Success Story

Developing ML-based Automated License Plate Recognition (ALPR) software for an American surveillance technology provider.

About the Client

titleImage

The client is an American technology provider for the surveillance needs of law enforcement agencies. Beat units, detectives, and command staff can leverage the provider’s surveillance license plate recognition solutions to detect and prevent crime in their community. This technology has helped an average patrol officer to recognize over 30,000 license plates every month using a mobile reader ensuring contactless compliance.

  • 80,000+ vehicles scanned
  • 98.5% accuracy
  • 10x faster scanning & reporting

Business Situation

Existing technologies in the marketplace for Automated License Plate Recognition (ALPR) have been known to have several limitations. They incur heavy costs, are ineffective with high-speed vehicles, lead to data loss and hardware compatibility issues, which make them unfit for use in a high stakes domain such as law enforcement.

The American surveillance technology provider was looking to solve these issues by leveraging Machine Learning (ML) to develop an ALPR application. The end goal was to help law enforcement officers use this technology to close cold cases that had been bogged down by existing license plate recognition technologies that were too slow in terms of on-the-go capabilities, historical data analysis and accurate matching.

They were on the look around for a technology partner with extensive experience in AI/ML technologies to step in and develop the necessary ML models and the supporting web and mobile applications for the intended purpose. After scouring the Asia Pacific market for AI experts who could provide time-bound, cost-effective services, they became privy to Daffodil Software’s AI Center of Excellence (CoE).

Daffodil was chosen as their technology partner and proceeded to gather the American provider’s requirements and formed a developmental roadmap for the ML solution to comprise of the following capabilities:

  • Develop trained models for instantaneous number plate scanning within milliseconds
  • Should retrieve state of origin, car make and model instantly
  • Perform training and retraining of models for incrementally faster predictions
  • Equip the application to carry out most of the data processing on the client-side
  • Calibrate for improper image orientation, noise and other hurdles in capturing images
  • Update models frequently to enable compatibility alongside equipment upgrades
  • Keep the system abreast of latest technology changes in the AI/ML ecosphere

The Solution

The Daffodil team set about developing the ML-based web and mobile application that could eventually scan the license plates of fast-moving vehicles and retrieve valuable information about the vehicles and its owner. This was supposed to provide insights to law enforcement personnel to enable them to solve crimes more accurately using their existing mobile devices and stationary cameras. The following capabilities were then worked into the mobile and web-based application built on an ML foundation:

Custom ML Model

The team leveraged the TensorFlow platform to create and train the ML model with custom training loops. In the ALPR pipeline, automobiles are first located in the frame using an object detection deep learning model, the license plate is then localized using a license plate detection model, and finally, the characters on the license plate are recognized. Eighty percent of the data is used for the training set and twenty percent is used for the validation set. The dataset is then randomly split into the necessary number of records using a series of instructions.

Instantaneous Data Retrieval

When a wanted plate is seen, the automatic system captures, compares, and analyses vehicle license plates and notifies the closest law enforcement agency or sheriff’s department within seconds. The system checks to make sure the information is accurate, the license plate is from the same state as the one where the vehicle is wanted, the reason this car or its owner was wanted or of interest, and if the case is still relevant. The searches can be made on the model, make and color with instantaneous data retrieval.

Cost-Effective ML Validation

Client-side validation was carried out in TensorFlow wherein the application developed by the team used ML directly on the client side. Instead of waiting for API results about the vehicle, client-side computing achieves analysis of live feed and the ML model inference in real time. There is also a reduction in costs since you no longer need to maintain ML servers for heavy computations.

Incrementally Improving Performance

The YOLO v2 real-time object detection convolutional model has been the basis for the creation of the custom law enforcement ALPR model created by team Daffodil. It enables the model to detect small objects in large groups, accounting for localization errors, with a high-resolution classifier. With a single forward propagation through a neural network, the prediction in the entire image is done in a single algorithm run, which helps reduce the computation time.

The Impact

With the AI technological expertise of Daffodil, the American technology provider was able to enable law enforcement agencies and police departments to leverage their existing infrastructure for proactive policing. They can now implement faster follow-ups with historical leads on the National Crime Information Center (NCIC) and clear up decades of backlogs related to cold cases using the ML-enabled ALPR technology. Daffodil received commendations for producing expedited results that went above and beyond the provider’s initial expectations.

  • 80,000+ vehicles scanned
  • 98.5% accuracy
  • 10x faster scanning & reporting

Read Related Case Studies

Get in Touch

Sign up for a 30 min no-obligation strategic session with us

Let us understand your business objectives, set up initial milestones, and plan your software project.

At the end of this 30 min session, walk out with:
  • Validation of your project idea/ scope of your project
  • Actionable insights on which technology would suit your requirements
  • Industry specific best practices that can be applied to your project
  • Implementation and engagement plan of action
  • Ballpark estimate and time-frame for development