Scribd
Upload
2 views33 pages

Module 5-4

Module 5 covers morphological image processing, focusing on operations such as erosion, dilation, opening, and closing, which manipulate object boundaries in images. It also discusses the hit-or-miss transform for shape detection and differentiates between quantitative and structural patterns in image classification. Finally, it introduces prototype matching and minimum-distance classifiers for pattern classification.

Uploaded by

Harini Rajendran
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
2 views33 pages

Module 5-4

Module 5 covers morphological image processing, focusing on operations such as erosion, dilation, opening, and closing, which manipulate object boundaries in images. It also discusses the hit-or-miss transform for shape detection and differentiates between quantitative and structural patterns in image classification. Finally, it introduces prototype matching and minimum-distance classifiers for pattern classification.

Uploaded by

Harini Rajendran
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 33

Module 5

Morphological Image Processing


Feature Extraction
Image Pattern Classification
Morphological Image Processing
• Morphological operations are defined in terms of sets.
• In image processing, morphology with two types of sets of pixels:
• objects and
• structuring elements (SE’s).
• Objects are defined as sets of foreground pixels. Structuring elements
can be specified in terms of both foreground and background pixels.
• The concept of set reflection and translation are used extensively in
morphology in connection with structuring elements.
• The reflection of a set (structuring element) B about its origin, denoted
෠ is defined as
by 𝐵,
Erosion and Dilation
• The erosion and dilation operations are fundamental to
morphological processing.
• Erosion: Removes pixels from the boundaries of objects in an
image. This can be used to remove small objects or thin
structures, separate overlapping objects, and find object
boundaries.
• Dilation: Adds pixels to the boundaries of objects in an image.
• erosion, which is a shrinking or thinning operation, dilation
“grows” or “thickens” objects in a binary image. Dilation expands
the components of a set.
Erosion
Dilation
Opening and Closing
• The other 2 important morphological operations:
• opening and closing.
• Opening generally smoothes the contour of an object, breaks
narrow isthmuses, and eliminates thin protrusions.
• Closing also tends to smooth sections of contours, but, as
opposed to opening, it generally fuses narrow breaks and long thin
gulfs, eliminates small holes, and fills gaps in the contour.
Equation (9-10) has a simple geometrical interpretation: The opening of A by B is the union of all
the translations of B so that B fits entirely in A.
Hit-or-Miss transform
• The morphological hit-or-miss transform (HMT) is a basic tool for shape
detection.
• Let I be a binary image composed of foreground (A) a background pixels,
respectively.
• The hit-or-miss transform is defined as follows: ⊛ = ( ⊖ ) ∩ ( ⊖)
• This transform is useful in locating all pixel configurations that match the B1
structure (i.e a hit) but do not match that of B2 (i.e. a miss).
PATTERNS AND PATTERN CLASSES
• In image pattern classification, the two principal pattern
arrangements are quantitative and structural.
• Quantitative patterns are arranged in the form of pattern vectors.
• Structural patterns typically are composed of symbols, arranged
in the form of strings, trees, or, less frequently, as graphs.
PATTERN VECTORS
• Pattern vectors are represented by lowercase letters, such as x, y,
and z, and have the form
• Pattern vectors can be formed directly from image pixel intensities
by vectorizing the image
STRUCTURAL PATTERNS

• Pattern vectors are not suitable for applications in which objects are
represented by structural features, such as strings of symbols.
• Although they are used much less than vectors in image processing
applications, patterns containing structural descriptions of objects
are important in applications where shape is of interest.
PATTERN CLASSIFICATION BY PROTOTYPE
MATCHING
• Prototype matching involves comparing an unknown pattern
against a set of prototypes, and assigning to the unknown pattern
the class of the prototype that is the most “similar” to the
unknown. Each prototype represents a unique pattern class, but
there may be more than one prototype for each class.
MINIMUM-DISTANCE CLASSIFIER

• One of the simplest and most widely used prototype matching


methods is the minimum-distance classifier which, as its name
implies, computes a distance-based measure between an
unknown pattern vector and each of the class prototypes.
• It then assigns the unknown pattern to the class of its closest
prototype.
• The prototype vectors of the minimum-distance classifier usually
are the mean vectors of the various pattern classes:

You might also like