Facial anonymization by artificial intelligence executed entirely at the edge. Real-time privacy protection without compromising the evidentiary quality of enforcement records.
Blur Inteligente is Areatec's proprietary facial anonymization system, developed to ensure compliance with data protection legislation without compromising the evidentiary integrity of traffic enforcement images.
The system uses convolutional neural networks trained on datasets representative of the Brazilian demographic to detect and anonymize faces in real time. All processing occurs at the edge, directly on the embedded devices of Olho Vivo patrol vehicles, eliminating the need to transmit biometric data to external servers.
The selective blur architecture allows license plates, signage, and vehicle characteristics to remain intact while faces of pedestrians, passengers, and bystanders are irreversibly anonymized before any storage or transmission.
Real-time video stream from embedded cameras
Neural network identifies all faces in the frame
Anonymization applied only to facial regions
Frame saved with anonymized faces and preserved license plates
Blur Inteligente implements the concept of Privacy by Design as an architectural requirement, not as an optional feature. Anonymization is an integral part of the capture pipeline, ensuring that no frame containing identifiable faces is persisted at any point in the system.
This approach directly meets the principles of necessity and adequacy provided for in Article 6 of the LGPD and the concept of Data Protection by Design and by Default in Article 25 of the GDPR.
Limitation of processing to the minimum necessary. Faces are not necessary data for traffic enforcement.
Technical measures capable of protecting personal data. Irreversible anonymization at the edge as a primary measure.
Data protection integrated into the system design. Architecture that makes the storage of facial data impossible.
Local execution on the device, without cloud dependency. Zero latency in anonymization with dedicated embedded processing.
Simultaneous identification of multiple faces in complex scenes. Pedestrians, passengers, and bystanders processed in parallel.
Operation in low-light and infrared conditions. Adaptive algorithms for patrol vehicle IR cameras.
Full preservation of license plates and vehicle details while faces are anonymized. Intact evidence, guaranteed privacy.
Complete record of each anonymization event for regulatory compliance. Immutable logs with timestamp and metadata.
No biometric data is stored. Faces are anonymized before any recording or transmission of the frame.
Processing speed at 1080p
Facial detection rate
False positive rate
Latency per frame
Supported native resolution
Continuous operation without degradation
Blur Inteligente operates as a native module of the Olho Vivo Patrol vehicle processing pipeline. Anonymization is executed on the same embedded hardware that performs OCR license plate recognition, without the need for additional equipment.
Anonymization events are recorded and made available on the Aretron AI dashboard, allowing real-time monitoring of the compliance status of the entire fleet. Audit reports can be generated automatically for presentation to regulatory bodies.
Patrol vehicle with embedded cameras
Real-time facial anonymization at the edge
Centralized management and monitoring platform
Anonymized frames with audit trail
The system applies selective anonymization that protects the identity of individuals while keeping all elements relevant to traffic enforcement intact.
No. The system operates on the principle of data minimization. Anonymization occurs in the processing pipeline before any recording or transmission. No facial template, embedding, or biometric data is persisted at any stage of the process.
Blur Inteligente implements Privacy by Design in accordance with Article 6 of the LGPD and Article 25 of the GDPR. Anonymization is irreversible, applied at the edge before storage, with a complete audit trail. There is no collection of sensitive personal data at any point in the flow.
No. Selective blur acts exclusively on facial regions detected by the neural network. License plates, vehicle characteristics, signage, and road context remain intact and in original resolution for evidentiary purposes.
Zero measurable impact. Blur Inteligente operates in the same processing pipeline as the OCR, with latency under 5ms per frame. The parallel processing architecture ensures that anonymization does not compete with license plate recognition resources.
Yes. The facial detection model was trained with datasets that include night, infrared, backlight, and low-light scenarios. The 99.2% detection rate covers 24-hour operation in real urban patrol conditions.
Each anonymization event generates an immutable record containing a timestamp, device identifier, detected face count, and processing status. These logs are accessible via the Aretron AI dashboard and can be exported for regulatory audits.
Implement AI facial anonymization in your enforcement fleet. Edge processing, zero storage of biometric data, provable regulatory compliance.