Difference: GMT_4YP_17_3 (1 vs. 3)

Revision 3
27 Mar 2017 - gmt11
Line: 1 to 1
 
META TOPICPARENT name="GMTTeaching"
Dr Graham Treece, Department of Engineering
Line: 11 to 11
 

Digital image filters, which replace each pixel in an image with some combination of pixels from a local neighbourhood, are extremely useful in image processing. Of particular note are those which attempt to reduce noise levels in an image whilst preserving the location and extent of any feature gradients, or 'edges'. Multiple linear filters have been developed for this purpose, for instance anisotropic diffusion, non-local means and image-guided filtering. These are all able to automatically adjust whichpixels are used in the local neighbourhood around each pixel to give the best reduced-noise new estimate.

Changed:
<
<		 Morphological filters, which rely on data ranking rather than level, have received much less attention. This is despite their inherent insensitivity to the extent of an edge which should make them good candidates for edge-preserving filtering. It has recently been demonstrated that the bitonic filter, a morphological filter based on 'opening' and 'closing' operations, can perform nearly as well as anisotropic diffusion and non-local means for noisy natural images. This is despite the use of a local 'structuring element' which does not automatically adjust to the data around each pixel. Hence there is good reason to expect that local variation of the shape of this structuring element will significantly improve performance. However, there has been very little work on locally-varying morphological filters, partly perhaps due to difficulties in both implementation and analysis. The aim of this project is to investigate such filters, which could encompass implementation efficiency, what determines the local shape of the structuring element, or a number of other avenues.
>
>		 Morphological filters, which rely on data ranking rather than level, have received much less attention. This is despite their inherent insensitivity to the extent of an edge which should make them good candidates for edge-preserving filtering. It has recently been demonstrated that the bitonic filter, a morphological filter based on 'opening' and 'closing' operations, can perform nearly as well as anisotropic diffusion and non-local means for noisy natural images. This is despite the use of a local 'structuring element' which does not automatically adjust to the data around each pixel. Hence there is good reason to expect that local variation of the shape of this structuring element will significantly improve performance. However, there has been very little work on locally-varying morphological filters, partly perhaps due to difficulties in both implementation and analysis. The aim of this project is to investigate such filters, which could encompass implementation efficiency, what is the best shape of the structuring elements, or what determines the local orientation of the structuring elements.
 

This is an algorithmic development / software project, so experience of writing software is essential, though the development could also be done using Matlab.

Revision 2
24 Mar 2017 - gmt11
Line: 1 to 1
 
META TOPICPARENT name="GMTTeaching"
Dr Graham Treece, Department of Engineering

F-GMT11-3: Image processing with structurally-varying morphological operators

Changed:
<
<
grapes_noise.png grapes_gaussian.png grapes_bitonic_fixed.png grapes_bitonic_varying.png
This is a typical natural image of some grapes, with a fairly high level of additional noise. Removing such noise is a very common image processing operation. A typical Gaussian filter removes noise well but also blurs the edges a lot in the process There are countless more complex linear filters that improve on this result. However, it is also possible to use morphological operators to do so. These are non-linear and use data ranking (sorting). They hence naturally preserve edges and are insensitive to threshold levels. Morphological operators rely on a local structuring element. In contrast to linear filters, there has been very little work on trying to locally vary this structuring element - but as can be seen from the image above, such local variation has much potential.
>
>
grapes_noise.png grapes_gaussian.png grapes_bitonic_fixed.png grapes_bitonic_varying.png
This is a typical natural image of some grapes, with a fairly high level of additional noise. Removing such noise is a very common image processing operation. A typical Gaussian filter removes noise well but also blurs the edges a lot in the process There are countless more complex linear filters that improve on this result. Morphological operators (non-linear and involving sorting the data) are much less common, but they naturally preserve edges and are insensitive to threshold levels. Morphological operators rely on a local structuring element. In contrast to linear filters, there has been very little work on trying to locally vary this structuring element - but as can be seen from the image above, such local variation has much potential.
 

Changed:
<
<		 Bone fracture is a major issue affecting millions of people annually, and we have recently been involved in research which has the potential to contribute significantly to both our understanding of fracture and how various preventative measures affect bone. The advances have come from a much more precise (and hence much more sensitive) measurement of the bone cortex (the denser layer surrounding the less dense bone in the middle). As well as contributing to our understanding of fracture, this technique is also being used to underpin models of bone used in mechanical analysis, for instance to see what happens to the skull during a head injury. It is also increasingly being used by paleo-anthropologists who are interested in the properties of very old bones.
>
>		 Digital image filters, which replace each pixel in an image with some combination of pixels from a local neighbourhood, are extremely useful in image processing. Of particular note are those which attempt to reduce noise levels in an image whilst preserving the location and extent of any feature gradients, or 'edges'. Multiple linear filters have been developed for this purpose, for instance anisotropic diffusion, non-local means and image-guided filtering. These are all able to automatically adjust whichpixels are used in the local neighbourhood around each pixel to give the best reduced-noise new estimate.
 
Changed:
<
<		 The existing technique expects to measure a single 'layer' of bone, but in many places bone has more than one layer: for instance in the skull there are two layers: an inner and an outer table, which in some places join together to become just one. In these areas it is not clear whether our measurements relate to one or both tables. The aim of this project is to try to disentangle this information, by looking at measurements made from either side of the bone, to try to split them into clean sets of measurements of either the whole skull thickness, or just the inner and outer tables. This is an interesting problem in computational geometry, but also in surface-based registration and visualisation
>
>		 Morphological filters, which rely on data ranking rather than level, have received much less attention. This is despite their inherent insensitivity to the extent of an edge which should make them good candidates for edge-preserving filtering. It has recently been demonstrated that the bitonic filter, a morphological filter based on 'opening' and 'closing' operations, can perform nearly as well as anisotropic diffusion and non-local means for noisy natural images. This is despite the use of a local 'structuring element' which does not automatically adjust to the data around each pixel. Hence there is good reason to expect that local variation of the shape of this structuring element will significantly improve performance. However, there has been very little work on locally-varying morphological filters, partly perhaps due to difficulties in both implementation and analysis. The aim of this project is to investigate such filters, which could encompass implementation efficiency, what determines the local shape of the structuring element, or a number of other avenues.
 
Changed:
<
<		 This is an algorithmic development / computational geometry / software project, so experience of writing software is essential, though the development could also be done using Matlab.
>
>		 This is an algorithmic development / software project, so experience of writing software is essential, though the development could also be done using Matlab.
 

Click here for other imaging projects offered by Graham Treece.

View topic | History: r3 < r2 < r1 | More topic actions...
 
No permission to view System.WebBottomBarExample