Clarify refuelling processes faster
Connect vehicle movements, licence plates and open transactions to a traceable event.

AI video analytics automatically detects defined visible events, informs your team and provides the relevant visual context for a quick check.
Initial position and goal
Vehicles, customers, employees, goods and hazardous substances meet in a narrow space on your petrol station area. At the same time, dispensers, shop, access points, washing areas and technical facilities often have to be monitored around the clock.
Vehicles, customers, employees and hazardous substances meet around the clock on your petrol station area. AI video analytics detects visible events, links them with your rules and supports your team in a quick, traceable response.
Connect vehicle movements, licence plates and open transactions to a traceable event.
Detect visible smoke, fire, larger liquid leaks or people in defined danger zones.
Document in and out and out and equal number plate where it is legally permitted with shared reference lists.
Notify staff or the control centre only when a defined situation requires review.
The most important areas of application – Other useful applications
You determine which visible situation should be identified, when it is relevant and which action follows. The AI detects visible events, but neither the intentions nor the legal meaning of a situation.
The video analytics detects when a vehicle passes through a defined exit area. In connection with your cash register, petrol pumping or petrol station management system, it can be checked whether there is still an open refuelling process at this time.
If a corresponding combination is detected, your team automatically receives an event with:
This allows you to check and document a possible case more quickly. The video analytics alone does not determine whether there is a crime or an intentional action. The final evaluation is carried out by your employees.
At your entrances, exits or petrol station areas, vehicle registration plates can be automatically read and assigned to an event.
The licence plate recognition supports you, for example:
An automatic list comparison is only used where there is a sufficient legal basis. A hit serves as a note for your authorised personnel and should be checked before further action.
Visible smoke or flame developments can be detected in the area of your fuel dispensers, your shop, your washing hall, your waste rooms or your technical facilities in a camera area.
In the event of a detected event, your system can be immediately:
The video analytics supplements your prescribed fire alarm and security systems with a visual detection layer. It does not replace fire detectors, emergency shutdowns or technical safety systems.
Larger liquid accumulations and clearly visible in the camera view can be detected as an inspection event.
This can be useful, for example, in the following areas:
Whether a liquid is dangerous or which substance it is cannot be reliably determined by means of the camera view. The reporting therefore serves the quick visual inspection by your personnel and supplements existing leakage and hazardous-material sensors.
You can place virtual zones via the camera view and connect it to individual rules. The analysis detects, for example, when a person enters a technical area, a tanker zone or a closed working area.
The rules can be defined depending on the following factors:
This only reports those situations that need to be checked in accordance with your operational specifications.
The video analytics can detect visible events in your publicly accessible operating areas. These include, for example:
A detected event is sent to your staff for review. The AI does not infer intent or assess how dangerous a person may be.
On the petrol station area, people move between entering, parking and ranking vehicles. Video analytics can detect critical close approaches in previously defined areas.
This is relevant, for example:
Recognised situations can be documented and evaluated. This gives you a better foundation to identify confusing areas and optimise your traffic guidance.
The AI detects when people enter protected areas outside defined times or without intended operation.
These may include:
In the event of a detected access, your security service, your control centre or another responsible team may be informed.
Longer stays or defined movement sequences at fuel dispensers, machines, washing systems, advertising systems or technical equipment can trigger an automatic test.
The video analytics cannot determine vandalism in every situation immediately and beyond any doubt. However, it can identify visible actions or changes in condition and make your team aware of a situation that needs to be reviewed.
Defined items or displays can be monitored for visible distance or change.
This may affect, for example, the following areas:
A message is triggered when a monitored object is removed or leaves its specified area. Your staff will check whether a withdrawal is entitled.
If the video analytics detects a fallen person or motionless for a specified period, your staff may be notified automatically.
This supports you in particular:
The detection serves as a supplementary level of security and does not replace any organisational first aid and emergency measures.
Newly switched off items or unsupervised items over a defined period of time can be automatically reported as an event.
Your team will receive the camera location, the time and a suitable image or video clip. This allows the situation to be checked more quickly without having to observe all camera images permanently.
The video analytics detects when a defined door or gate remains visibly open or is used outside the set time window.
This is suitable for:
If a specified opening time is exceeded, a message can be triggered to your responsible team automatically.
Defined paths and areas can be monitored for visible obstacles.
