ACKNOWEDGEMENT

The knowledge and satisfaction that accompany the successful completion of

any task would be incomplete without mention of people who made it possible,
whose guidance and encouragement crowned our effort with success. We would
like to thank all and acknowledge the help we have received to carry out this mini
project. We would like to convey our thanks to the Principal Dr. Aswath.M.U and
Head of Department Dr. Jyothi.D.G for being kind enough to provide the necessary
support to carryout the mini Project.
We would also like to acknowledge the support we have received from the
mini project coordinators Prof. Shruthiba A and Prof. Sahana M for
understanding the requirements to be met at each phase of the mini project.
We are most humbled to mention the enthusiastic influence provided by
our guide Prof. Shruthiba A on the mini project for the ideas, time to time
suggestions for being a constant support and co-operation showed during the
venture and making this mini project a great success. We would also take this
opportunity to thank my friends and family for their constant support and help.
We are very much pleasured to express our sincere gratitude to the friendly
co-operation showed by all the staff members of Artificial Intelligence and Machine
Learning Department, BIT.
 ABSTRACT

The main objective of License Plate Detection project is to reduce the criminal
activity like stolen vehicle, road traffic monitoring that involve use of motor vehicles. In
this project we will be processing the image of the vehicle such that the vehicle number
from the image will be extracted. This report discuses a method for the vehicle number
plate recognition from the image using mathematical morphological operations. The main
objective is to use different morphological operations in such a way that the number plate
of vehicle can be identified accurately. This is based on various operation such as image
enhancement, morphological transformation, edge detection and extraction of number
plate from vehicle image. After this segmentation is applied to recognize the characters
present on number plate.
CHAPTER 1
INTRODUCTION

License plate recognition is a part of digital image processing which is generally

used in vehicle transportation system to categorize the vehicle. License plate recognition
systems are having varieties of application such as traffic maintenances, tracing stolen
cars, automatic electronic Toll collection system etc. But the main aim is to control the
traffic management system. In India the traffic management system is developing day by
day. License plates of various styles and colors are found in different countries. The
license plate format of every country is different. So every country develops an
appropriate LPR system which suits their transport vehicle license plate format. In India,
the License plate containing white background with black foreground color is used for
private cars and for the commercial vehicles yellow is used as background and black as
foreground color. The goal of license plate detection is to automatically locate and extract
license plate regions from images. This comprehensive introduction aims to provide an
in-depth overview of license plate detection, its significance and popular techniques
employed in the field.

In License plate recognition, a camera captures the vehicle images and a computer
processes them and recognizes the information on the number plate by applying various
image processing and optical character recognition techniques. Prior to the character
recognition, the number plates must be separated from the background images. This task
is considered as the most crucial step in the License plate recognition system, which
influences the overall accuracy and processing speed of the whole system significantly.
Since there are problems such as poor image quality, image perspective distortion, other
disturbance characters or reflection on vehicle surface, and the color similarity between
the number plate and the background vehicle body, the number plate is often difficult to
be located accurately and efficiently.

Generally vehicle number plate recognition is divided into several steps including
number plate extraction,image region which contains a number plate, character
segmentation, and character recognition. Generally, in order to recognize a vehicle
number plate, the region of the number plate should be extracted from a vehicle image.
Accurate detection of the plate region is essential process to go over to the step of
character recognition.

Dept.,of AI&ML(2020 Batch) 3 2022-23

Dept.,of AI&ML(2020 Batch) 4 2022-23
CHAPTER 2
LITERATURE REVIEW
PAPER 1:
Title: "Automatic Number Plate Recognition" ,Rai, Vanshika, and Deepali Kamthania.
In Proceedings of the International Conference on Innovative Computing &
Communication (ICICC). 2021

Summary:
In this paper an attempt has been made to develop an automatic number plate
detection and recognition system for Indian vehicles. The proposed system first detects
the vehicle and then captures the vehicle image. Vehicle number plate region is extracted
using the image segmentation and characters are recognized using optical character
recognition technique. The system can handle noisy, low illuminated, cross angled, non-
standard font number plates. The morphological transformation, Gaussian smoothing, and
Gaussian thresholding the different image processing techniques, has been used in the
pre-processing stage. The contours have been applied by border following and contours
are filtered based on character dimensions and spatial localization for number plate
segmentation and then for character recognition K-nearest neighbour algorithm has been
used. The proposed system has good accuracy and can be used for e-challan surveillance,
stolen vehicle detection and many other applications.

