Legacy Documentation

Digital Image Processing (2000)

This is documentation for an obsolete product.
Current products and services
User's Guide / Area Operations /
Previous section-----Next section

5.6 Region-of-Interest Processing

It is sometimes of interest to process a single subregion of an image leaving other regions unchanged. This is commonly referred to as region-of-interest (ROI) processing. ROI processing is distinctly different from BlockProcessing where a selected area operator is applied to all rectangular subregions of an image. The subregions in block processing are assumed to completely tile the image. Conversely, in ROI processing a single region, frequently irregularly shaped, is of interest. Function RegionProcessing implements the basic features of ROI processing.

Region-of-interest processing functions.

Image subregions may be conveniently specified by using Mathematica Graphics primitives, such as Circle, Disk, Point, Line, Rectangle, Polygon, or simply as a list of vertex positions. We follow with a few demonstrations.
This loads the package.
In[1]:=
Here we create a small data array.
In[2]:=
Out[2]//MatrixForm=Out[2]//MatrixForm=
Point operations are mapped to every point in a ROI by using the function RegionPoints, which returns a list of all coordinate pairs within a region, given a region specification. Here is an example of using RegionPoints to extract all diagonal elements of a matrix.
In[3]:=
Out[3]=Out[3]=
In[4]:=
Out[4]=Out[4]=
Finally, we show an example of edge detection in a triangular ROI using a real image. This loads the example image.
In[5]:=
Image regions may be conveniently selected using several GUIKit-based user interfaces specifically designed for this purpose. These interfaces permit the selection of any image regions that map to the available 2D graphics primitives Point, Line, Circle, Rectangle, Polygon, and Disk. The regions may be selected using simple click or click-and-drag actions.

Region selection functions.

Here we define a triangular ROI using a polygon selection tool.
In[6]:=
Out[6]=Out[6]=
This shows the result of applying a Sobel edge detector to the region defined by roi.
In[7]:=
Out[7]=Out[7]=
As a second example, consider the application of a threshold operator to an irregular ROI. The vertices were selected using PolygonSelectGUI.
In[8]:=
Out[8]=Out[8]=
This shows the result of applying the threshold operator to the ROI.
In[9]:=
Out[9]=Out[9]=
Here is another example of ROI processing.

LineProfile functions.

The function LineProfile is useful in obtaining details of amplitude variations (i.e., intensity profile) in an image by returning pixel values along some arbitrary user-selected path. Examples of row or column profiles were shown in Sections 2.2 and 2.5.
Here we load an example image.
In[10]:=
Here we use LineProfile to extract data values along any path defined by Line. The path was selected using LineSelectGUI.
In[11]:=
This displays the original image with a thin red line depicting the selected path and the intensity profile along the selected path.
In[12]:=
Out[12]=Out[12]=