SCS1
Series
Instruction
Manual
148
The advantages offered by the
Blob analysis
consist in high elaboration speed and possibility of
tolerating and measuring object orientation and dimension variations.
While the constraints consist mainly in the difficulty of distinguishing objects from backgrounds with
low contrasts, objects in touch with each other or overlapped objects as well as the difficulty of
recognising similar objects.
12.6. Contour match
Contour match
recognised objects comparing the contour characteristics. The analysis is carried-out
on binary images with a specific threshold.
The group of black pixels defines the image (BinaryImageSet, BIS).
Figure 32
The contour is defined as a pixel sequence that follows the binary image border. This sequence is
obtained setting relations between adjacent pixels. The contour does not belong to the BIS. The
algorithm can detect both external contours (indicated in red) as well as internal contours (indicated in
green).
A template and some relative information (position, pixel number, dimensions, etc.) has to be acquired
before recognising a target-object. This information is normalised and entered in a vector and then
stored in a database. Object recognition is obtained comparing the target-object’s vector and the
template’s vector previously stored. The distance between the vectors has to be defined to carry-out
the comparison. The vector (in the database) closer to the target-object is selected.
This method guarantees a discrete elaboration speed and can tolerate and measure object orientation
and dimension variations. Moreover, it offers a higher accuracy respect to
Blob analysis
, but has
difficulty in distinguishing objects from backgrounds with low contrasts as well as objects in touch with
each other or overlapped objects.
12.7. Edge detection
A brief introduction to “projection” is necessary.
Edge detection
functions on 256 grey tone values and
not on binary images. Considering that a specific image area can be represented by a matrix of lines
and columns. Each matrix element contains a brightness value associated to each pixel. The following
example has three lines and ten columns.
69 75 78 85 98 120 128 135 133 135
66 69 76 81 88 102 118 129 134 136
68 74 81 87 101 119 128 136 134 135
The sum of each column value is:
203 218 235 253 287 341 374 400 401 406
Phone: 800.894.0412 - Fax: 208.368.0415 - Web: www.ctiautomation.net - e.mail: [email protected]
Phone: 800.894.0412 - Fax: 888.723.4773 - Web: www.clrwtr.com - Email: [email protected]
Phone: 800.894.0412 - Fax: 888.723.4773 - Web: www.clrwtr.com - Email: [email protected]
Phone: 800.894.0412 - Fax: 888.723.4773 - Web: www.clrwtr.com - Email: [email protected]