RAJIV GANDHI PROUDYOGIKI VISHWAVIDYALAYA,

BHOPAL New Scheme Based On AICTE

Flexible Curricula Computer Science and Engineering,

Scan for YouTube

Unit lll

Explain Region Analysis: Region properties, External points, Spatial moments

Region
gion analysis involves the study and evaluation of the properties of a particular region or area
in an image. Regions can be any shape or size and may have different properties that are of
interest to the analyst. Region analysis is an important tool for u
understanding
nderstanding the properties of
an image and can be used in a variety of applications, including image processing, computer
vision, and pattern recognition.

1. Region Properties:
2. These are basic features of a region that describe its size and shape. These pro properties
perties include:
 Area: The area of a region is the number of pixels contained within it. It is usually expressed in
square pixels.
 Perimeter: The perimeter of a region is the length of its boundary. It can be used to estimate the
shape of the region and the e degree of its irregularity.
 Centroid: The centroid of a region is the center of mass of the region. It can be used to estimate
the position of the region in the image.
 Bounding Box: The bounding box is the smallest rectangle that encloses the region. It can be
used to estimate the size and shape of the region.
2. External Points: These are points that lie outside the region but are used to describe its
properties. External points can be used to compute the eccentricity of the region, which is a
measure of how w elongated or circular it is. Other external points include the major and minor
axes of the region, which can be used to estimate its orientation.
3. Spatial Moments: These are mathematical calculations that are used to describe the shape and
position of a region. The zeroth--order
order spatial moment is equal to the area of the region, while
the first-order
order spatial moments are used to calculate the centroid of the region. Higher-order
Higher
spatial moments can be used to estimate the shape and orientation of the region.
4. Mixed Spatial Gray-Level
Level Moments: These are calculations that take into account both the
spatial position and the intensity values of the pixels in the region. They are used to evaluate the
texture and contrast of a region. These moments can be used to es estimate
timate the texture and
contrast of the region, which can be useful in applications such as medical image analysis.

In summary, region analysis is an important tool for understanding the properties of an image
and can be used in a variety of applications. R
Region
egion properties, external points, spatial moments,

YouTube Channel: RGPV

PV Exam official 2023 https://www.youtube.com/@RGPVExamOfficial
 and mixed spatial gray-level moments are some of the key properties of regions that can be
analyzed to extract useful information from images.

Explain Boundary analysis: Signature properties, Shape numbers

Boundary analysis is a technique used to study and analyze the properties of the boundaries of
objects or regions in an image. It involves extracting information from the boundaries of objects
in order to classify, identify, or segment them. Two key properties of boundary analysis are
signature properties and shape numbers.

1. Signature Properties: Signature properties refer to a series of numbers that represent the shape
and structure of a boundary. The signature can be calculated using various methods, such as
Fourier descriptors or complex moments. Signature properties can include curvature, tangent
angles, and boundary length. They can be used to classify and identify objects based on their
shape, as well as to track object movement and deformation.
2. Shape Numbers: Shape numbers are numerical descriptors that are used to quantify the shape
of a boundary. They are often calculated using geometric features such as area, perimeter, and
compactness. Some commonly used shape numbers include circularity, aspect ratio, and
eccentricity. These shape numbers can be used to distinguish between different types of objects
or to compare the shapes of different regions in an image.

Here are some additional details on these properties:

 Curvature: Curvature is a measure of the rate of change of the tangent angle along the
boundary. It can be used to distinguish between curved and straight boundaries, as well as to
detect corners or inflection points in the boundary.
 Tangent angles: Tangent angles are the angles between the tangent line and the horizontal axis
at each point along the boundary. They can be used to quantify the orientation of the boundary
and to identify symmetrical patterns.
 Boundary length: Boundary length is the length of the boundary of an object. It can be used to
estimate the size of an object or to track its movement over time.
 Circularity: Circularity is a measure of how closely a boundary resembles a perfect circle. It is
calculated as the ratio of the object's perimeter to the perimeter of a circle with the same area. A
perfectly circular object will have a circularity of 1, while more irregular objects will have lower
circularity values.
 Aspect ratio: Aspect ratio is a measure of the elongation of an object. It is calculated as the ratio
of the object's height to its width. Objects that are longer in one direction than in the other will
have high aspect ratios, while objects that are roughly square will have aspect ratios close to 1.
 Eccentricity: Eccentricity is a measure of how elongated or flattened an object is. It is calculated
as the ratio of the distance between the foci of an ellipse that fits the object to its major axis
length. Objects that are more elongated will have higher eccentricities, while more circular
objects will have eccentricities close to 0.

YouTube Channel: RGPV Exam official 2023 https://www.youtube.com/@RGPVExamOfficial

 In summary, signature properties and shape numbers are important properties of boundary
analysis. They can be used to characterize the shape and structure of objects in an image, and to
distinguish between different types of objects based on their shape. These properties can be
used in a wide range of applications, including pattern recognition, image segmentation, and
object tracking.

Explain General Frame Works For Matching

The general frame for matching is a set of techniques used to compare and match objects or
patterns in an image with a reference template or database. The general frame consists of four
main components: the distance relational approach, ordered structural matching, view class
matching, and models database organization.

1. Distance relational approach: The distance relational approach involves comparing the
distances between feature points in the reference and target images. These distances can be
used to compute a similarity score, which indicates how well the target image matches the
reference image. The distance relational approach is commonly used for point-based features,
such as corners or edges.
2. Ordered structural matching: The ordered structural matching approach involves comparing
the structures of the reference and target images. This involves identifying the objects or regions
in the image and comparing their relative positions and shapes. The ordered structural matching
approach is commonly used for shape-based features, such as contours or silhouettes.
3. View class matching: The view class matching approach involves matching objects or patterns
across different viewing angles or orientations. This involves building a database of views or
templates for each object or pattern, and then comparing the target image to the appropriate
view or template in the database. The view class matching approach is commonly used for 3D
object recognition or tracking.
4. Models database organization: The models database organization involves organizing the
reference templates or models in a database for efficient searching and retrieval. This involves
indexing the templates based on their features, such as shape or texture, and using efficient
search algorithms, such as k-d trees or hash tables, to quickly retrieve the closest matches.

In summary, the general frame for matching provides a set of techniques for comparing and
matching objects or patterns in images. These techniques include the distance relational
approach, ordered structural matching, view class matching, and models database organization.
Each technique has its strengths and weaknesses, and the appropriate technique depends on
the specific application and the features of the objects or patterns being matched.

YouTube Channel: RGPV Exam official 2023 https://www.youtube.com/@RGPVExamOfficial