1 of 10

Problem�Statement - 2

Detect license plate from CCTV:

The objective is to enhance CCTV footage for detecting license plate.

Detect human presence from live stream from CCTV:

The objective is to process live stream from CCTV to detect presence of humans on irregular hours.

2 of 10

OUR TEAM

SHUBHAM SINGH RATHOUR

DIVESH MANDHYAN

ANKITA DATTA

(Team Lead)

3 of 10

IDEA

We propose an efﬁcient real-time automatic license plate recognition (ALPR) framework, particularly designed to work on CCTV video footage obtained from cameras that are not dedicated to the use in ALPR. At present, license plate detection, tracking, and recognition are reasonably well-tackled problems. However, the existing ALPR algorithms are based on the assumption that the input video will be obtained via a dedicated, high-resolution, high-speed camera and is/or supported by a controlled capture environment, with appropriate camera height, focus, exposure/shutter speed, and lighting settings.

This idea presents an efﬁcient and robust framework that can perform localization, tracking and recognition of multiple vehicle license plates in a real-time scenario.

We have also proposed the Human Detection and Counting System from live stream CCTV.

4 of 10

APPROACH

It is intelligent enough to automatically adjust for varying camera distances and diverse lighting conditions, a requirement for a video forensic tool that may operate on videos obtained by a diverse set of unspeciﬁed, distributed CCTV cameras.

5 of 10

“Deep learning is hard, we know, so we have simplified it”.

“

6 of 10

TECH STACK

PYTHON

NUMPY

OPENCV

API

LBPH ALGORITHM

TENSORFLOW

FLASK

KERAS

CNN ALGORITHM

9 of 10

RESULTS

In this, an automatic license plate recognition method, which works in challenging illumination conditions in real-time scenarios. Experimental results shows that the detection rate of the proposed method is much higher than existing methods, with an overall detection rate of 96.72% and a recognition rate of 98.02% in multiple LPs and varying illumination scenarios. The proposed algorithm might be suitable for real-time ITS applications.