NAME

IPA - Image Processing Algorithms

DESCRIPTION

IPA stands for Image Processing Algorithms and represents the library of image processing operators and functions. IPA is based on the Prima toolkit ( http://www.prima.eu.org ), which in turn is a perl-based graphic library. IPA is designed for solving image analysis and object recognition tasks in perl.

USAGE

IPA works mostly with grayscale images, which can be loaded or created by means of Prima toolkit. See Prima::Image for the information about Prima::Image class functionality. IPA methods are grouped in several modules, that contain the specific functions. The functions usually accept one or more images and optional parameter hash. Each function has its own set of parameters. If error occurs, the functions call die, so it is advisable to use eval blocks around the calls.

The modules namespaces can be used directly, e.g. use IPA::Local qw(/./), use IPA::Point qw(/./) etc, with each module defining its own set of exportable names. In case when all names are to be exported, it is possible to use IPA.pm exporter by using use IPA qw(Local Point) syntax, which is equivalent to two separate 'use' calls above. Moreover, if all modules are to be loaded and namespace exported, special syntax use IPA 'all' is available.

A code that produces a binary thresholded image out of a 8-bit grayscale image is exemplified:

use Prima;
use IPA qw(Point);
my $i = Prima::Image-> load('8-bit-grayscale.gif');
die "Cannot load:$@\n" if $@;
my $binary = threshold( $i, minvalue => 128);

The abbreviations for pixel types are used, derived from the im::XXX image type constants, as follows:

im::Byte     - 8-bit unsigned integer
im::Short    - 16-bit signed integer
im::Long     - 32-bit signed integer
im::Float    - float
im::Double   - double
im::Complex  - complex float
im::DComplex - complex double

Each function returns the newly created image object with the result of the operation, unless stated otherwise in API.

API

IPA::Point

IPA::Point module contains functions that perform single point transformations and image arithmetic.

Single-point processing is a simple method of image enhancement. This technique determines a pixel value in the enhanced image dependent only on the value of the corresponding pixel in the input image. The process can be described with the mapping function

s = M(r)

where r and s are the pixel values in the input and output images, respectively.

combine [ images, conversionType = conversionScale, combineType = combineSum, rawOutput = 0]

Combines set of images of same dimension and bit depth into one and returns the resulting image.

Supported types: Byte, Short, Long.

Parameters:

images ARRAY

Array of image objects.

conversionType INTEGER

An integer constant, one of the following, that indicates how the resulting image would be adjusted in accord to the minimal and maximal values of the result. Trunc constants cut off the output values to the bit maximum, for example, a result vector in 8-bit image [-5,0,100,300] would be transformed to [0,0,100,255]. Scale constants scale the whole image without the cutoff; the previous example vector would be transformed into [0,4,88,255]. The Abs suffix shows whether the range calculation would use the whole domain, including the negative values, or the absolute values only.

conversionTruncAbs
conversionTrunc
conversionScale
conversionScaleAbs

Default is conversionScale.

combineType INTEGER

An integer constant, indicates the type of action performed between pixels of same [x,y] coordinates.

combineMaxAbs          - store the maximal absolute pixel value
combineSignedMaxAbs    - compute the maximal absolute value, but store its original ( before abs()) value
combineSumAbs          - store the sum of absolute pixel values
combineSum             - store the sum of pixel values
combineSqrt            - store the square root of the sum of the squares of the pixel values

Default is combineSum.

rawOutput BOOLEAN

Discards conversionType parameter and performs no conversion. If set to true value, the conversion step is omitted.

Default is 0.

threshold IMAGE [ minvalue, maxvalue = 255]

Performs the binary thresholding, governed by minvalue and maxvalue. The pixels, that are below minvalue and above maxvalue, are mapped to value 0; the other values mapped to 255.

Supported types: Byte

gamma IMAGE [ origGamma = 1, destGamma = 1]

Performs gamma correction of IMAGE by a product of origGamma and destGamma.

Supported types: Byte

remap IMAGE [ lookup ]

Performs image mapping by a passed lookup array of 256 integer values. Example:

IPA::Point::remap( $i, lookup => [ (0) x 128, (255) x 127]);

is an equivalent of

IPA::Point::threshold( $i, minvalue => 128);

Supported types: 8-bit

subtract IMAGE1, IMAGE2, [ conversionType = conversionScale, rawOutput = 0]

Subtracts IMAGE2 from IMAGE1. The images must be of same dimension. For description of conversionType and rawOutput see combine.

Supported types: Byte

mask IMAGE [ test, match, mismatch ]

Test every pixel of IMAGE whether it equals to test, and assigns the resulting pixel with either match or mismatch value. All test, match, and mismatch scalars can be either integers ( in which case mask operator is similar to threshold ), or image objects. If the image objects passed, they must be of the same dimensions and bit depth as IMAGE.

