Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 14

AUTOMATIC NUMBER PLATE

RECOGNITION SYSTEM.

SIGNALS AND SYSTEM PROJECT


REVIEW 3

BY:-
1. SHABD PRAKASH (18BIS0031)
2. AMAN RAJ GUPTA(18BEC321)
3.
OBJECTIVE

The main objective of our project is to recognize and display the


Registration numbers of vehicles which are scanned by the camera.
EXPERIMENTAL PROCEDURE
The work is divided into several parts:
 Image Reading
 Gray Code Conversion
 Filtering
 License Plate Detection
 Segmentation
 Character Recognition
 Display Output
EXPERIMENTAL PROCEDURE
• Image Reading:
We input the raw image which is actually an RGB image (analog input) which is
considered for conversion.
• Gray Code Conversion:
This RGB image is converted to Gray Scale code which is again binarized to 8-
bit Gray code.
• Filtering:
After conversion the image is filtered using paper and salt noise reduction
which consists of eroding, filling and dilating. The corresponding 3*3 masks
get eight neighboring values.
• License Plate Detection:
This step is the most important step in the detection of this system and
algorithm which is based on the belief that license plates are rectangular in
EXPERIMENTAL PROCEDURE
• Segmentation:
This step is again the next most important step of where the license plate is segmented
using the character detection technique which is mainly given by their vertical projection.
• Character Recognition:
This step we compare different characters which are usually combinations of Numbers
and Alphabets of varying length. These characters are compared using the database of
templates given by the use of unique indexes. Which are numbered so as to compare the
no. of end points.
• Display Output:
Then each output is displayed and each step is shown which renders and reads and then
finally displays the characters.
BLOCK DIAGRAM
BLOCK DIAGRAM
CODES
CODES
CODES
RESULT
CONTRIBUTIONS

• NITISH DUA-Topic Selection, Plate Detection code, Research on various Research


Papers, Report Writing, Review-1 and Review 3 preparation.
• SHABD PRAKASH- Definition of Problem statement, Letter Detection code,
Explanation of the code, Review-1 and Review 2 preparation.
• ROHAN BERA- Letter Detection code, Template Detection code, Report Writing,
Review-2 preparation.
• SAROJ ANAND TRIPATHY- Plate Detection code, Template Detection code, Report
Writing, Review-1 and Review-3 preparation.
• We all did the simulation together
CONCLUSION

The results are as expected but not to up to the mark. The time of compilation and
processing is varying depending on the size of image and the resize function.
Around 12 images have been tested by us at different angles but even if we
increase the range of resize function it is detecting only of the characters. We were
able to find out that bigger images which have other vehicles or writings in
background could not be processed correctly. We weren’t able to do extrapolation
and interpolation. More changes can be executed done on this model. A GUI based
software can be developed using more complex code and the existing algorithm.
REFERENCES

• Qadri, Muhammad Tahir, and Muhammad Asif. "Automatic number plate recognition
system for vehicle identification using optical character recognition." Education
Technology and Computer, 2009. ICETC'09. International Conference on. IEEE, 2009.
• Kurdi, Moustafa M., et al. "Lebanese automated number plate reading based on neural
network recognition." Computer Engineering Conference (ICENCO), 2017 13th
International. IEEE, 2017.
• Panahi, Rahim, and Iman Gholampour. "Accurate detection and recognition of dirty
vehicle plate numbers for high-speed applications." IEEE Transactions on intelligent
transportation systems 18.4 (2017): 767-779.

You might also like