tracking turbulent 3d features lu zhang nov. 10, 2005
TRANSCRIPT
Tracking Turbulent 3D Features
Lu Zhang
Nov. 10, 2005
Motivations
Introduction Visualization techniques can help scientists to identify observed
phenomena both in scientific simulation or practical circumstance.
Application Storm, Hurricane, Ocean wave, Cloud….
Common features:• multiple evolution • time-varying• huge dataset• non-rigid
Outline
Segmentations and Region growing Thresholding Region growing
Features extraction Different features
Classification and Feature tracking Tracking methods Classes and structures
Overview
The original dataset
Flowchart and Modulus
Input imagesSegmentation
Feature extraction
Classification
Graph building
Basic features
classes
Directed acyclic graph
Segmentations and Region growing Thresholding Global thresholding vs optimal thresholding
Region Growing method Iterative region growing method [1]
Segmentations and Region growing Region Growing Basic features
timeID viewID x y R G B
Features extraction
Feature structure
After gaining region information from segmentation stage, we can browse each region to find basic features Areas – The count of all pixels in the region. Center of Gravity –The center of all points in one region. Diameter - Diameter is the distance between two points on the
boundary of the region whose mutual distance is the maximum. Perimeter - The number of pixels under each edge label. Fourier descriptors – Fourier transform of boundary points.
Features extraction
Output from Feature extraction module viewID mx my areas labeling timeID …..
Classification /Feature tracking
ClassificationAfter feature extraction module, we can gain a list of feature information for each region in different views.
One AssumptionBecause all the views have strictly time order, we can assume the difference between a pair of views should not vary too much.
Classification /Feature tracking
Evolution in time-varying imagesThere are five different changes of regions between a pair of views.
Continuation: one feature continues from dataset at t1 to the next dataset at t2
Creation: new feature appear in t2 Dissipation: one feature weakens and becomes part of the background Bifurcation: one feature in t1 separates into two or more features in t2. Amalgamation: two or more features merge from one time step to the
next.
Classification /Feature tracking
Classification Several pattern recognition methods can be used here,
eg. Euclidean Distance classifier:
KNN classifier: Find the K-Nearest Neighbor feature clusters in datas
et t1 and dataset t2.
Aeuc xAxd ,
Classification /Feature tracking
Output from Classification module
I create a new class to preserve the output dataset from Classification module: class LabelTrack(). It preserve the information:
1. ViewID: camera positions, we will move camera around the object in order to restore 3D object.
2. timeID: time order, for each camera position , we will take several time- varying images
3. classID: class number after correspondence computation between a pair of images in time order
4. Label: the original region numbers before correspondence computaton
5. R, G, B: the color information for each pixel
6. Coordinate x, y: the 2D coordinate of the projection of 3D object.
7. Forward pointer: preserve the labeling information of the previous dataset
8. Backward pointer: preserve the labeling information of the next dataset
Computation Time
The importance of computation time Size of database:
512*512*24*40(time orders)*N(camera positions)
In [5], the resolution is 128^3 with the computation time: 40 minutes.
In my project, I use 3 minutes for 512*512*24*40. Because this is the framework of the whole project, there are a lot of I/O operations to see the temporary results. My expectation is 1 minutes for each camera position finally.
REFERENCES
[1] Snyder and Cowart, “An Iterative Approach to Region Growing”, IEEE transaction on PAMI, 1983
[2] Wesley E.Snyder and Hairong Qi, “Machine Vision”, Cambridge [3] Richard O.Duda, Peter Hart, David Stork, “Pattern Classification”, Prenti
ce Hall [4] Rafael Gonzalez, Richard Woods,”Digital Image Processing”, 2nd, Prenti
ce Hall [5] D.Silver, Xin Wang, ”volume tracking”, Visualization '96. Proceedings.27
Oct.-1 Nov. 1996
Thanks
Any questions?