Scribd
Upload
31 views7 pages

Vanity Plate Identification

The paper proposes a unified deep neural network that can detect license plates and recognize characters in one pass. This approach jointly solves detection and recognition with a single network, avoiding separate steps and accumulated errors, while accelerating processing speed. Experiments on various scene images demonstrate the effectiveness of the proposed method.

Uploaded by

swathi8903422971
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
31 views7 pages

Vanity Plate Identification

The paper proposes a unified deep neural network that can detect license plates and recognize characters in one pass. This approach jointly solves detection and recognition with a single network, avoiding separate steps and accumulated errors, while accelerating processing speed. Experiments on various scene images demonstrate the effectiveness of the proposed method.

Uploaded by

swathi8903422971
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 7

VANITY PLATE IDENTIFICATION USING

CONVOLUTION NEURAL NETWORK


Paper presentation for Conference

Submitted by
Megala K 210420205034
Swathisri R 210420205062
In partial fulfillment for the award of the degree of

BACHELOR OF TECHNOLOGY
IN
INFORMATION TECHNOLOGY
CHENNAI INSTITUTE OF TECHNOLOGY, CHENNAI

ANNA UNIVERSITY : CHENNAI 600 025

INTRODUCTION
- The main scope of this project is to implement intelligent urban surveillance
system for automated Number Plate Recognition.

- The automated object detection algorithm is really important component in the


smart city application.

- In urban surveillance application, the image sensor or camera plays an


important role in digitizing the scene or environment.

- To process the digitized images, searching for a particular object, smart


vehicle vanity plate is a huge task as it will need high cpu and memory power.

- To achieve this kind of functionality with distributing the processing is the best
way to solve.

- Image processing technology to search for a number plate in a given image is


a vital task.

- AUTOMATIC car license plate detection and recognition plays an important


role in intelligent transportation system.

- It has a variety of potential applications ranging from security to security to


traffic control and attracts considerable research attentions during recent
years.

- To our knowledge, this is the first work that integrates both license plate
detection and recignition into a single end to end trainable network and solves
them at the same time.

ABSTRACT
In this paper, we tackle the problem of car license plate detection and recognition in
natural scene images. We propose a unified deep neural network, which can localize
license plate and recognise the letters simultaneously in a single forward pass.
The whole network can be trained end to end. In contrast to existing approaches
Which take license plate detection and recognition as two separate tasks and settle
them step by step, our method jointly solves these two tasks by a single network. It
not only avoids intermediate error accumulation but also accelerates the processing
speed. For performance evaluation, four data sets including images captured from
various scenes under different conditions are tested. Extensive experiments show
the effectiveness and efficiency of our proposed approach.

It access the image locally and convolution neural network (CNN) is a technique
which is used in detecting number plate region. Recently, convolution neural
network(CNN) based detectors achieve the state of the art performance. However, it
is difficult to efficiently detect the vehicle and license plate simultaneously in most
cases.
PROBLEM STATEMENT

- In existing approaches which take license plate detection and recognition as

two separate tasks and settle them step by step. This method solves these

two task by a single network.

- It not only avoids intermediate error accumulation but also accelerates the

processing speed. For performance evaluation four data sets including

images captured from various scenes under different conditions are tested.

extensive experiments show the effectiveness and the efficiency of proposed

approach.

- The main focus in this research project is to experiment deeply with, and find

alternative solution to the image segmentation and character recognition

problems within the license plate recognition framework.

MODULES

1. Image Capturing

2. Image Frame cleaning and preprocessing

3. Detecting number plate using contours

4. Transforming Number plate image to text.

IMAGE CAPTURING
- It is a basic infrastructure for establishing the proposed picture of image

handling in Python.

- It prepares the python environment to access input device, in this case

camera is used.

IMAGE FRAME CLEANING AND PREPROCESSING


- Once the image is read in Python it uses the different image processing

technique on the image.

- The pixel will have the value of one byte per pixel which means the value

between 0 to 255 is used.

DETECTING NUMBER PLATE USING CONTOUR


- After edge detection, all high intensity pictures were scanned from left to right

in x-axis and top to bottom for y-axis.

- Concentration of white pixels will give a fair knowledge of where the more

number of edges where available in x-axis and also scan to get approximate

location of y-axis.

TRANSFORMING NUMBER PLATE IMAGE TO TEXT


- The region of interest for the number plate area was detected from the

previous module. We will try to fit the number plate area and ratio of width and

height.

- If the possible number plate region is identified the image is cropped and then

processed with the simple optical character recognition software to convert to

a text.

RESULT
CONCLUSION
The paper reviews intelligent surveillance video analysis techniques. Reviewed
paper cover wide variety of applications. The techniques, tools and dataset identified
were listed in form of tables. The issues identified in existing methods were listed as
future directions to provide efficient solution.

REFERENCES

1. J. Redmon and A. Farhadi , “YOLO9000 : Better, faster, stronger,” in

Proc. IEEE Conf. Comput. Vis. Pattern. Recognit., Jul.2017.

2. J. Tian, G. Wang, J. Liu, and Y. Xia, “License plate detection in an open

environment by density – based boundary clustering,” J. Electron. Imag.,

vol.26, no. 3, 2017.

3. S. Kim, H. Jeon and H. Koo, “Deep-learning- based license plate detection

method using vehicle region extraction,” Electron. Lett., vol. 53, no. 15, 2017.

4. L. Neumann and J. Matas, “A method for text localization and recognition in

real-world images,” in Proc. Asian Conf. Comput. Vis., 2011.

You might also like