You are reading the documentation for version 1.1 of OpenStructure. You may also want to read the documentation for: 1.2 1.3 1.4 1.5 1.6 1.7 1.7.1 1.8 1.9 1.10 1.11 2.0 2.1 2.2 devel

# img - Images and Density Maps¶

## Introduction : The ImageHandle¶

OpenStructure offers extensive processing capabilities for planar 2d images and 3d maps using the img module. The main datastructure for images is the ImageHandle class.

ImageHandles provide a clean and efficient interface to interact with images and maps. An ImageHandle can store an image in either real (SPATIAL) or Fourier (FREQUENCY) space and always keep track of the currently active domain. This means,for example that one can apply a Fourier transformation to an ImageHandle containing a SPATIAL image and the image will correctly identify the new active domain as FREQUENCY. The ImageHandle also understands, for example, that applying a Fourier Transform to a centrosymmetric FREQUENCY image results in a real SPATIAL image, but applying it to a non-centrosymmetric one results in a complex SPATIAL image.

Furthermore, the image will make sure that real and Fourier space information about the image are always in sync. If, for example, the pixel sampling is changed while the current active domain is real space, the pixel sampling in Fourier space will be adjusted accordingly, and vice versa.

## Creating and visualizing ImageHandles¶

As a first step, enter the following lines in the OpenStructure python console:

im=img.CreateImage(img.Size(200,200))


This will create an empty, 2D image, with a height and width of 200 pixels, whose origin (ie the pixel with the coordinates <0,0>) is in the top-left corner.

v=gui.CreateDataViewer(im)


A viewer window will pop up (see below), showing a white frame on a black background. The inner area of the white frame is the image, which is empty.

## Reading and writing into an image¶

Data can be read and written from and into an image using the following commands:

# writes the real value 23.4 into pixel 10,10
im.SetReal(img.Point(10,10),23.4)
# reads the value in pixel 10,10
val=im.GetReal(img.Point(10,10))


The complex equivalents are also available

# writes the complex value value 2+3j into pixel 10,10
im.SetComplex(img.Point(10,10),2+3j)
# reads the value in pixel 10,10
val=im.GetComplex(img.Point(10,10))


The image knows in which domain it is, and will adjust the type of data being written accordingly. For example, if one writes a complex value in a SPATIAL image, the value will be automatically converted to a real one by taking the amplitude of the complex number. On the other hand, if one writes a real value in a FREQUENCY image, the value is automatically converted to complex by setting the imaginary part to 0.

## Image properties¶

### The data domain¶

The data domain of an image specifies wether the image contains data in the spatial or frequency domain. A HALF_FREQUENCY domain also exists, representing centrosymmetric frequency data (such as the data coming from the Fourier transform of an image from the real spatial domain)

ost.img.SPATIAL

Real-valued spatial images

ost.img.COMPLEX_SPATIAL

Complex-valued spatial images, i.e. resulting from a Fourier transform of the FREQUENCY domain.

ost.img.FREQUENCY

Complex frequeny domain.

ost.img.HALF_FREQUENCY

Centrosymmetric frequency images

### The spatial origin¶

The spatial origin of an image is the first pixel of its extent. Specifically, this is the top left pixel for 2D images and the top-front-left corner for 3D images.

### The absolute origin¶

The absolute origin of an image describes the coordinates, in the absolute reference system used by OpenStructure, of the pixel in with all indexes equal to 0. Please note that the pixel does not necessarily need to belong to the extent of the image.

### Pixel sampling¶

The pixel sampling property of an image describes the size of its pixels. For the same image, the size of pixels in the SPATIAL and in the FREQUENCY The data domain are obviously interdependent. OpenStructure takes care of the transformation and allows access to both pixel sampling irrespective of the current image domain.

## The ImageHandle class¶

The public interface of the ImageHandle class provides many ways to manipulate image properties. What follows is a brief description of the most important methods and attributes of the ImageHandle class.

class ost.img.ImageHandle
size

The size of the image. Read-only.

Type : Size
extent

The extent of the image. Read-only.

Type : Extent
type

The DataType of an image represents the nature of the data it contains. An image can contain :objREAL or COMPLEX values.

absolute_origin

The absolute origin of an image describes the coordinates, in the absolute reference system used by OpenStructure, of the pixel in with all indexes equal to 0. Please note that the pixel does not necessarily need to belong to the extent of the image. Read-write.

Type : Vec3
spatial_origin

The spatial origin of an image is the first pixel of its extent. Specifically, this is the top left pixel for 2D images and the top-front-left corner for 3Dimages.

Type : Point
domain

The current domain of the image. See The data domain. Read-only.

Apply(algorithm)

Applies an algorithm on an image and returns a new ImageHandle containing the modified image. See img.alg - Image Processing Algorithms

Parameters: algorithm (Instance of an algorithm class - See img.alg - Image Processing Algorithms.) – algorithm ImageHandle
ApplyIP(algorithm)

Applies an algorithm on an image, overwriting the current content. See :doc:../alg/alg

Parameters: algorithm (Instance of an algorithm class - See: doc:../alg/alg.) – algorithm
CenterSpatialOrigin()

Sets the The spatial origin of an image in such a way that the central pixel has all 0 indexes, e.g. (0,0) or (0,0,0) for 3d images.

CoordToIndex(coord)

Returns the indexes of an image corresponding to the specified absolute coordinates. (See The absolute origin). A given set of absolute coordinates will almost never fall exactly at the center of a pixel, so this method return fractional indexes.

