
Image analysis technologies can automatically identify individual cells and measure them
Quantitative Imaging: tools for extracting high quality data from images
We develop computer vision tools to maximise the quantity and quality of information that can be extracted from images. Our research spans 3D imaging, dynamic imaging and visualisation.
-
27 July 2011 | Updated 12 April 2013
Overview
Automating image analysis
We develop computer vision tools that, like the human eye, can recognise and discriminate between objects but are also able to make detailed, accurate measurements. And do it much better and faster than people can.
Our image analysis research supports the automation of imaging applications, while boosting throughput, reliability and
reproducibility.
Applications of quantitative imaging
Our Quantitative Imaging team is applying a growing portfolio of image analysis technologies to problems like:
- Imaging virus entry into cells
- Profiling stem cells
- Tracing neurites in Alzheimer’s Diseased brain cells
- Detecting breast cancer
- Retinal imaging
- Bacterial dynamics
- Automated histology
- 3-D segmentation of rocks
We're excited that our neuron outgrowth model has been taken up by instrument manufacturers"
Our imaging capabilities
We use techniques including
- Image segmentation
- Unmixing of hyperspectral datasets
- Stereo/multi-view vision
- Dynamic imaging – motion analysis
- Software engineering and high performance computing
We can create stunning 3D models by fusing multiple views of objects and scenes. We apply this technique in plant phenotyping and medical imaging.
We analyse time-lapse imaging sequences to characterise and understand processes including cell migration and biofilms.
Our visualisation team is taking advantage of cloud computing by building a visualisation toolbox using the NECTAR [External Link] cloud infrastructure. Visualisation helps maximise human understanding of data. We use immersive environments and human computer interaction technologies for visualisation.
As part of CSIRO's Computational and simulation sciences initiative, we are also exploring applications of graphics processing units (GPUs) to speed up image analysis tasks.
Our commercial imaging tools
We invite contact with companies, research institutions and researchers interested in working with us on a range of imaging challenges. Please use the contact details in the right sidebar to get in touch.
Our image analysis software has already been built into commercially-available instruments such as:
HCA-Vision: software for cytomicsOur HCA-Vision tool helps researchers find more effective, safer pharmaceuticals and better understand the mechanisms of disease. HCA-Vision is an automated image analysis software package that rapidly and reproducibly measures functional features of neuronal cells in images.
Better melanoma diagnosis with SolarScan™Australian medical instrumentation company Polartechnics came to CSIRO to develop image analysis software for a device that supports diagnosis of melanoma, skin cancer’s deadliest form.
CSIRO image analysis technology is built into SolarScan®, a device that helps doctors detect melanoma earlier.
Spot: software for DNA microarray image analysisCSIRO has a track record in developing innovative solutions to practical problems. Our Spot software, a microarray spot detection and characterisation package, helps researchers get better gene expression information. Spot 3.0 is free.
Image analysis projects
We also have contributed to image analysis in projects concerning:
Automated detection of road cracksCSIRO’s automated road sensing system, RoadCrack, is able to find cracks in road paving down to a width of one millimetre at high speeds, reducing the cost and increasing the accuracy of road crack detection.
Hyperspectral imaging for analysing materialsChecking the contents of tablets, finding new ore bodies to mine, identifying tumours – these are just some of the applications of hyperspectral imaging. This technology has been the focus of CSIRO research which helps determine the composition of different materials and how composition varies throughout the material.
Faster drug discoveryCSIRO helped Axon instruments develop image analysis software for their ImageXpress™ cellular screening system. ImageXpress™ helps researchers discover safer drugs more quickly.
Image analysis for 3D phenotypingImaging techniques can provide faster and more accurate measures of organism characteristics than manual methods. Our 3D phenotyping group uses two dimensional images of living things to create three-dimensional models. We work most often with plants and insects.
Cloud-based image analysis and processing toolboxWe're using the National eResearch Collaboration Tools and Resources (NeCTAR) cloud infrastructure to provide improved access to biomedical image processing and analysis software packages to researchers.
Learn more about our work in Business and Services Analytics.
-
SolarScan™ is a trade mark of Polartechnics
-
ImageXpress™ is a trade mark of Axon Instruments
-
RoadCrack™ is a trade mark of CSIRO Australia.
Share this CSIRO content using: