Services
We develop technologies that merges the gap between technological advancements and user convenience
Image Upload
Upload a Image to identify and recognize the License_Plate and Vehicles Information.
Video Upload
Upload a Video to identify and recognize the License_Plate and Vehicles Information.
Live Data
Live Identification and Detection From Traffic Camera.
Features
Excited to know what we do.. Here it is
Is your lisence plate authentic.. Be aware
Technology we built detects
- Fake lisence plates
- Black Listed Registrations
- Stolen vehicles
Rules are to be followed and we make sure of it
For your safety. We can detect
- over speeding
- Helmet violations
- Traffic rules violation
Get the Violation ticket in your mail and sms, So you Won't get Sunburns
Technology can make everything convenient.. Even Penalty
Accident Detection in Real time, So that the Victim won't be let down
Hit and Run is not a problem anymore
Approach
This is what we do!!!
DataSet
We Create a customized and self labelled dataset which is trained using yolo
Gaussian Blur
Then we convert the input image to grayscale and apply a small Gaussian blur to smooth it out
Otsu Method
The image is then thresholded to white text with black background and has Otsu's method also applied. This white text on black background helps to find contours of image.
Open CV & Contuor
We dilate the image using OpenCV and apply contour, further we sort the contour and filter them and remove the unwanted contour
Tesseract-5 OCR
To obtain the number and the letter in the number plate ,we segment the contour and using Tesseract-5 OCR ,we get the value of each contour
Lisence Plate Detection
The number from the license plate is used to evaluate its other data elements like type, make, model, color using the dataset, then we look up in the provided dataset for any blacklisted registration numbers
Technology Stack
We Use cutting edge technologies which makes our product more reliable and convenient
AI
Python
TensorFlow 2 API
Open CV
Tesseract
Yolo v3 - Custom trained model
BACK-END
Python
Flask
MongoDB
FRONT-END
HTML
CSS
JavaScript
Bootstrap