Parameters: coord (Vec3) – Absolute coordinates Vec3
Copy()

Creates a and returns a copy of the ImageHandle. The new handle does not point to the same underlying data as the old one, but to a complete and separate copy of the original data.

Return type: ImageHandle
Extract(extent)

Creates and returns a new image that contains a copy of a portion of the original image. The extracted image keeps the same data-type of the original image, but extractions from images in the ‘FREQEUNCY’ or ‘HALF FREQUENCY’ domains result in COMPLEX  :obj:SPATIAL images. This transformation is necessary, since the there is no guarantee that the extracted FREQUENCY sub-image is centered around the origin and hence back-transformable to SPATIAL.

Parameters: extent (Extent) – Portion of the image to extract ImageHandle
FractionalIndexToCoord(frac_pixel)

Same as IndexToCoord(), but introduces subpixel precision by accepting fractional numbers for pixel indexes.

Parameters: frac_pixel (Vec3) – Fractional pixel indexes Vec3
GetAbsoluteOrigin()

Returns the The absolute origin of an image

Return type: Vec3
GetComplex(pixel)

Returns the complex value of the specified image pixel. If the image holds data of the ‘REAL’ data-type, the method return s complex value with the pixel content as real part and a null imaginary part.

Parameters: pixel (Point) – Image pixel complex
GetDomain()

See domain

GetExtent()

See extent

GetFrequencySampling()

Returns the Pixel sampling of the image in the FREQUENCY The data domain

Return type: Vec3
GetIntpolComplex(frac_pixel)

Returns the interpolated complex value of the virtual pixel corresponding to the specified fractional indexes. This is computed by calculating a weighted vector sum of the values of the surrounding pixels. If the image holds data of the ‘REAL’ data-type, the method computes the interpolated value using bilinear interpolation (trilinear for 3D images), then returns a complex value with the interpolated value as real part and a null imaginary part.

Parameters: frac_pixel (Vec3) – Fractional pixel indexes complex
GetIntpolReal(frac_pixel)

Returns the interpolated value of the virtual pixel corresponding to the specified fractional indexes. This is computed by using bilinear interpolation (trilinear for 3D images). If the image holds data of the COMPLEX  :ref:data-type, the method computes the interpolated value as a weighted vector sum of the values of the surrounding pixels, then returns the amplitude of the interpolated value.

Parameters: frac_pixel (Vec3) – Fractional pixel indexes float
GetPixelSampling()

Returns the Pixel sampling of the image in the current The data domain.

Return type: Vec3
GetReal(pixel)

Returns the value of the specified image pixel. If the image holds data of the COMPLEX  :ref:`data-type, the method return the amplitude of the pixel content.

Parameters: pixel (Point) – Image pixel float
GetSize()

Returns the Size of the image.

Return type: Size
GetSpatialOrigin()

Returns the The spatial origin of the image.

Return type: Point
GetSpatialSampling()

Return the Pixel sampling of the image in the SPATIAL The data domain.

Return type: Vec3
GetType()

Returns the data-type of the image (REAL or COMPLEX)

Return type: DataType ???????????
IndexToCoord(pixel)

Returns the absolute coordinates (See The absolute origin) corresponding to the pixel with the specified indexes. Please note this method always returns the coordinates corresponding to the center of the pixel. When sub-pixel precision is needed, the method FractionalIndexToCoord() can be used.

Parameters: pixel (Point) – Vec3
IsFrequency()

Returns true if the The data domain of the image is FREQUENCY or HALF-FREQUENCY

Return type: bool
IsSpatial()

Returns true if the The data domain of the image is SPATIAL.

Return type: bool
IsValid()

Returns true, when the image handle is valid, false if not. It is an error to call any method on the image handle other than IsValid when this method returns false.

Return type: bool
Paste(source_image)

Copies the content of a different image into the current one, overwriting pre-existing data . The values of pixels with given indexes in the source image are copied into pixels with the same indexes in the target image. If the two images have different extents (see Extent), pixels that do not exist in both are ignored. Please notice that this method only copies the pixel content: all other properties of the image are left untouched.

Parameters: source_image (ImageHandle) – Source image that gets pasted into the current one
SetAbsoluteOrigin(coord)

Sets the The absolute origin of the image to the specified coordinates

Parameters: coord (Vec3) – Absolute coordinates
SetComplex(pixel, value)

Sets the content of the specified pixel to the provided value. If the image holds data of the ‘REAL’ data-type, the method sets the pixel to the amplitude of the provided. value.

Parameters: pixel (Point) – Image pixel value (complex) – Value
SetPixelSampling(sampling)

Sets the Pixel sampling of the image to the provided values in the current The data domain.

Parameters: sampling (Vec3) – Size of a pixel in x,y and z direction respectively
SetReal(pixel, value)

Sets the content of the specified pixel to the provided value. If the image holds data of the COMPLEX data-type, the method sets the pixel to a value has a real part equal to the provided value and a null complex part.

Parameters: pixel (Point) – Image pixel value (float) – Value
SetSpatialOrigin(pixel_indexes)

Sets the The spatial origin of the image to the provided indexes.

Parameters: pixel_indexes (Point) – Indexes of the first pixel of the image None
SetSpatialSampling(sampl)

Sets the Pixel sampling if the image to the provided values in the spatial The data domain

Parameters: sampl (Vec3) – Size of a pixel in x,y and z direction respectively

## Search

Enter search terms or a module, class or function name.

## Previous topic

conop – Connectivity and Topology of Molecules

## Next topic

Point, Size, Extent