Supported types: Byte, Short, Long.

average LIST

Combines images of same dimensions and bit depths, passed as an anonymous array in LIST and returns the average image.

Supported types: Byte, Short, Long, 64-bit integer.

equalize IMAGE

Returns a histogram-equalized image.

Supported types: Byte

IPA::Local

Contains functions that operate in the vicinity of a pixel, and produce image where every pixel is dependant on the values of the source pixel and the values of its neighbors. The process can be described with the mapping function

      |r(i,j),r(i+1,j)...|
s = M |...               |
      |r(j+1,i) ...      |

where r and s are the pixel values in the input and output images, respectively.

crispening IMAGE

Applies the crispening algorithm to IMAGE and returns the result.

Supported types: Byte

sobel IMAGE [ jobMask = sobelNWSE|sobelNESW, conversionType = conversionScaleAbs, combineType = combineMaxAbs, divisor = 1]

Applies Sobel edge detector to IMAGE.

Supported types: Byte

Parameters:

jobMask INTEGER

Combination of the integer constants, that mask the pixels in Sobel 3x3 kernel. If the kernel is to be drawn as

| (-1,1) (0,1) (1,1) |
| (-1,0) (0,0) (1,0) |
| (-1,-1)(0,-1)(1,-1)|

Then the constants mask the following points:

sobelRow      - (-1,0),(1,0)
sobelColumn   - (0,1),(0,-1)
sobelNESW     - (1,1),(-1,-1)
sobelNWSE     - (-1,1),(1,-1)

(0,0) point is always masked.

divisor INTEGER

The resulting pixel value is divided to divisor value after the kernel convolution is applied.

conversionType and <combineType> parameters described in combine.

GEF IMAGE [ a0 = 1.3, s = 0.7]

Applies GEF algorithm ( first derivative operator for symmetric exponential filter) to IMAGE.

Supported types: Byte

SDEF IMAGE [ a0 = 1.3, s = 0.7]

Applies SDEF algorithm ( second derivative operator for symmetric exponential filter) to IMAGE.

Supported types: Byte

deriche IMAGE [ alpha ]

Applies Deriche edge detector.

Supported types: Byte

filter3x3 IMAGE [ matrix, expandEdges = 0, edgecolor = 0, conversionType = conversionScaleAbs, rawOutput = 0, divisor = 1 ]

Applies convolution with a custom 3x3 kernel, passed in matrix.

Supported types: Byte

Parameters:

matrix ARRAY

Array of 9 integers, a 3x3 kernel, to be convoluted with IMAGE. Indexes are:

|0 1 2|
|3 4 5|
|6 7 8|

expandEdges BOOLEAN

If false, the edge pixels ( borders ) not used in the convolution as center pixels. If true, the edge pixels used, and in this case edgecolor value is used to substitute the pixels outside the image.

edgecolor INTEGER

Integer value, used for substitution of pixel values outside IMAGE, when expandEdges parameter is set to 1.

divisor INTEGER

The resulting pixel value is divided to divisor value after the kernel convolution is applied.

conversionType

See combine

rawOutput

See combine

median IMAGE [ w = 0, h = 0 ]

Performs adaptive thresholding with median filter with window dimensions w and h.

unionFind IMAGE [ method, threshold ]

Applies a union find algorithm selected by method. The only implemented method is average-based region grow ( 'ave' string constant ). Its only parameter is threshold, integer value of the balance merger function.

Supported types: Byte

hysteresis IMAGE, thresold => [ thr0, thr1], neighborhood => 4 or 8

Perform binary hysteresis thresholding of Byte image with two thresholds, thr0 and thr1. A pixel is set to 1, if its value is larger than thr1 or if it is larger than thr0 and the pixel is adjacent to already marked pixels.

Default value of neighborhood is 8.

Supported types: Byte

IPA::Global

Contains methods that produce images, where every pixel is a function of all pixels in the source image. The process can be described with the mapping function

s = M(R)

where s is the pixel value in the output images, and R is the source image.

close_edges IMAGE [ gradient, maxlen, minedgelen, mingradient ]

Closes edges of shapes on IMAGE, according to specified gradient image. The unclosed shapes converted to the closed if the gradient spot between the suspected dents falls under maxlen maximal length increment, mingradient the minimal gradient value and the edge is longer than minedgelen.

