Research describes as to what Image processing consist of.

project on investigation of Image processing overview and its method

























We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

There has been well noted research carried over since the
first algorithm for face detection ever witnessed by the world was invented by Takeo
Kanade. As we know, everything that exist has some sort of signals that defines
its geometry, location or orientation of it which is nothing but signals and is
a part of Signal processing, mathematical analysis and computer science field.
So far there are various patented face detection and face recognition algorithm
or more generally computer vision, which is continuously enhanced by reducing
time of face detection on any image. Moreover, the field is further classified
into supervised, trained images database and detection, and unsupervised, no
classifier or database required, learning methods. There is various methodology
when it comes to face detection since no method can prove to be best for any
condition thus, face detection has many different approaches such as face
detection in images with controlled background, finding face by color, finding
faces by motion and finding faces in unconstrained scenes which includes edge
orientation face detection, hausdorff distance face detection and week classifier
cascade. With every research there came an improvement over the stated method
of its accuracy for example, viola jones algorithm is an improvement over
haar-like features (week classifier) for face detection by using Adaboost and
integral imaging 1.

It is important to know the foundation and its constituent before
even thinking for making a building. Image processing forms the basis for any form
of vision be it computer vision, video processing etc. To be specific computer
vision is used when detection, recognition is to be carried while manipulating
an image such as deblurring etc are to be carried. Thus, in this research is
carried out what exactly is Image processing, reading, writing and processing
Image over on the grounds of its representation in gray, HSV, ycbcr color
space. Moreover, through this research region of interest and threshold
manipulation will be carried. All this research will be constructive towards
future work in the field of Image processing which will be listed in the
upcoming chapter of this research project. To describe the world of AI, Computer
vision and Image processing, figure 1 describes as what constitutes to be an
Artificial intelligence.

Fig 1: Artificial Intelligence classification (source: Neota Logic)

Figure 2 classifies components of computer vision namely AI,
image processing, Patter recognition and so on and finally figure 3 describes
as to what Image processing consist of.

Fig 2: Computer vision components (source: KDnuggets by Pulkit Khandelwal,
VIT university)

Fig 3: Digital Image processing components (source: Shunji Murai, Professor
and Dr. Eng. Online course)


Though there are possibilities to do Image processing using Octave,
python, java, C, C++, C# and moreover it is highly recommended to use python
nowadays as it is the simplest software language. But MATLAB stands out from
these languages as MATLAB offers concise coding for this functioning without
invoking new directories or new .dll for image recognition or any of the
purpose that is needed for this task completion. Moreover, MATLAB is more
reliable on mathematics and saving features than C C++ and C#. Example to save
two numbers, in C++/C# it is needed to invoke pre-defined functions then add
namespace, then add class, then define two variables, add them, save into other
variable, take care of errors free coding and then execute.

While same thing in MATLAB will just need a=2, b=3, sum=a+b;
with no bracket involvement and other error that is needed to take care of.
Thus, saving time in thinking over logic than thinking over code error or
conversion problems. MATLAB has already built in function for almost all my
requirement, that saves time while coding for the same. While on other hand,
though C# is best for more lengthy programs but that requires invoking new
libraries, threads etc. MATLAB also offers direct code conversion to C++.

Though to do a work, any tool could be used because what
exactly matters is the work. But, during this research MATLAB platform will be used.

processing components

i) Image Sensors

reference to sensing, two elements are required to acquire digital

image. The
first is a physical device that is sensitive to the energy

by the object we wish to image and second is specialized

processing hardware.

ii) Specialize image processing hardware

consists of the digitizer just mentioned, plus hardware that performs other

operations such as an arithmetic logic unit, which performs

such addition and subtraction and logical operations in parallel on


iii) Computer

It is a general-purpose
computer and can range from a PC to a

depending on the application. In dedicated applications,

specially designed computer is used to achieve a required level of


iv) Software

It consists
of specialized modules that perform specific tasks a well designed

also includes capability for the user to write code, as a minimum,

the specialized module. More sophisticated software packages allow

integration of these modules.

v) Mass storage

capability is a must in image processing applications. An image of size

1024 x1024
pixels, in which the intensity of each pixel is an 8- bit quantity

one megabytes of storage space if the image is not compressed.

processing applications falls into three principal categories of storage

i) Short
term storage for use during processing

ii) On
line storage for relatively fast retrieval

Archival storage such as magnetic tapes and disks

vi) Image displays

displays in use today are mainly color TV monitors. These monitors are

driven by
the outputs of image and graphics displays cards that are an

part of computer system

vii) Hardcopy devices

devices for recording image includes laser printers, film cameras, heat

devices inkjet units and digital units such as optical and CD ROM disk.

provide the highest possible resolution, but paper is the obvious medium

of choice
for written applications.

viii) Networking

It is
almost a default function in any computer system in use today because of

the large
amount of data inherent in image processing applications. The key

consideration in image transmission

steps in Digital Image processing

There are
two categories of the steps involved in the image processing

1. Methods
whose outputs are input are images.

2. Methods whose outputs are
attributes extracted from those images.

Table 1: Fundamental steps in Digital Image Processing

Image acquisition

It could
be as simple as being given an image that is already in digital form. Generally,

image acquisition stage involves
processing such as scaling.

Image Enhancement

It is
among the simplest and most appealing areas of digital image processing. The

this is to bring out details that are obscured or simply to highlight certain

of interest in image. Image enhancement is a very subjective area of image


Fig 4:
Before and after Image enhancement

Image Restoration

It deals
with improving the appearance of an image. It is an objective approach, in the

sense that
restoration techniques tend to be based on mathematical or probabilistic

models of
image processing. Enhancement, on the other hand is based on human

subjective preferences regarding what
constitutes a “good” enhancement result

Fig 5: Image restoration input and output

Color image processing

It is an
area that is been gaining importance because of the use of digital images over

Color image processing deals with basically color models and their

in image processing applications.

and Multiresolution Processing

These are the foundation for
representing image in various degrees of resolution


It deals
with techniques reducing the storage required to save an image, or the

required to transmit it over the network. It has to major approaches:

Lossless Compression

b) Lossy Compression

Morphological processing

It deals
with tools for extracting image components that are useful in the
representation and

of shape and boundary of objects. It is majorly used in automated inspection