrieval Local Feature Matching Verification.
Image alignment Image from
Semi-automatic Range to Range Registration: A Feature-based Method Chao Chen & Ioannis Stamos Computer Science Department Graduate Center, Hunter College.
A Versatile Depalletizer of Boxes Based on Range Imagery Dimitrios Katsoulas*, Lothar Bergen*, Lambis Tassakos** *University of Freiburg **Inos Automation-software.
Reverse Engineering Niloy J. Mitra.
Registration of two scanned range images using k-d tree accelerated ICP algorithm By Xiaodong Yan Dec
자동차 보험 고객 관리 프로그램 학 과: 정보통신공학과 학 번: 이 름: 김민형 지도교수: 신영석
A Computationally Efficient Approach for 2D-3D Image Registration Juri Minxha Medical Image Analysis Professor Benjamin Kimia Spring 2011 Brown University.
Mosaics con’t CSE 455, Winter 2010 February 10, 2010.
CENG 789 – Digital Geometry Processing 06- Rigid-Body Alignment Asst. Prof. Yusuf Sahillioğlu Computer Eng. Dept,, Turkey.
Multiple View Geometry
Iterative closest point algorithms
CS CS 175 – Week 2 Processing Point Clouds Registration.
Curve Analogies Aaron Hertzmann Nuria Oliver Brain Curless Steven M. Seitz University of Washington Microsoft Research Thirteenth Eurographics.
Matching and Recognition in 3D. Moving from 2D to 3D Some things harderSome things harder – Rigid transform has 6 degrees of freedom vs. 3 – No natural.
Stereo Computation using Iterative Graph-Cuts
Automatic Image Alignment (feature-based) : Computational Photography Alexei Efros, CMU, Fall 2006 with a lot of slides stolen from Steve Seitz and.
3D full object reconstruction from kinect Yoni Choukroun Elie Semmel Advisor: Yonathan Afflalo.
Lecture 10: Robust fitting CS4670: Computer Vision Noah Snavely.
Accurate, Dense and Robust Multi-View Stereopsis Yasutaka Furukawa and Jean Ponce Presented by Rahul Garg and Ryan Kaminsky.
Chapter 6 Feature-based alignment Advanced Computer Vision.
CSE554AlignmentSlide 1 CSE 554 Lecture 8: Alignment Fall 2014.
CSE554Laplacian DeformationSlide 1 CSE 554 Lecture 8: Laplacian Deformation Fall 2012.
 발표 날짜  팀명  팀원 이름 Project1 Fall, 2013; Midterm.
Multimodal Interaction Dr. Mike Spann
Computer Architecture Project #2 Cache Simulator
Advanced Computer Vision Feature-based Alignment Lecturer: Lu Yi & Prof. Fuh CSIE NTU.
Many-SC 아키텍처 기반 OpenCL 프레임워크상에서의 응용프로그램 구현 및 최적화
Medical Imaging Dr. Mohammad Dawood Department of Computer Science University of Münster Germany.
Example: line fitting. n=2 Model fitting Measure distances.
Feature-Based Stereo Matching Using Graph Cuts Gorkem Saygili, Laurens van der Maaten, Emile A. Hendriks ASCI Conference 2011.
Medical Image Analysis Image Registration Figures come from the textbook: Medical Image Analysis, by Atam P. Dhawan, IEEE Press, 2003.
Copyright © 2010 Siemens Medical Solutions USA, Inc. All rights reserved. Hierarchical Segmentation and Identification of Thoracic Vertebra Using Learning-based.
Computer Vision, Robert Pless
A Region Based Stereo Matching Algorithm Using Cooperative Optimization Zeng-Fu Wang, Zhi-Gang Zheng University of Science and Technology of China Computer.
Parameter estimation. 2D homography Given a set of (x i,x i ’), compute H (x i ’=Hx i ) 3D to 2D camera projection Given a set of (X i,x i ), compute.
Medical Image Analysis Dr. Mohammad Dawood Department of Computer Science University of Münster Germany.
EFFICIENT VARIANTS OF THE ICP ALGORITHM
Geometric Transformations
Using simplified meshes for crude registration of two partially overlapping range images Mercedes R.G.Márquez Wu Shin-Ting State University of Matogrosso.
Medical Image Analysis Dr. Mohammad Dawood Department of Computer Science University of Münster Germany.
Improvements on Automated Registration CSc83020 Project Presentation Cecilia Chao Chen.
CENG 789 – Digital Geometry Processing 07- Rigid-Body Alignment Asst. Prof. Yusuf Sahillioğlu Computer Eng. Dept,, Turkey.
Efficient 2D-to-3D Correspondence Filtering for Scalable 3D Object Recognition Yeong-Jun Cho Computer Vision and Pattern Recognition,2013.
Mobile Robot Localization and Mapping Using Range Sensor Data Dr. Joel Burdick, Dr. Stergios Roumeliotis, Samuel Pfister, Kristo Kriechbaum.
CSCI 631 – Foundations of Computer Vision March 15, 2016 Ashwini Imran Image Stitching.
Copyright  2008 by CEBT Survey Schedule  Survey – Context-aware computing 팀원 간 자료 공유 – IDS Wiki Updated (
Panorama Tools H2T2 Group Capstone Project. H2T2 Group TungNS Project Manager HoaHM DesignerHuongP DeveloperThoND Tester.
Computer Photography -Scene Fixed 陳立奇.
Deformation Modeling for Robust 3D Face Matching Xioguang Lu and Anil K. Jain Dept. of Computer Science & Engineering Michigan State University.
CSCI 631 – Foundations of Computer Vision March 15, 2016 Ashwini Imran Image Stitching Link: singhashwini.mesinghashwini.me.
776 Computer Vision Jan-Michael Frahm Spring 2012.
CSE 554 Lecture 8: Alignment
Danfoss Visual Inspection System
CENG 789 – Digital Geometry Processing 08- Rigid-Body Alignment
University of Ioannina
Real Time Dense 3D Reconstructions: KinectFusion (2011) and Fusion4D (2016) Eleanor Tursman.
Geometrically Stable Sampling for the ICP Algorithm
Homography From Wikipedia In the field of computer vision, any
3D Scan Alignment Using ICP
Image Registration 박성진.
MultiModality Registration using Hilbert-Schmidt Estimators
Presentation transcript:

Intraoral Laser Scan Data와 Dental CT 영상 접합 및 개별 치아 분할 2015년 10월 2일 Team c 강명구, 박유군, 이상호 Creative integrated design Fall, 2015 TEAM C

Creative integrated design Fall, 2015 Contents Overview Goal/Problem & Requirement Approach Architecture Further Plan Basic Spec Development Environment Current Status Division and Assignment of Work Schedule Creative integrated design Fall, 2015 TEAM C

Creative integrated design Fall, 2015 Overview 연구 주제: Dental Image Registration & Segmentation Dental CT 영상 데이터와 Intraoral laser scan data의 정 합 정합된 Volume data로부터 개별 치아 분할 및 numbering 연구 단계 Image processing에 관한 사전 지식 이해 세 점 입력 UI 및 CT-Mesh registration 모듈 구현 개별 치아 분할 인터페이스 및 알고리즘 구현 Creative integrated design Fall, 2015 TEAM C

Goal / Problem & Requirement CT volume data 2D image를 쌓은 3D volume data 내부 구조의 정보 획득 가능 해상도가 높지 않고 noise 발생 및 artifact 영향 고려 필요 Creative integrated design Fall, 2015 TEAM C

Goal / Problem & Requirement Laser scan mesh data 레이저 기반 거리 측정 기술로 구강 내 위치 정보 획득 높은 정밀도의 치아 및 잇몸의 표면 정보 내부 구조의 정보 획득 불가 Creative integrated design Fall, 2015 TEAM C

Goal / Problem & Requirement CT-Mesh Registration Tooth Segmentation 치과 수술 정밀도 향상 및 가이드 설계 간편화 Creative integrated design Fall, 2015 TEAM C

Approach : CT-Mesh Registration Transformation Model : Rigid Transformation Rotation + Translation 6 DoF(Degree of Freedom) Creative integrated design Fall, 2015 TEAM C

Approach : CT-Mesh Registration Similarity Metric Feature-based vs. Voxel-based Multi-Modality Registration Similarity metric based on point to point correspondence Finding correspondences also needed. Creative integrated design Fall, 2015 TEAM C

Approach : CT-Mesh Registration Optimization Method Initialization + Local Optimizer Initial Transformation Calculation User input : 3 pairs of points Noise / Outliers : particle filtering / RANSAC* RANSAC* : Random sample consensus https://www.youtube.com/watch?v=Wi1WJlAIwBs Creative integrated design Fall, 2015 TEAM C

Approach : CT-Mesh Registration Optimization Method (Contd.) Local Optimizer : Iterative Closest Point(ICP) Closed-Form Solution Using SVD Creative integrated design Fall, 2015 TEAM C

Approach : Tooth Segmentation CT Data vs. Scanned Mesh Using full 3D data Without loss of information The segmentation plane can be used for both 2D and 3D. Based on Hui Gao and Oksam Chae, “Automatic Tooth Region Separation for Dental CT Images,” ICCIT Third International Conference on Convergence and Hybrid Information Technology, vol. 1, 2008. Creative integrated design Fall, 2015 TEAM C