Supported types: Byte

Parameters:

gradient IMAGE: Specifies the gradient image
maxlen INTEGER: Maximal edge length
minedgelen INTEGER: Minimal edge length
mingradient INTEGER: Minimal gradient value

fill_holes IMAGE [ inPlace = 0, edgeSize = 1, backColor = 0, foreColor = 255, neighborhood = 4]

Fills closed shapes to eliminate the contours with holes in IMAGE.

Supported types: Byte

Parameters:

inPlace BOOLEAN: If true, the original image is changed
edgeSize INTEGER: The edge breadth that is not touched by the algorithm
backColor INTEGER: The pixel value used for determination whether a pixel belongs to the background.
foreColor INTEGER: The pixel value used for hole filling.
neighborhood INTEGER: Must be either 4 or 8. Selects whether the algorithm must assume 4- or 8- pixel connection.

area_filter IMAGE [ minArea = 0, maxArea = INT_MAX, inPlace = 0, edgeSize = 1, backColor = 0, foreColor = 255, neighborhood = 4]

Identifies the objects on IMAGE and filters out these that have their area less than minArea and more than maxArea. The other parameters are identical to those passed to fill_holes.

identify_contours IMAGE [ edgeSize = 1, backColor = 0, foreColor = 255, neighborhood = 4]

Identifies the objects on IMAGE and returns the contours as array of anonymous arrays of 4- or 8- connected pixel coordinates.

The parameters are identical to those passed to fill_holes.

Supported types: Byte

IPA::Morphology

Quote from http://www.dai.ed.ac.uk/HIPR2/morops.htm:

Morphological operators often take a binary image and a structuring element as input and combine them using a set operator (intersection, union, inclusion, complement). They process objects in the input image based on characteristics of its shape, which are encoded in the structuring element.

Usually, the structuring element is sized 3x3 and has its origin at the center pixel. It is shifted over the image and at each pixel of the image its elements are compared with the set of the underlying pixels. If the two sets of elements match the condition defined by the set operator (e.g. if the set of pixels in the structuring element is a subset of the underlying image pixels), the pixel underneath the origin of the structuring element is set to a pre-defined value (0 or 1 for binary images). A morphological operator is therefore defined by its structuring element and the applied set operator.

Morphological operators can also be applied to gray-level images, e.g. to reduce noise or to brighten the image.

BWTransform IMAGE [ lookup ]

Applies 512-byte lookup LUT string ( look-up table ) to image and returns the convolution result ( hit-and-miss transform). Each byte of lookup is a set of bits, each corresponding to the 3x3 kernel index:

|4 3 2|
|5 0 1|
|6 7 8|

Thus, for example, the X-shape would be represented by offset 2**0 + 2**2 + 2**4 + 2**6 + 2**8 = 341 . The byte value, corresponding to the offset in lookup string is stored in the output image.

IPA::Morphology defines several basic LUT transforms, which can be invoked by the following code:

IPA::Morphological::bw_METHOD( $image);

or its alternative

IPA::Morphology::BWTransform( $image, lookup => $IPA::Morphology::transform_luts{METHOD}->());

Where METHOD is one of the following string constants:

dilate: Morphological dilation
erode: Morphological erosion
isolatedremove: Remove isolated pixels
togray: Convert binary image to grayscale by applying the mean filter
invert: Inversion operator
prune: Removes 1-connected end points
break_node: Removes node points that connect 3 or more lines

Supported types: Byte

dilate IMAGE [ neighborhood = 8 ]

Performs morphological dilation operation on IMAGE and returns the result. neighborhood determines whether the algorithm assumes 4- or 8- pixel connectivity.

Supported types: Byte, Short, Long, Float, Double

erode IMAGE [ neighborhood = 8 ]

Performs morphological erosion operation on IMAGE and returns the result. neighborhood determines whether the algorithm assumes 4- or 8- pixel connectivity.

Supported types: Byte, Short, Long, Float, Double

opening IMAGE [ neighborhood = 8 ]

Performs morphological opening operation on IMAGE and returns the result. neighborhood determines whether the algorithm assumes 4- or 8- pixel connectivity.

Supported types: Byte, Short, Long, Float, Double

closing IMAGE [ neighborhood = 8 ]

Performs morphological closing operation on IMAGE and returns the result. neighborhood determines whether the algorithm assumes 4- or 8- pixel connectivity.

Supported types: Byte, Short, Long, Float, Double

gradient IMAGE [ neighborhood = 8 ]

Returns the result or the morphological gradient operator on IMAGE. neighborhood determines whether the algorithm assumes 4- or 8- pixel connectivity.

Supported types: Byte, Short, Long, Float, Double

algebraic_difference IMAGE1, IMAGE2 [ inPlace = 0 ]

Performs the algebraic difference between IMAGE1 and IMAGE2. Although this is not a morphological operator, it is often used is conjunction with ones. If the boolean flag inPlace is set, IMAGE1 contains the result.

