Pilot image recognition advertisements provided façade image

Municipality of The Hague

The Hague Pandbrigade wants to arrive at a modern digital working method for Careful Facade Image. To take a step in that direction, the HPB is doing this learning project with image recognition of advertisements.

Last change on 17th of January 2025, at 9:48 (CET) | Publication Standard 1.0

• Publication category: Other algorithms
• Impact assessment: Field not filled in.
• Status: In development

General information

Theme

Space and Infrastructure

Begin date

2023-09

Contact information

[e-mail]

Responsible use

Goal and impact

The Hague Pledge Brigade checks built advertisements. Before this pilot, these were always carried out manually by inspectors who go out on the streets to check whether all advertisements comply with permit requirements. Image recognition has the potential to be a tool with high return at small investment, which can be targeted to areas.

The purpose of the pilot is to explore the potential of the tool. The aim of this pilot is not to end violations, and therefore entrepreneurs will not experience any impact from the pilot.

Considerations

The municipality has the task of enforcing the licence requirement for built advertisements. Although advertising in principle does not contain personal data, business owners may still incorporate personal information in their advertisements, such as photos of themselves, names or personal phone numbers. The image recognition technology does not analyse this information specifically or purposefully, but looks at all images in a general sense, including advertising images containing such personal information.

Human intervention

The results from the image recognition algorithm are assessed by inspectors from the Hague Pandbrigade. The outcome of this assessment is used to gather information

Risk management

The system is not risky in terms of privacy. It does not record personal data and only takes pictures of displays in public spaces. The biggest risk here is that the location of the advertisement becomes known, which could possibly make it possible to trace the address of a person or company. This is also possible on the street. Unauthorised persons cannot access the stored images due to the realised information security. Security risks are controlled by the overall security set-up on systems and connections. A privacy quick scan has been performed.

Legal basis

Municipalities Act; Environment Act (art. 5(1)(a) and (2)(a))

Operations

Data

• id
• bagid
• position
• width
• height
• area
• type
• date_first_detection
• date_latest_detection
• date_removal_detected
• change_detection_sug
• StreetSmartURL

Permits:

• submission date
• status
• description
• address

Technical design

It involves a self-learning AI system for image recognition. This system recognises from streetsmart images with convolutional neural networks. The found advertisements are linked to the licences based on address or bag-id.

External provider

Cyclomedia