1. 3. Introduction to Digital Image Analysis

2. CONTENT Segmentation Introduction Preprocessing Discrete Transform
System Model
Preprocessing
ROI Image Geometry
Image Algebra
Spatial Filters
Image Quantization
Edge/Line Detection
Roberts Operator
Sobel Operator
Prewitt Operator
Kirsch Compass Masks
Robinson Compass Masks
Laplacian Operators
Frei-Chen Masks
Edge Operator Performance
Hough Transform
Segmentation
Region Growing and Shrinking
Clustering Techniques
Boundary Detection
Combined Approaches
Morphological Filtering
Discrete Transform
Fourier Transform
Cosine Transform
Walsh-Hadamard Transform
Filtering
Wavelet Transform
Feature Extraction and Analysis
Feature vectors & Feature Spaces
Binary Object Features
Histogram Features
Color Features
Spectral Features
Pattern Classification

3. INTRODUCTION
Image analysis involves manipulating the image data to determine exactly the information necessary to help solve a computer imaging problem.
This analysis is typically part of a larger process, is iterative in nature, and allows us to answer application specific questions:
How much spatial & brightness resolution is needed?
Will existing methods solve the problem?
Is color information needed?
Do we need to transform the image data into freq. domain?
Do we need to segment the image to find object info?
What are the important features in the images?
Is the hardware fast enough for the application?

4. IA is primarily a data reduction process
In CV, end product is typically the extraction of high level information for computer analysis or manipulation
High level information like shape parameters
In IP, IA methods may be used to help determine the type of processing required & the specific parameters needed for that processing
Eg, enhancement, degradation, compression

5. System Model can be broken down into three primary stages:
Preprocessing
Data Reduction
Feature Analysis .
Input image
Preprocessing
Data Reduction
Feature Analysis

6. INTRODUCTION
Preprocessing - to remove noise and eliminate irrelevant, visually unnecessary information.
Noise is unwanted information that can result from the image acquisition process.
Other preprocessing steps might include gray-level or spatial quantization (reducing the number of bits per pixel or the image size) or
finding regions of interest for further processing.

7. INTRODUCTION
Data reduction - reducing the data in the spatial domain or transforming it into another domain called the frequency domain (Figure below) and
then extracting features for the analysis process.
Input image
Preprocessing
Frequency (Spectral)
Domain
Feature Analysis
Spatial

8. INTRODUCTION
Feature analysis - features extracted by the data reduction process are examined and evaluated for their use in the application.

9. Preprocessing
To makes the primary data reduction and analysis task easier.
They include operations related to extracting regions of interest, performing basic algebraic operations on images, enhancing specific image features and reducing data in both resolution and brightness.
a stage where the requirements are typically obvious and simple, such as the removal of artifacts from images, or the elimination of image information that is not required for the application.
For example, in one application we needed to eliminate borders from the images that had been digitized from film (the film frames); in another we had to mask out rulers that were present in skin tumor slides.
Another example of a preprocessing step involves a robotic gripper that needs to pick and place an object; for this, we reduce a gray-level image to a binary (two-valued) image that contains all the information necessary to discern he object's outline.