Classification and numbering of teeth in dental bitewing images

Slides:



Advertisements
Similar presentations
Applications of one-class classification
Advertisements

Complex Networks for Representation and Characterization of Images For CS790g Project Bingdong Li 9/23/2009.
RGB-D object recognition and localization with clutter and occlusions Federico Tombari, Samuele Salti, Luigi Di Stefano Computer Vision Lab – University.
Robust statistical method for background extraction in image segmentation Doug Keen March 29, 2001.
Sreya Chakraborty Under the guidance of Dr. K. R. Rao Multimedia Processing Lab (MPL) University of Texas at Arlington.
Beyond bags of features: Part-based models Many slides adapted from Fei-Fei Li, Rob Fergus, and Antonio Torralba.
Lecture 20 Object recognition I
A new face detection method based on shape information Pattern Recognition Letters, 21 (2000) Speaker: M.Q. Jing.
Rodent Behavior Analysis Tom Henderson Vision Based Behavior Analysis Universitaet Karlsruhe (TH) 12 November /9.
Prénom Nom Document Analysis: Data Analysis and Clustering Prof. Rolf Ingold, University of Fribourg Master course, spring semester 2008.
Multiple Human Objects Tracking in Crowded Scenes Yao-Te Tsai, Huang-Chia Shih, and Chung-Lin Huang Dept. of EE, NTHU International Conference on Pattern.
Pattern Classification All materials in these slides were taken from Pattern Classification (2nd ed) by R. O. Duda, P. E. Hart and D. G. Stork, John Wiley.
Handwritten Thai Character Recognition Using Fourier Descriptors and Robust C-Prototype Olarik Surinta Supot Nitsuwat.
Face Processing System Presented by: Harvest Jang Group meeting Fall 2002.
Maximum-Likelihood Image Matching Zheng Lu. Introduction SSD(sum of squared difference) –Is not so robust A new image matching measure –Based on maximum-likelihood.
DIGITAL SIGNAL PROCESSING IN ANALYSIS OF BIOMEDICAL IMAGES Prof. Aleš Procházka Institute of Chemical Technology in Prague Department of Computing and.
Jason Li Jeremy Fowers Ground Target Following for Unmanned Aerial Vehicles.
July 11, 2001Daniel Whiteson Support Vector Machines: Get more Higgs out of your data Daniel Whiteson UC Berkeley.
Shape Recognition and Pose Estimation for Mobile Augmented Reality Author : N. Hagbi, J. El-Sana, O. Bergig, and M. Billinghurst Date : Speaker.
New Segmentation Methods Advisor : 丁建均 Jian-Jiun Ding Presenter : 蔡佳豪 Chia-Hao Tsai Date: Digital Image and Signal Processing Lab Graduate Institute.
Presented by Tienwei Tsai July, 2005
LEAF BOUNDARY EXTRACTION AND GEOMETRIC MODELING OF VEGETABLE SEEDLINGS
Perception Introduction Pattern Recognition Image Formation
Ajay Kumar, Member, IEEE, and David Zhang, Senior Member, IEEE.
COMMON EVALUATION FINAL PROJECT Vira Oleksyuk ECE 8110: Introduction to machine Learning and Pattern Recognition.
Topic 10 - Image Analysis DIGITAL IMAGE PROCESSING Course 3624 Department of Physics and Astronomy Professor Bob Warwick.
Shape Based Image Retrieval Using Fourier Descriptors Dengsheng Zhang and Guojun Lu Gippsland School of Computing and Information Technology Monash University.
Intelligent Vision Systems ENT 496 Object Shape Identification and Representation Hema C.R. Lecture 7.
X-ray Image Segmentation using Active Shape Models
COMPARISON OF IMAGE ANALYSIS FOR THAI HANDWRITTEN CHARACTER RECOGNITION Olarik Surinta, chatklaw Jareanpon Department of Management Information System.
Texture scale and image segmentation using wavelet filters Stability of the features Through the study of stability of the eigenvectors and the eigenvalues.
DIGITAL IMAGE PROCESSING
Automated Target Recognition Using Mathematical Morphology Prof. Robert Haralick Ilknur Icke José Hanchi Computer Science Dept. The Graduate Center of.
Handwritten Recognition with Neural Network Chatklaw Jareanpon, Olarik Surinta Mahasarakham University.
Feature Vector Selection and Use With Hidden Markov Models to Identify Frequency-Modulated Bioacoustic Signals Amidst Noise T. Scott Brandes IEEE Transactions.
Fourier Descriptors For Shape Recognition Applied to Tree Leaf Identification By Tyler Karrels.
Visual Categorization With Bags of Keypoints Original Authors: G. Csurka, C.R. Dance, L. Fan, J. Willamowski, C. Bray ECCV Workshop on Statistical Learning.
1 Iris Recognition Ying Sun AICIP Group Meeting November 3, 2006.
Chapter 12 Object Recognition Chapter 12 Object Recognition 12.1 Patterns and pattern classes Definition of a pattern class:a family of patterns that share.
Ivica Dimitrovski 1, Dragi Kocev 2, Suzana Loskovska 1, Sašo Džeroski 2 1 Faculty of Electrical Engineering and Information Technologies, Department of.
Biologically Inspired Approaches to Automated Feature Extraction and Target Recognition Speaker: Yi-Chun Ke Adviser: Bo-Chi Lai.
MSc Project Musical Instrument Identification System MIIS Xiang LI ee05m216 Supervisor: Mark Plumbley.
Nottingham Image Analysis School, 23 – 25 June NITS Image Segmentation Guoping Qiu School of Computer Science, University of Nottingham
Shadow Detection in Remotely Sensed Images Based on Self-Adaptive Feature Selection Jiahang Liu, Tao Fang, and Deren Li IEEE TRANSACTIONS ON GEOSCIENCE.
Gaussian Mixture Model classification of Multi-Color Fluorescence In Situ Hybridization (M-FISH) Images Amin Fazel 2006 Department of Computer Science.
EE368 Final Project Spring 2003
Another Example: Circle Detection
Watermarking Scheme Capable of Resisting Sensitivity Attack
Image Representation and Description – Representation Schemes
Medical Image Analysis
IT472: Digital Image Processing
Digital Image Processing Lecture 20: Representation & Description
Speech Enhancement with Binaural Cues Derived from a Priori Codebook
PRESENTED BY Yang Jiao Timo Ahonen, Matti Pietikainen
Classification and numbering of teeth in dental bitewing images
A new data transfer method via signal-rich-art code images captured by mobile devices Source: IEEE Transactions on Circuits and Systems for Video Technology,
REMOTE SENSING Multispectral Image Classification
REMOTE SENSING Multispectral Image Classification
Visual Tracking of Cell Boundaries and Geometries
Centrality Bias Measure for High Density QR Code Module Recognition
An Introduction to Supervised Learning
Where did we stop? The Bayes decision rule guarantees an optimal classification… … But it requires the knowledge of P(ci|x) (or p(x|ci) and P(ci)) We.
Generally Discriminant Analysis
Mathematical Foundations of BME
Outline A. M. Martinez and A. C. Kak, “PCA versus LDA,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 23, no. 2, pp , 2001.
Translations.
Fourier Transform of Boundaries
A Block Based MAP Segmentation for Image Compression
Using Association Rules as Texture features
Source: Pattern Recognition Letters, VOL. 27, Issue 13, October 2006
Presentation transcript:

Classification and numbering of teeth in dental bitewing images M. H. Mahoor and M. Abdel-Mottaleb Pattern Recognition, Vol. 38, No. 4, pp. 577-586, April 2005. Speaker: Cheng-Hsiung Li Date: 2005-06-02

Outline Introduction Method Feature extraction and pre-classification Final classification and numbering Experiments and results Conclusion

Introduction - ADIS An automated dental identification system Bitewing Segmentation Feature extraction and search Bitewing * Bitewing of X-ray image (X光的照片) DB Identification Somebody of death Missing people

Introduction - Motivation The authors limit the comparison of the teeth to the ones that have the same number. Decrease the search space Increase the robustness of the system Segmentation Feature extraction (FDs) and Bayesian classification of molars and premolars Final classification and numbering

Method – Adult dentition system The adult dentition contains 32 teeth, 16 teeth in each jaw. molars premolars

Method – teeth segmentation First method -Segmentation Second method -Segmentation Segmentation Feature extraction Classification

Feature extraction and pre-classification(1) Complex coordinates signature Fourier descriptors (FDs) are one of the most popular techniques for shape analysis and description. The contour of the teeth as a complex signal u(n) defined based on the coordinates, x(n) and y(n). X jy(n) u(n) = x(n) + jy(n), n = 0,1,…,N-1 *DFT(傅立葉轉換)->高頻對應細節,低頻則是決定整體的形狀 Fourier transform to above complex signal Fourier coefficients: Segmentation Feature extraction Classification

Feature extraction and pre-classification(2) Centroid distance The centroid distance function is expressed by the distance of the boundary points from the centroid (xc, yc) of the shape. (xc, yc) Fourier coefficients: Segmentation Feature extraction Classification

Bayesian classification of teeth ci denote tooth class i, i.e., molar(c1) or premolar(c2) x denote the feature vector complex coordinates signature or centroid distance Suppose we know the prior probability p(ci) and the conditional densities p(x|ci). Posteriori probability Say c2 Say c1 P(x|c1) P(x|c2) P(x|ci) Segmentation Feature extraction Classification

Final classification and numbering Arrangement of teeth in dental bitewing images. (a) left quadrant (b) right quadrant. (a) (b) (c) (d) * When the confidence measure of the tooth is greater than threshold and their class membership are different, then we consider the tooth with low confidence measure as a misclassified tooth and assign it to the same class. Classification and numbering of the teeth in dental bitewing images. (c) left quadrant (d) right quadrant

Experiments and results(1) Training set The authors used 25 bitewing images as a training set to estimate the prior distribution p(ci) and the conditional distribution p(x|ci). Testing set For classification, 50 images, containing 220 molar and 180 premolar.

Experiments and results-(2) Pre-classification of teeth using first method of segmentation Pre-classification of teeth using second method of segmentation

Experiments and results-(3) Final classification of teeth using first method of segmentation Final classification of teeth using second method of segmentation

Experiments and results-(4) Missing teeth Missclassification teeth

Conclusion The authors introduced a method for robust classification and numbering of molar and premolar teeth in bitewing images using Bayesian classification.

Distinguish between method 1 and method 2 (c) (d) (e) (f) (a) Original image; (b) Result of enhancement; (c) Result of adaptive threshold; (d) Result of segmented teeth using morphological operation; (e) Bones image; (f) Final result of separated roots and crowns. Source: Automatic Human Identification based on Dental X-Ray Images

Fourier coefficients Fourier transform (DFT) Fourier transform (DFT) … Original image (S = 64) P = 2 P = 62 P = 64

Morphological image processing Dilation d (c) d/8 d (a) . d/4 (b) (a) Set A. (b) Square structuring element (dot is the center). (c) Dilation of A by B.