PAPER 2:
Title: "Vehicle number plate detection using image processing",Singh, Abhay, Anand
Kumar Gupta, Anmol Singh, Anuj Gupta, and Sherish Johri. International Research
Journal of Engineering and Technology (IRJET) 5, no. 03 (2018).
Summary:
In this technology we will be working on input images given. The These input
images are converted to grayscale and characters are segmented and recognised using
OCR. There are some conditions for this software to work:
1) Vehicle plates should be white and according to the rules given by the government of
India.
2) Image should be of appropriate brightness and contrast: In this, a software is designed
which detects the vehicle number plate number using MATLAB.

Dept.,of AI&ML(2020 Batch) 5 2022-23

 License Plate Detection using Image Processing
In this technique we will be performing several methods step by step to find the vehicle
number. Then using that vehicle number found we will be comparing that number from
our database.
PAPER 3:
Title: "Automatic vehicle number plate recognition for vehicle parking management
system." Jadhav, Ganesh R., and Kailash J. Karande. Computer Engineering and
Intelligent Systems, ISSN (2014): 2222-1719.
Summary:
This paper discusses a method for the vehicle number plate recognition from the
image using mathematical morphological operations. The main objective is to use
different morphological operations in such a way that the number plate of vehicles can be
identified accurately. This is based on various operations such as image enhancement,
morphological transformation, edge detection and extraction of number plates from
vehicle images. After this segmentation is applied to recognize the characters present on
the number plate using template matching. This algorithm can recognize the number plate
quickly and accurately from the vehicle's image. The goal of the research is to investigate
the possibility to create a comprehensive system for Indian vehicle identification based on
the license plate recognition. In that case no additional hardware, such as e.g. transmitters,
mounted on a vehicle, and responders will be required. The system performs well on
various types of vehicle license plate images. The recognition errors of letters and
numbers mainly occur in some of the characters with the very similar main structures but
some detailed differences, such as B and 8, O and 0, S and 5 You have to resize each
character based on standardized size in this method, which becomes an added step in pre-
processing and increases time needed.

2.2 Existing System

The existing system of the License plate Detection includes several steps.Firstly
we are capturing the image through the device, then input the captured image. Next the
preprocessing techniques take place where we are converting the image to the gray scale
to reduce the complexities in the image.Applying the median filters to reduce the noise in
the image.Then, detect the edges in the image by using the sobel edge detector.The Sobel
edge detector uses the kernels or masks,these kernels are convoluted with the image so
we will get the edges in horizontal and vertical directions using respective kernels.After
detecting the edges applying thresholding to the image it removes the lower pixels and
produces the binary image.Then apply the flood fill algorithm to get the region of License
plate and extract the License plate by cropping the region of interest[1].Atlast the

Dept.,of AI&ML(2020 Batch) 6 2022-23

 License Plate Detection using Image Processing
recognition of characters is done by applying the bounding box algorithm to segmentation
of characters and identifying the characters with the help of dataset.

2.2 Proposed System

License Plate Recognition using an efficient OCR engine like Pytesseract along
with major and vast libraries of OpenCV for image processing. As we have seen so far
LPR covers as a solution to most of the problems we have posed. The proposed system
for License plate recognition consists of two steps: license plate detection and character
recognition. In the first stage, image acquisition that is the input of the image which is
captured from highly resoluted camera,so that the image should be clear for the
processing and to get accurate results.In the next stage the preprocessing of the image
starts with the conversion of color image to grayscale as it reduces the complexities in the
image and then applying the bilateral filter to sharpen the edges. In addition to these
techniques, the system also uses canny edge detection to identify the edges of the license
plate characters. This helps in separating the characters from the background and
improving the accuracy of character recognition. Overall, the combination of binarization,
noise removal, and edge detection techniques helps in handling challenges like noise and
low illumination in the pre-processing stage.
The contours are drawn for the image for the localization of the License
plate.These contours are sorted based on their area or the perimeter.After this we will get
some sorted contours and get the license plate region by cropping the rectangle part after
looping through each contour. Then feed it to pytesseract to recognise the numbers and
characters.
The proposed system helps to perform successful and efficient preprocessing on
the raw RGB image and to exploit the high performance and effectiveness of OpenCV
and Pytesseract framework to detect and recognize license plate of vehicles,to improve
our system reliability.

