Computer vision using python and open CV

Dr. Vibhakar Pathak , Professor ,Aman Sharma ,B.Tech. Scholar, Computer Science Department,
Arya College of Engineering & I .T. , Jaipur Rajasthan, India
Email :[email protected]

ABSTRACT:

Computer vision enables a computer to elements (pixels), each with finite, discrete
extract information or insights from the quantities of numeric value.
images, videos or any other visual inputs.
Computer vision is now almost comparable Image Processing:
to human vision, computer vision lacks So the image is just a set of numerical data
human vision as humans have a lifetime of ranging from 0-255, so for digital image
context from which they are able to processing we just perform some steps on the
distinguish between things, whereas in data with the help of our Digital Image
computers we need to train the model with Processing algorithms.
lots and lots of input data by following the examples of such actions are - color
classical machine learning paradigms. transformation, edge detection, morphing,
gray scale leveling, etc.
INTRODUCTION :
Image : An image consists of picture

1
 These are the three main reasons why I have
OpenCv : chosen this library to do research about and
OpenCv is one of the most popular learn some fundamentals about computer
Computer Vision Libraryand a great tool for vision.
digital image processing, it is an open source
library too so that students who are desiring
to learn and understand the concepts of METHODOLOGY :
computer vision and how a computer In this section I am going to write some
actually works can use these powerful and fundamental image processing methods
free tools to have an excellent head start. using openCv functions in python language.

Q. Why OpenCV ?
ans. reasons to use this library are as follows
:
1. Free and open source, everyone can 1. Reading the input image :
use it and learn it. For this we need to import the open
2. It has multiple language support like cv library into our project.
java, python, c++. Like import cv2 as cv
3. OpenCv is heavily used to perform Now we can read the image with in
tasks like object detection, built function imread();
and in self-driving cars that work on Soimage =
Artificial Intelligence. cv.imread(“/User/me/desktop/elon”)

2. Display the image :

2
 We will just use the inbuilt function Object detection :Visual recognition is
of openCvi.eimshow() used for this application i.e the object
imshow(“output”, image) detection here we draw a box let it be a
rectangle or a square and the object reside
inside this box will be detected by
comparing with the trained model data.

3. Color transformation : It distinguishes the object inside the box

Here we try to manipulate the color with other classes' objects, for example if
of the image to have another the box contains a human(person) then it
perspective of the input image and will distinguish it from the animal, car or a
get some information out of it. toy, etc.To do this it performs different
So we write, operations like scaling of the image,
manipulate the lighting condition or we can
call it the color transformation and after
color_Tranform = cv.color(image, these several steps it accurately shows the
cv.COLOR_BGR2HSV) output as text onto the screen.

APPLICATION OF COMPUTER
VISION : LIMITATIONS :
One of the most heavily used domains There is no technology with zero limitation
which uses computer vision primarily is and same applies to the computer vision, i
Object detection. am going to discuss one case study which is

3
able to showcase the loop holes where this
technology lacks CONCLUSION AND FUTURE SCOPE :

As the use cases of artificial intelligence is

Case study of tesla :Elon musk who owns
increasing in our daily life the demand for
and runs the tesla motors, tesla motors is
this domain is increasing heavily, So this
able to achieve a huge milestone when they
domain is going to be the most promising
launch a car which can run on its own or we
domain in the near future.
can say autopilot mode.
So it has been seen that on one day a person It also helps to increase the employment as
driving a Tesla turn on the autopilot mode the need for the person who knows all these
and ride was smooth but after some time it technological stuff like cloud computing,
begins to slow down. machine learning, DL libraries, 3D space
modeling and many more will be increased.

References :

● For future scope :

https://developer.nvidia.com/blog/
the-future-of-computer-vision/
#:~:text=Computer%20vision%20is%20a
%20rapidly,to%20them%20in%20real
%20time.

By this incident we can say that there are ● For object detection :

some cases in which computer vision lacks https://towardsdatascience.com/

and need a lot of improvements. everything-you-ever-wanted-to-

know-about-computer-vision-

4
heres-a-look-why-it-s-so-awesome-
e8a58dfb641e