Supported types: Byte, Short, Long, Float, Double

watershed IMAGE [ neighborhood = 4 ]

Applies the watershed segmentation to IMAGE with given neighborhood.

Supported types: Byte

reconstruct IMAGE1, IMAGE2 [ neighborhood = 8, inPlace = 0 ]

Performs morphological reconstruction of IMAGE1 under the mask IMAGE2. Images can be two intensity images or two binary images with the same size. The returned image, is an intensity or binary image, respectively.

If boolean inPlace flag is set, IMAGE2 contains the result.

neighborhood determines whether the algorithm assumes 4- or 8- pixel connectivity.

Supported types: Byte, Short, Long, Float, Double

thinning IMAGE

Applies the skeletonization algorithm, returning image with binary object maximal euclidian distance points set.

Supported types: Byte

IPA::Geometry

Contains function for mapping pixels from one location to another

mirror IMAGE [ type ]

Mirrors IMAGE vertically or horizontally, depending on integer type, which can be one of the following constants:

IPA::Geometry::vertical
IPA::Geometry::horizontal

Supported types: all

shift_rotate IMAGE [ where, size ]

Shifts image in direction where, which is one of the following constants

IPA::Geometry::vertical
IPA::Geometry::horizontal

by the offset, specified by integer size.

Supported types: all, except that the horizontal transformation does not support 1- and 4- bit images.

IPA::Misc

Contains miscellaneous helper routines.

split_channels IMAGE, [ MODE = 'rgb' ]

Splits IMAGE onto channels, with the selected MODE, which currently can be 'rgb' or 'hsv' string constants. Returns channels as anonymous array of image objects.

rgb: Supported types: RGB . Returns: 3 Byte images .
hsv: Supported types: RGB . Returns: 3 float images - with hue, saturation, and value . Ranges: hue: 0-360, saturation: 0-1, value: 0-1 .

combine_channels [IMAGES], [ MODE = 'rgb' ]

Combines list of channel IMAGES into single image, with the selected MODE, which currently can be 'rgb' or 'hsv' string constants. Returns the combined image.

rgb: Supported types: Byte . Returns: RGB image .
hsv: Supported types: Float . Returns: RGB image . Channel ranges: hue: 0-360, saturation: 0-1, value: 0-1

histogram IMAGE

Returns anonymous array of 256 integers, each representing number of pixels with the corresponding value for IMAGE.

Supported types: 8-bit

REFERENCES

M.D. Levine. Vision in Man and Machine. McGraw-Hill, 1985.
R. Deriche. Using Canny's criteria to derive a recursively implemented optimal edge detector. International Journal on Computer Vision, pages 167-187, 1987.
R. Boyle and R. Thomas Computer Vision. A First Course, Blackwell Scientific Publications, 1988, pp 32 - 34.
Image Processing Learning Resources. http://www.dai.ed.ac.uk/HIPR2/hipr_top.htm
William K. Pratt. Digital Image Processing. John Wiley, New York, 2nd edition, 1991
John C. Russ. The Image Processing Handbook. CRC Press Inc., 2nd Edition, 1995
L. Vincent & P. Soille. Watersheds in digital spaces: an efficient algorithm based on immersion simulations. IEEE Trans. Patt. Anal. and Mach. Intell., vol. 13, no. 6, pp. 583-598, 1991
L. Vincent. Morphological Grayscale Reconstruction in Image Analysis: Applications and Efficient Algorithms. IEEE Transactions on Image Processing, vol. 2, no. 2, April 1993, pp. 176-201.

AUTHORS

Anton Berezin <tobez@tobez.org>, Vadim Belman <voland@lflat.org>, Dmitry Karasik <dmitry@karasik.eu.org>

	Global
`s`	Focus search bar
`?`	Bring up this help dialog

	GitHub
`g` `p`	Go to pull requests
`g` `i`	go to github issues (only if github is preferred repository)

	POD
`g` `a`	Go to author
`g` `c`	Go to changes
`g` `i`	Go to issues
`g` `d`	Go to dist
`g` `r`	Go to repository/SCM
`g` `s`	Go to source
`g` `b`	Go to file browse

	Search terms
module: (e.g. module:Plugin)
distribution: (e.g. distribution:Dancer auth)
author: (e.g. author:SONGMU Redis)
version: (e.g. version:1.00)

NAME

DESCRIPTION

USAGE

API

IPA::Point

IPA::Local

IPA::Global

IPA::Morphology

IPA::Geometry

IPA::Misc

REFERENCES

AUTHORS

SEE ALSO

Module Install Instructions