Dept.,of AI&ML(2020 Batch) 7 2022-23

CHAPTER 3
SYSTEM ARCHITECTURE
3.1 Data Flow Diagram

3.2 Methodology
License plate detection is a common task in image processing and computer vision. It
involves identifying and localizing license plates in images or video frames. Here is a
general methodology for license plate detection using image processing:

1) Image Acquisition: The first step is to capture an image of the vehicle and its license
plate. This can be done using a camera installed at a fixed location or a mobile device
such as a smartphone.

2) Image Preprocessing: Start by preprocessing the input image to enhance the quality
and make subsequent processing more effective. Common preprocessing steps include
resizing, noise removal, and contrast enhancement.

Dept.,of AI&ML(2020 Batch) 8 2022-23

 Gray-scale Conversion: Convert the preprocessed image to gray-scale. This simplifies
subsequent processing steps and reduces the computational complexity.

Dept.,of AI&ML(2020 Batch) 9 2022-23

 License Plate Detection using Image Processing
 Noise reduction: Reducing noise in license plate detection can help improve the
accuracy and reliability of the algorithm.Here we are using the Bilateral Filter,it is
useful for noise reduction while preserving edges. Gaussian and Median filters tend to
blur edges,but bilateral filter it can reduce unwanted noise very well while keeping
edges fairly sharp.The bilateral filter helps in license plate detection by reducing noise
and smoothing the image while preserving important edges and details. It can improve
the accuracy of subsequent processing steps by providing a cleaner image for contour
extraction and region selection.
 Edge Detection: Apply an edge detection algorithm, such as the Canny edge detector,
to identify the edges in the gray-scale image. This helps in locating the boundaries of
objects, including license plates. The Canny edge detection algorithm consists of the
following steps:

 Gaussian blur: The first step is to apply a Gaussian blur to the image. This is done
to smooth the image and reduce noise.

 Gradient calculation: The next step is to calculate the gradient of the image. This
is done by calculating the derivatives of the image in the x and y directions.

 Non-maximum suppression: The next step is to perform non-maximum

suppression. This is done by suppressing all pixels that are not local maxima in
their neighborhood.

 Hysteresis thresholding: The final step is to perform hysteresis thresholding. This

is done by setting two thresholds, a low threshold and a high threshold. All pixels
with a gradient magnitude greater than the high threshold are classified as edges.
All pixels with a gradient magnitude between the low threshold and the high
threshold are classified as potential edges. Potential edges are then classified as
edges if they are connected to a high-threshold edge.

3) Contour Extraction: Extract contours from the edge-detected image. Contours are
continuous curves that represent the boundaries of objects in an image. Most image
processing libraries provide functions or methods to find contours, such as OpenCV's
findContours() function. The contour detection algorithm traces the continuous curves
that represent the boundaries of connected components in the binary image.

Dept.,of AI&ML(2020 Batch) 10 2022-23

 License Plate Detection using Image Processing
4) Sorting the Contours: Sort the extracted contours based on certain criteria to identify
potential license plate regions. Based on the area we are sorting the 10 largest
contours.

5) Contour approximation: Contour approximation is an algorithm for reducing the

number of points in a curve with a reduced set of points thus, an approximation. This
algorithm is commonly known as the Ramer-Douglas-Peucker algorithm, or simply
the split-and-merge algorithm.The general assumption of this algorithm is that a curve
can be approximated by a series of short line segments. And we can thus approximate
a given number of these line segments to reduce the number of points it takes to
construct a curve.Overall, the resulting approximated curve consists of a subset of
points that were defined by the original curve.The algorithm is implemented in
OpenCV via the cv2.approxPolyDP function. Once we have the approximated
contour, we check the length to see how many vertices (i.e. points) our approximated
contour has. If our approximated contour has a 4 vertices, we can thus mark it as a
License plate.

6) Region of Interest (ROI) Extraction: Once potential license plate regions are
identified, extract the corresponding regions of interest (ROIs) from the original
image.

