#image-processing

Steps of DIP with block diagram

Image acquisition - is an image capturing process by image capturing devices and converted into manageable entity
Enhancement - deblur, denoise, improve contrast, sharpen
Restoration - improving the appearance of image - Based on mathematical and probabilistic models
Color image processing - color models - rgb, cmy etc
Compression - for storage and transmission
Segmentation - partition image into constituent parts or objects - It is most difficult task in DIP - edge detection
Representation and description - e input of this process is output of segmentation - indicate external shape characteristics, such as corners and inflections - Complete region: focus on internal properties
Object detection

image and pixel

Intensity, brightness and contrast

In order to become suitable for digital processing, an image function f(x,y) must be digitized both spatially and in amplitude. Hence in order to create an image which is digital, we need to covert continuous data into digital form. There are two steps in which it is done:
• Sampling
• Quantization
The sampling rate determines the spatial resolution of the digitized image, while the quantization level determines the number of grey levels in the digitized image.

Types of distances

Types of adjancies

Types of zooming

Mathematical transformation are applied to the signals to obtain a further information from the signal that is not readily available in the raw signal. The information that cannot be seen in timedomain can be seen in frequency-domain. Used in image analysis and processing to provide information regarding the rate at which the gray levels change within an image – the spatial frequency content of an image

Polygon approximation techniques:

• Minimum Perimeter Polygons - We visualize this enclosure as two walls corresponding to the outside and inside boundaries of the strip of cells, and think of the object boundary as a rubber band contained within the wall.
• Merging techniques - based on average error - merge points along a boundary until a least square error line of the points merged so far exceeds a preset threshold - procedure repeated, merging new points
• Splitting techniques - subdivide a segment successively into 2 parts until a criterio is met
• Signature - a 1D representation of the boundary in various ways - one way - plot distance from centroid to the boundary as a function of angle - invariant to translation but change with rotation and scaling

The main objective of enhancement is to process an image so that the results is more suitable than original image for a specific application. Enhancement of quality => increasing the dominance of some features ( at the cost of suppression of some features).

Enhancement techniques:
• Point operations : Each pixel is modified by an equation that is not dependent on other pixel values
• Mask operations : Each pixel is modified according to the values in a small neighborhood (subimage)
• Global operations : All pixel values in the image are taken into consideration

Thresholding

1st order derivative

2nd order derivative

Blurring vs. sharpening

Contrast stretching and its graph

grey level slicing

bit-plane slicing

histogram equalisation

Basically, there are two ways to represent a region involves two choices. The external representation is used when the primary focus is on shape characteristics. The internal representation is used when the primary focus is on regional properties.

In image analysis, for the same image the output is the detail description of the scene (in the image). Segmentation is the first step of image analysis techniques. - Segmentation is to subdivide an image into constituent regions or objects. Image segmentation algorithms are generally based on one of the two basic properties of intensity value: discontinuity and similarity.

Different line detection masks

Different region splitting and merging techniques:

• Seeded Region growing
• unseeded region growing
• mean shift algorithm - nonparametric clustering technique - advantage is good in shape of match. And a disadvantage is the slow speed
• fast scanning algorithm - resembles unseeded region growing

redudancy in 2D images

Fidelity criteria

Image compression block diagram

encoding process

decoding process