JPEG Image Compression All of the Gory Details

JPEG Image Compression

All of the Gory Details

Biological ImagingRoss Whitaker

University of Utah

Clinical Imaging• Paradigm

– Diagnosis/prognosis (individuals)

– Radiologists/experts• Trends

– More/larger datasets (+3D)

– Surgery/planning– Quantification

QuickTime™ and aTIFF (Uncompressed) decompressor

are needed to see this picture.

QuickTime™ and aTIFF (Uncompressed) decompressor

are needed to see this picture.

Biological Imaging• Understanding organisms/populations• Growth–explosive

– Trends: instrumentation + science• Very large datasets

– E.g. terabytes – Acquisition times

• Paradigm– Quantification/exploration– Desperate need of tools

Biological Imaging Examples

• Mouse phenotyping – M. Capecchi

• Retinal mapping – R. Marc

The Capecchi Laboratory

• Knock-out mouse– Mouse as test tube

for genetics• Understanding the

effects of particular genes

• Mutations to study mouse-bat relationships– Skeleton

Mouse Phenotyping• State of the art

– Dozens of animals– Dissection, etc.

• Future– Hundreds/thousands

of animals– 3D imaging (MRI,CT)

Image-Based PhenotypingA Significant Engineering Challenge

ImagesAtlas/protoype

MutantsImages

Geometric/StatisticalAnalysis

Mutants

Imaging•Reconstruction•Image processing (denoising)

Shape & Deformation•Atlas/image registration•Statistics of shape/deformation

•Fundamental questions•Visualization

•Algorithm development and tuning•End user

Atlases/Prototypes•Image Registration

•Scale of problem•Deformation vs articulation

•Modeling•Representing variability •Structures or pixels

•Segmentation•User interaction/visualization•Automation–Computer Vision

Marc Laboratory

Retinal Mapping

•Mosaicing•Distortion and position•Very large images (100mps)

•Filtering and segmentation•Noise, texture

•3D tracking

Visualization in Biological Imaging

• As a visual front end for other tools– User assisted analysis

• Exploration– Heterogeneous data– Large databases

• Interpretation– Statistics (high dimensions)– Anatomical variability

Done