7) Character Recognition: Recognition of characters from the image is using the

pytesseract.Python-tesseract is an optical character recognition (OCR) tool for python.
That is, it will recognize and “read” the text embedded in images.
pytesseract.image_to_string() function extracts the string from the image.It uses the
Page segmentation mode for extracting the characters for more accuracy.Page
segmentation mode defines how your text should be treated by Tesseract. For
example, if your image contains a single character or a block of text, you want to
specify the corresponding psm so that you can improve accuracy.

 PSM 11:Sparse Text: Find as Much Text as Possible in No Particular Order

Detecting sparse text can be useful when there is lots of text in an image you need to
extract.When using PSM 11, Tesseract performs the following steps:
 The image is analyzed for script and orientation information to assist in accurate
recognition.
 The text regions are identified and segmented based on the script and orientation
analysis.

Dept.,of AI&ML(2020 Batch) 11 2022-23

 License Plate Detection using Image Processing
 Tesseract performs OCR on each segmented region separately, using the
appropriate language model and OCR settings.

8) Output: Finally, display or output the detected license plate regions, along with any
recognized characters if character recognition was performed.

3.3 Algorithm
The sequence of processes associated with Number plate recognition is
given below. The file upload process is initiated by the recognition of License
Plate Number.
Input: Uploading the image file from camera
Output: Vehicle number plate in characters
1) Read the original image or Capture the image
2) Resize the image
3) Convert it to grayscale.
4) Apply Bilateral Filter.A bilateral filter is a non-linear, edge preserving, and noise-
reducing smoothing filter for images. It replaces the intensity of each pixel with a
weighted average of intensity values from nearby pixels.
5) Identify and store the Canny edges. The Canny edge detector is an edge detection
operator that uses a multi-stage algorithm to detect a wide range of edges in images.
6) Find the contours in from the edges detected and sort the top 10 contours.
7) Get the perimeter of each contour and select those with 4 corners.
8) Mask all other parts of the image and show the final image.
9) Read the text using Tesseract OCR.
On upload of the image file to the system, the number plate recognition system performs
its functions to provide the output.

Dept.,of AI&ML(2020 Batch) 12 2022-23

 License Plate Detection using Image Processing

Dept.,of AI&ML(2020 Batch) 13 2022-23

CHAPTER 4
IMPLEMENTATION

4.1 CODE

import cv2
import imutils
import numpy as np
import pytesseract
pytesseract.pytesseract.tesseract_cmd = r'C:\Program Files\Tesseract-OCR\tesseract.exe'

img = cv2.imread(‘car.jpg',cv2.IMREAD_COLOR)
img = cv2.resize(img, (600,400) )
cv2.imshow('orignal Image',img)
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
cv2.imshow('Grayscale Image',gray)
gray = cv2.bilateralFilter(gray, 13, 15, 15)
cv2.imshow('Filtered Image',gray)
edged = cv2.Canny(gray, 30, 200)
cv2.imshow('Edged',edged) #drawing edges
contours = cv2.findContours(edged.copy(), cv2.RETR_TREE,
cv2.CHAIN_APPROX_SIMPLE)
contours = imutils.grab_contours(contours)
contours = sorted(contours, key = cv2.contourArea, reverse = True)[:10]
screenCnt = None

for c in contours:
peri = cv2.arcLength(c, True)
approx = cv2.approxPolyDP(c, 0.018 * peri, True)
if len(approx) == 4:
screenCnt = approx
break
if screenCnt is None:
detected = 0
print ("No contour detected")
Dept.,of AI&ML(2020 Batch) 14 2022-23
 License Plate Detection using Image Processing
else:
detected = 1

if detected == 1:
cv2.drawContours(img, [screenCnt], -1, (0, 0, 255), 3)

mask = np.zeros(gray.shape,np.uint8)
new_image = cv2.drawContours(mask,[screenCnt],0,255,-1,)
new_image = cv2.bitwise_and(img,img,mask=mask)

(x, y) = np.where(mask == 255)

(topx, topy) = (np.min(x), np.min(y))
(bottomx, bottomy) = (np.max(x), np.max(y))
Cropped = gray[topx:bottomx+1, topy:bottomy+1]

text = pytesseract.image_to_string(Cropped, config='--psm 11')

print("NUMBER PLATE RECOGNITIONS\n")
print("Detected License Plate Number is:",text)
img = cv2.resize(img,(500,300))
Cropped = cv2.resize(Cropped,(400,200))
cv2.imshow('car',img)
cv2.imshow('Cropped',Cropped)

cv2.waitKey(0)
cv2.destroyAllWindows()

