🚀 go-pugleaf

m_yearwo@csd.brispoly.ac.uk (Mike Yearworth) writes:

>Hi,
>	I'm looking for some references on extracting objective
>measurements of image quality, for example the signal to noise ratio
>of an image or its entropy, with a view to judging how good a particular
>image restoration technique is. I'd be interested in obtaining bibliographies
>and/or references to good textbooks. Thankyou.
>|================================================================|
>|Dr Michael Yearworth       | eMAIL:  m_yearwo@uk.ac.brispoly.csd|
>|Bristol Polytechnic        | Tel:    +44 272 656261 ext. 3159   | 
>|Frenchay, Bristol BS16 1QY | FAX:    +44 272 763860             |
>|U.K.                       |                                    |
>|================================================================|
> 
We use the volume between two surfaces(original image and restored
one)normalized by the surface area
as an invariant(w.r.t. arbitrary coordinate) measure to compare
results from different reconstruction algorithms.
We came up with a very efficient way of computing
the volume between
two surfaces and the surface area. Time
complexity is theta(n)
where n is the number of pixels. 
			 
This computing method is used in our paper entitled
"Discontinuity-preserving and Viewpoint Invariant Reconstruction
 of Visible Surfaces using a First Order Regularization"
which is to appear in the IEEE Int'l Conf. on Automation, robotics, and
Computer Vision, Sigapore, September 1992.

-- yi@ecn.purdue.edu

In article <yi.714841254@fife.ecn.purdue.edu> yi@fife.ecn.purdue.edu (Juneho Yi) writes:
>m_yearwo@csd.brispoly.ac.uk (Mike Yearworth) writes:
>>	I'm looking for some references on extracting objective
>>measurements of image quality, for example the signal to noise ratio
>>of an image or its entropy, with a view to judging how good a particular
>>image restoration technique is. I'd be interested in obtaining bibliographies
>>and/or references to good textbooks. Thankyou.
>
>We use the volume between two surfaces(original image and restored
>one)normalized by the surface area
>as an invariant(w.r.t. arbitrary coordinate) measure to compare
>results from different reconstruction algorithms.
>We came up with a very efficient way of computing
>the volume between
>two surfaces and the surface area. Time
>complexity is theta(n)
>where n is the number of pixels. 

I'm assuming that the original poster is interested in objective measures
that have some correlation to the assessment a human observer would make.
Does this measure have any relationship to subjective evaluation?  What is
the point of normalizing with respect to surface area?

The simplest objective measure is mean-square error, but it doesn't always
correlate closely with a subjective evaluation of image quality.  Granted, 
the above technique is *an* objective measure, but the question is whether 
it is a good one for the purposes of the original poster.


--
Stan Reeves
Auburn University, Department of Electrical Engineering, Auburn, AL  36849
INTERNET: sjreeves@eng.auburn.edu

sjreeves@eng.auburn.edu (Stan Reeves) writes:

>In article <yi.714841254@fife.ecn.purdue.edu> yi@fife.ecn.purdue.edu (Juneho Yi) writes:
>>m_yearwo@csd.brispoly.ac.uk (Mike Yearworth) writes:
>>>	I'm looking for some references on extracting objective
>>>measurements of image quality, for example the signal to noise ratio
>>>of an image or its entropy, with a view to judging how good a particular
>>>image restoration technique is. I'd be interested in obtaining bibliographies
>>>and/or references to good textbooks. Thankyou.
>>
>>We use the volume between two surfaces(original image and restored
>>one)normalized by the surface area
>>as an invariant(w.r.t. arbitrary coordinate) measure to compare
>>results from different reconstruction algorithms.
>>We came up with a very efficient way of computing
>>the volume between
>>two surfaces and the surface area. Time
>>complexity is theta(n)
>>where n is the number of pixels. 

>I'm assuming that the original poster is interested in objective measures
>that have some correlation to the assessment a human observer would make.
>Does this measure have any relationship to subjective evaluation?  What is
>the point of normalizing with respect to surface area?

By normalizing the volume by the surface area, the value of the measure 
can be compared for different size of images.

>The simplest objective measure is mean-square error, but it doesn't always
>correlate closely with a subjective evaluation of image quality.  Granted, 
>the above technique is *an* objective measure, but the question is whether 
>it is a good one for the purposes of the original poster.

The Mean-Square error is kinda L2 metric.
Apparently, if MSE is used as a measure, where the slope is high,
the difference value is emphasized more than in flat regions.
I am saying that MSE measure is ** not invariant ** w.r.t. a coordinate system.
The original poster was looking for methods to compare various algorithms on
image restoration, wasn`t it ?

>--
>Stan Reeves
>Auburn University, Department of Electrical Engineering, Auburn, AL  36849
>INTERNET: sjreeves@eng.auburn.edu


-- yi@ecn.purdue.edu