Spot quantification apparatus, spot quantification method, and recording medium

A spot quantification apparatus comprising:

• image input means for inputting an image including a spot;
• filtering means for filtering the image input through said image input means a plurality of times while changing a parameter of a kernel within a predetermined range;
• and
• spot quantifying means for quantifying a feature of the spot based on a luminance distribution in each image which has been filtered a plurality of times by said filtering means,
• wherein:
• said filtering means changes a bandwidth of the kernel within a predetermined range to perform the filtering a plurality of times;
• said spot quantification means finds peak values of a normalized luminance distribution in each image which has been filtered a plurality of times by said filtering means, finds a maximum value in the peak values, and quantifies the bandwidth of the kernel used for an image in which the maximum value has been found;
• at least one of the image input means, the filtering means and the spot quantification means is implemented by a processor;
• said kernel is a concentric kernel defined by:
• K
• ⁡
• (
• x
• )
• =
• 1
• π
• ⁢
• ⁢
• r
• 1
• 2
• ⁢
• I
• {
• 
• x
• 
• ≤
• r
• 1
• }
• -
• 1
• π
• ⁡
• (
• r
• 2
• 2
• -
• r
• 1
• 2
• )
• ⁢
• I
• {
• r
• 1
• <
• 
• x
• 
• ≤
• r
• 2
• }
• ,
• where K(x) is a filter function, x is an input value for luminance, r1 is a radius of an internally concentric circle of the kernel, and r2 is a radius of an external concentric circle of the kernel;
• and
• said filtering means changes a radius of an internal circle of the kernel within a predetermined range to perform the filtering a plurality of times.

A spot quantification method comprising:

• inputting an image including a spot;
• filtering the input image a plurality of times while changing a parameter of a kernel within a predetermined range;
• and
• quantifying a feature of the spot based on a luminance distribution in each image which has been filtered a plurality of times,
• wherein:
• filtering further comprises changing a bandwidth of the kernel within a predetermined range to perform the filtering a plurality of times;
• quantifying further comprises finding peak values of a normalized luminance distribution in each image which has been filtered a plurality of times, and quantifying the bandwidth of the kernel used for an image in which the maximum value has been found;
• said kernel is a concentric kernel defined by:
• K
• ⁡
• (
• x
• )
• =
• 1
• π
• ⁢
• ⁢
• r
• 1
• 2
• ⁢
• I
• {
• 
• x
• 
• ≤
• r
• 1
• }
• -
• 1
• π
• ⁡
• (
• r
• 2
• 2
• -
• r
• 1
• 2
• )
• ⁢
• I
• {
• r
• 1
• <
• 
• x
• 
• ≤
• r
• 2
• }
• ,
• where K(x) is a filter function, x is an input value for luminance, r1 is a radius of an internally concentric circle of the kernel, and r2
• a radius of an external concentric circle of the kernel;
• and
• filtering further comprises changing a radius of an internal circle of the kernel within a predetermined range to perform the filtering a plurality of times.

A non-transitory computer readable recording medium which has recorded thereon a program for causing a computer to perform these functions:

• perform a filtering procedure configured to filter an image including a spot a plurality of times while changing a parameter of a kernel within a predetermined range;
• and
• perform a spot quantification procedure configured to quantify a feature of the spot based on a luminance distribution in each image region which has been filtered a plurality of times by said filtering procedure,
• wherein:
• the filter procedure is further configured to change a bandwidth of the kernel within a predetermined range to perform the filtering a plurality of times;
• the spot quantification procedure is further configured to find peak values of a normalized luminance distribution in each image which has been filtered a plurality of times, and quantify the bandwidth of the kernel used for an image in which the maximum value has been found;
• said kernel is a concentric kernel defined by:
• K
• ⁡
• (
• x
• )
• =
• 1
• π
• ⁢
• ⁢
• r
• 1
• 2
• ⁢
• I
• {
• 
• x
• 
• ≤
• r
• 1
• }
• -
• 1
• π
• ⁡
• (
• r
• 2
• 2
• -
• r
• 1
• 2
• )
• ⁢
• I
• {
• r
• 1
• <
• 
• x
• 
• ≤
• r
• 2
• }
• ,
• where K(x) is a filter function, x is an input value for luminance, r1 is a radius of an internally concentric circle of the kernel, and r2 is a radius of an external concentric circle of the kernel;
• and
• the filter procedure is further configured to change the radius of an internal circle of the kernel within a predetermined range to perform the filtering a plurality of times.