Transparency in ANPR Data Collection: Upholding Privacy and Building Trust

Transparency in ANPR Data Collection: Upholding Privacy and Building Trust

In our pursuit of improving Automatic Number Plate Recognition (ANPR) models, transparency stands as a cornerstone of our operations. We are committed to upholding privacy and building trust through open communication about our data collection practices. This article outlines key aspects of our approach, including our lawful basis, types of data collected, processing methods, retention policies, security measures, data sharing practices, and respect for individual rights.

Purpose of Data Collection

Our primary goal is to enhance our ANPR and object detection AI models, which are used to identify vehicles appearing on our customers' driveways. To ensure accurate results across various conditions, we need to collect a large number of images that account for the diverse external factors affecting image quality and settings.

Lawful Basis

Our data processing is grounded in:

  1. Public interest
  2. Fulfilment of tasks carried out in the public domain
  3. Legitimate interest in providing effective and precise ANPR services to our customers

Types of Data Collected

  • Randomly collected images of vehicles on public roads, including:
    • Vehicles themselves
    • License plates
  • No additional personally identifiable information.
  • Customers participating in training contributions for this use.

We focus on training our ANPR model to recognize consistent patterns across images while excluding extraneous background details. We commit to updating this policy if any additional data collection becomes necessary.

Data Processing

  1. Secure transfer of collected images to our local scOS server
  2. Hygiene process: Deletion of unsuitable images (e.g., blurry, vehicles not clearly visible)
  3. Manual annotation by scOS team members:
    • Drawing bounding boxes around vehicles and license plates.
  4. AI model training:

Data Retention

We retain collected images for as long as our model training events require it. However, you can request us to erase data that matches to your vehicle(s) using the contact details below. This approach balances our need for continuous model improvement with responsible data management practices.

Data Security

We prioritize data security through:

  • Encrypted storage on our local server
  • Secure transmission to AWS (when required)
  • Stringent confidentiality and security protocols during manual annotation

Data Sharing

This data is not shared and is confidential to scOS.

Individual Rights

We respect GDPR rights, including:

  • Right to access personal data
  • Right to request rectification of inaccuracies
  • Right to request data deletion

To exercise these rights or for any inquiries, contact:

We commit to prompt responses and appropriate actions in accordance with applicable data protection laws.

Legal Basis and Policy Changes

This policy complies with relevant data protection laws, including GDPR. We regularly review our practices to ensure ongoing compliance. Any modifications will be communicated through our website or via direct notifications, as required by law.

Further Information

For more readable policies and articles about data, please visit our Data Ethics Zone.

Your trust is paramount to us. By maintaining transparency in our ANPR data collection practices, we aim to foster a relationship of trust with all stakeholders while delivering innovative and effective services.