4.2 Testing
License plate recognition (LPR) systems can face several challenges that can impact
their accuracy and performance. Here are some common problems faced in license plate
recognition:
 Parameter Sensitivity: Canny edge detection requires tuning several parameters,
such as the threshold values for edge detection and hysteresis, which can be
challenging. Selecting inappropriate parameter values can lead to inaccurate edge
detection or missing important edges.
 Computational Complexity: The Canny edge detection algorithm involves several
computationally intensive steps, such as convolution with Gaussian filters and non-
Dept.,of AI&ML(2020 Batch) 15 2022-23
 License Plate Detection using Image Processing
maximum suppression. These operations can be time-consuming, especially for large
images or real-time applications.
 Sensitivity to Noise: Canny edge detection is sensitive to noise in the input image.
Noise can produce spurious edges or cause the algorithm to miss true edges. Pre-
processing steps, such as noise reduction or smoothing, are often required to improve
the accuracy of edge detection.
 Single Edge Width: The Canny edge detection algorithm produces edges with a fixed
width, determined by the width of the Gaussian filter used in the process. This
limitation means that it may not accurately represent edges with varying widths in the
image.
 Discretization: Contours are typically represented by a series of connected points or
line segments. This discretization process can lead to information loss, resulting in an
approximation of the true edge. The accuracy of the contour representation depends
on the number of points used, but using too many points can be computationally
expensive.
 Object connectivity: Contour-based edge detection algorithms may struggle to
handle objects that are touching or overlapping each other. In such cases, the contours
of adjacent objects can merge, resulting in a single contour that represents both
objects. Resolving object connectivity accurately is a challenging task for contour-
based methods.
 Fragmentation: Conversely, in certain situations, contours may break up into
multiple smaller segments due to intensity variations within an object or region. This
fragmentation can result in incomplete or fragmented edge representations, making it
difficult to accurately identify the boundaries of object.
 Edge-Based Object Detection: If the goal is to detect objects or shapes in an image
solely based on their edges, without the need for contour information, contour
detection may not be necessary. Edge detection methods like Canny can be sufficient
for this purpose, followed by object recognition or analysis based on the detected
edges.
 Texture Analysis: Contour detection primarily focuses on the boundaries of objects
or regions, while texture analysis involves extracting information about the spatial
arrangement of pixels within an object or region. If the goal is to analyze textures in
an image, methods such as texture filters, texture descriptors (e.g., Local Binary

Dept.,of AI&ML(2020 Batch) 16 2022-23

 License Plate Detection using Image Processing
Patterns), or frequency domain analysis (e.g., Fourier Transform) may be more
appropriate than contour detection.
 Shape Matching: If the objective is to compare or match shapes based on their
overall structure or geometric properties, contour detection may not be the primary
technique. Shape matching algorithms, such as Hu moments, scale-invariant feature
transform (SIFT), or shape context, can be used to analyze and compare shapes based
on their distinctive features rather than contours alone.
 Variations in License Plate Designs: License plates can have various designs, fonts,
and styles, which can make it challenging to accurately recognize and interpret the
characters. Differences in character spacing, size, and orientation further complicate
the recognition process. LPR systems need to be robust enough to handle these
variations and adapt to different license plate formats.
 Poor Image Quality: The quality of the captured license plate images can greatly
affect recognition accuracy. Factors such as low lighting conditions, motion blur,
occlusions, reflections, and camera angle can degrade the image quality and make it
difficult to extract the license plate and accurately recognize the characters.
Preprocessing techniques, such as image enhancement and noise reduction, may be
necessary to improve image quality before recognition.
 Environmental Conditions: License plates can be subjected to challenging
environmental conditions, such as rain, snow, dust, or fog, which can further degrade
image quality and impact recognition accuracy. Adverse weather conditions can affect
the visibility of characters and introduce noise and artifacts in the image. Robust
algorithms and image processing techniques are required to handle such
environmental challenges.
To overcome these limitations, various advanced edge detection techniques have
been developed, such as gradient-based methods (e.g., Sobel, Canny), which rely on the
analysis of intensity gradients rather than contours alone. These techniques aim to address
some of the challenges associated with contour-based edge detection and improve edge
detection accuracy.

