Customer is a huge IT Park handling logistics of thousands of vehicles
- During the peak business hours of the IT park there is a tremendous challenge to have each vehicle verified before entry apart from the RFID Tag.
- Customer was looking at an Automatic solution that will scan vehicle number plates as they come in and go out and authenticate the barrier.
- Implemented License plate number recognition , used Tensorflow in Python on top of IBM Power AI Stack
- Used training data from MNIST database
- The Neural network is training using MNIST data in 30,000 iterations on IBM Power AI server
- The recognition accuracy of the Model is 95%
Benefits to Client
- Implementing the automated system helped the IT Park in reducing the process of noting down the vehicle numbers while entering and exiting
- Solved the possible security threat of un-authorized vehicles entering the premises
- Creates a repository of vehicle log data which helps in necessary reports and analytics.