Approach : Tooth Segmentation Automatic Dental Arch Fitting Segmentation of Teeth Region Using oral cavity structure Segmentation of Maxilla and Mandible Assumption : open bite constraint Arch Curve Fitting Assumed to be fourth order poly. Based on Hui Gao and Oksam Chae, “Automatic Tooth Region Separation for Dental CT Images,” ICCIT Third International Conference on Convergence and Hybrid Information Technology, vol. 1, 2008. Creative integrated design Fall, 2015 TEAM C

Approach : Tooth Segmentation Automatic Dental Arch Fitting (Contd.) Based on Hui Gao and Oksam Chae, “Automatic Tooth Region Separation for Dental CT Images,” ICCIT Third International Conference on Convergence and Hybrid Information Technology, vol. 1, 2008. Creative integrated design Fall, 2015 TEAM C

Approach : Tooth Segmentation Individual Tooth Segmentation Profiling Integral Intensity on Arch Curve local minimum at separating point Based on Hui Gao and Oksam Chae, “Automatic Tooth Region Separation for Dental CT Images,” ICCIT Third International Conference on Convergence and Hybrid Information Technology, vol. 1, 2008. Creative integrated design Fall, 2015 TEAM C

Architecture Acquired data Scanned mesh CT volume data Compute an initial transformation matrix based on user input Refine the matrix to minimize errors User input 3 pairs of points Aligned data Aligned mesh Aligned CT data Segmentation of Maxilla and Mandible Segmentation and numbering of individual tooth Creative integrated design Fall, 2015 TEAM C

Creative integrated design Fall, 2015 Further Plan CT-Mesh 정합 과정의 자동화 사용자 입력을 받지 않고 자동으로 대응점을 검출하여 정합 수행 feature detection의 활용 치아 결손, 부정교합 데이터에 대한 치아 분할 자동적인 치아 분할이 불가능한 경우 사용자로부터 가이드라인을 입력 받을 필요가 있음 입력을 받기 위한 사용자 친화적인 UI 설계 Creative integrated design Fall, 2015 TEAM C

Creative integrated design Fall, 2015 Basic Spec CT-Mesh Registration 정합된 mesh와 ground truth의 각 대응점간 평균 제곱근 오차(RMSE) 1mm 이하 512x512x200, 12bit CT 영상과 최대 100만개의 vertex 를 가지는 mesh의 registration 수렴 시간 30초 이하 Tooth Segmentation 치열 이상이나 금속이 없는 정상적 데이터에 대해 수동으로 분할한 결과와 비교하여 분할이 옳지 않은 치아의 수 2개 이 하 512x512x200, 12bit CT 영상의 전체 치아 분할 시간 1분 이하 Creative integrated design Fall, 2015 TEAM C

Development Environment Windows 7/8.1 C++ / C# 현재 회사 제품에 사용 중인 framework 제공 Microsoft Visual Studio Open source libraries Creative integrated design Fall, 2015 TEAM C

Creative integrated design Fall, 2015 Current Status 담당자와 프로젝트 방향 설정 프로젝트 개괄 이해, 평가 기준표 작성 구현 범위 구체화 및 구현 방안 선택 향후 일정 협의, 구현을 위한 framework 요청 사전지식 이해 3D 좌표계와 image processing 기본 이론 이해 Image registration과 segmentation을 위한 이론과 실제 구현 알고리즘 이해 Creative integrated design Fall, 2015 TEAM C

Division and Assignment of Work 항목 담당자 세 점 입력 UI 구현 팀 전원 initial transformation 행렬 계산 모듈 구현 fine tuning을 통한 optimization 개별 치아 분할 interface 구현 개별 치아 분할 알고리즘 구현 문제점 개선방안 및 추가 고려 사항 연구 필요할 시 분담 연구주제 특성상 팀원 모두 구현 방법과 과정의 이해 필요 문제점 개선방안 연구는 구성 단위 별로 분담할 수 있음 Creative integrated design Fall, 2015 TEAM C

Creative integrated design Fall, 2015 Schedule 2015 09 10 11 12 프로젝트 주제 선정 9/4 최초 담당자 미팅 및 회사평가표 작성, 제출 9/11 프로젝트 스펙 발표 10/2 프로젝트 중간 발표 10/23 최종 발표 전 시연 12/11 최종 발표 12/18 9/4 - 10/6 10/7 - 10/13 10/14 - 10/20 10/21 - 10/27 10/28 - 11/10 11/11 - 12/1 12/2 - 12/11 12/12 - 12/18 사전 지식 이해 및 개발 환경 구축 세 점 입력 UI 구현 initial transformation 변환행렬 계산 모듈 구현 fine tuning을 통한 optimization 개별 치아 분할 interface 구현 개별 치아 분할 알고리즘 구현 문제점 개선방안 및 추가 고려 사항 연구 최종 보고서 및 발표 준비 Creative integrated design Fall, 2015 TEAM C

Creative integrated design Fall, 2015 감사합니다. Creative integrated design Fall, 2015 TEAM C