Dept.,of AI&ML(2020 Batch) 17 2022-23

 License Plate Detection using Image Processing

Dept.,of AI&ML(2020 Batch) 18 2022-23

CHAPTER 5
RESULTS
1)Original Image:

Firstly Input the captured image in RGB format from the device.
2)Grayscale Image:

The RGB image is converted into grayscale image to reduce the complexities in the color
image for further morphological operations

Dept.,of AI&ML(2020 Batch) 19 2022-23

 License Plate Detection using Image Processing
3)Filtered Image:

The Grayscale image is futher processed for noise reduction using the bilateral filter for
sharpening the edges and smoothing of the image.
4)Edged Image:

Edges are detected using the canny edge detection which helps to draw the contours for
the image

Dept.,of AI&ML(2020 Batch) 20 2022-23

 License Plate Detection using Image Processing
5)License Plate Detected using Contours:

The License Plate region is extracted by sorting the contours in the image and a contour is
drawn for the region of License Plate.
6)Extracted image:

The region of License Plate is cropped from the whole image and displayed it.
The characters are recognized and output is displayed as below:
NUMBER PLATE RECOGNITIONS

Detected License Plate Number is: KL 21S 8086

Dept.,of AI&ML(2020 Batch) 21 2022-23

 License Plate Detection using Image Processing

CONCLUSION
Through this project it is possible to recognise Vehicle registration numbers
through digital image processing. The system has been designed using a modular
approach which allows easy upgrading and/or substituting of various sub-modules thus
making it potentially suitable for a large range of vision applications. The performances
of the system makes it a valid choice among its competitors especially in those situations
when the cost of the application has to be maintained at reasonable levels. Furthermore,
the modular architecture makes it extremely flexible and versatile. The earlier
methodologies which have been implemented have not been as accurate and efficient as
the designed Recognition system , this is because of the implementation of digital Image
Processing which gives an accuracy of 90% under normal conditions This Project is
based on automatic vehicle license plate recognition, in which it is observed that the
existing techniques don't pay much attention towards improving the system's efficiency in
terms of its power consumption. As the objective in our proposed design is to reduce
power consumption of the system, with the successful implementation of the same it will
play a very important role in traffic management and security systems such as automobile
theft prevention, parking lot management etc. implementations of the software algorithm
have shown promising results. The system can be made more robust if high precision
cameras can be used to increase overall accuracy if this system is implemented in real
time applications. Also a sensor can be designed to allow the camera to capture the image
only when required to save power.

Dept.,of AI&ML(2020 Batch) 22 2022-23

 License Plate Detection using Image Processing

Dept.,of AI&ML(2020 Batch) 23 2022-23

 License Plate Detection using Image Processing
REFERENCES

[1] Kukreja, Amit, Swati Bhandari, Sayali Bhatkar, Jyoti Chavda, and
Smita Lad. "Indian vehicle number plate detection using image
processing." Int Res J Eng Technol (IRJET) 4, no. 4 (2017).
[2] "Automatic Number Plate Recognition" ,Rai, Vanshika, and Deepali
Kamthania. In Proceedings of the International Conference on
Innovative Computing & Communication (ICICC). 2021
[3] "Automatic vehicle number plate recognition for vehicle parking
management system." Jadhav, Ganesh R., and Kailash J. Karande.
Computer Engineering and Intelligent Systems, ISSN (2014): 2222-
1719.
[4] "Vehicle number plate detection using image processing",Singh,
Abhay, Anand Kumar Gupta, Anmol Singh, Anuj Gupta, and
Sherish Johri. International Research Journal of Engineering and
Technology (IRJET) 5, no. 03 (2018).
[5] Ganta, Srikanth, and Praveen Svsrk. "A novel method for Indian
vehicle registration number plate detection and recognition using
image processing techniques." Procedia Computer Science 167
(2020): 2623-2633.
[6] Soumya, Kumary R., Angel Babu, and Laya Therattil. "License plate
detection and character recognition using contour analysis."
International Journal of Advanced Trends in Computer Science and
Engineering (2014).
[7] Saraswathi, S., Ravi Subban, T. Shanmugasundari, and S. Manogari.
"Research on License Plate Recognition Using Digital Image
Processing." In 2017 IEEE International Conference on
Computational Intelligence and Computing Research (ICCIC), pp. 1-
6. IEEE, 2017.

Dept.,of AI&ML(2020 Batch) 24 2022-23