These include, for example:
If the system detects a vehicle, delivery or other item in a free zone, your team will be informed.
The video analytics can count vehicles and record the utilization of defined areas. This allows you to detect, for example:
The evaluations help you to better adapt traffic management, personnel deployment and operational processes to actual use.
During goods deliveries and tank fillings, vehicles, employees and external service providers meet in sensitive areas.
The video analytics can be detected, among other things, if:
The specific rules are adapted to your workflows and safety regulations.
After a reported event, you can search extensive video recordings based on defined criteria.
Possible search criteria are:
This allows you to connect relevant video footage more quickly and do not need to manually search all recordings.
How it works
Your existing or new cameras capture the areas defined for the respective application.
The AI evaluates the required visual features, objects, movements or changes in their state.
Zones, time windows, dwell times, vehicle movements and other rules determine when a situation is classified as a relevant event.
In the event of a detected event, your employees, your control centre or connected systems are specifically informed.
Your team checks the situation and processes it according to a defined operational workflow.
The event can be documented in a traceable way with time, camera location, image, video clip and processing status.
Operational integration
Recognised events can be transmitted to your existing security, cash register and operating systems via interfaces. This results in a consistent process from visual recognition to operational testing to documented processing. Possible integrations are:
Operational added value
faster checking of possible departures with open tanks
automatic occupancy of licence plates and vehicle images
Previous detection of visible security risks
faster reaction to smoke, fire or larger liquid trips
better monitoring of dispensers and outdoor areas
More targeted alerting of your employees
Relief for low or unoccupied operating times
better protection for shop, storage and technical areas
Structured documentation of relevant events
faster searches for incidents
Use your existing camera infrastructure
Central management of several filling stations
cross-site event and utilization evaluations
Data protection and compliance
The implementation is based on the principle Privacy by Design. Only those areas and visual features required for your specified purpose are analysed.
In particular, label recognition, list matching and biometric procedures require a clear purpose limitation, a suitable legal basis and individually defined access and deletion concepts.
The reliability of video analytics depends on the quality and position of your cameras as well as the conditions on site.
Impairments may arise, among other things, by:
A camera view cannot reliably determine whether a person has paid, what intention he pursues or whether an action is punishable. A possible open refuelling process is therefore only recognisable by connecting to your cash register or petrol station management system.
Small flames, cigarettes, fuel vapours or minor liquid leaks are not reliably visible in every condition. Video analytics does not replace prescribed fire, gas, leak or emergency shutdown systems.
Automatically generated reports should be checked by authorised personnel before further action are initiated.
Privacy by Design
Depending on the application, the following measures are possible:
For locations in the EU form the GDPR, especially Art. 5, 6, 25 and 35 and the EDSA video surveillance guidelines the basis of the privacy audit. In addition, consider the law of the respective Member State, in particular employee data protection and co-determination. The duties from the EU AI Act check for the concrete use of AI.
Frequently Asked Questions
In many cases yes. The decisive factors are the viewing angle, lighting, frame rate and coverage of the required area. Specially positioned cameras may be required for licence plate recognition.
Not from the video image alone. The system can detect a vehicle leaving and link the event to an open refuelling transaction in the till or fuel-dispenser system. Staff must verify whether fuel theft has occurred.
Yes, when camera position, image quality, lighting and vehicle speed are suitable. Processing must have a defined purpose and comply with data protection law.
A list comparison is technically possible. However, it is only used where it is legally permissible and you have a sufficient legal basis, an authorisation concept and defined deletion periods. A hit should always be checked manually.
Clearly visible smoke and flame developments can be detected. The reliability depends on the camera view, distance, lighting and surroundings. The analysis does not replace your prescribed fire safety systems.
Larger, high-contrast and visible fluid accumulations in the camera view can trigger an event. Small leaks, colourless liquids and vapours cannot be reliably detected using cameras alone. Technical sensors are required.
The AI can report defined visible events such as removing a supervised item or entering a protected area. It cannot reliably determine whether an action is unauthorised or unlawful.
Facial recognition is technically possible, but is only used for clearly defined use cases in which it is legally permissible and proportionate. This requires a separate legal examination as well as strict access, storage and deletion rules.
Yes, provided your cameras provide sufficient images. Lighting, headlights, reflections and contaminated lenses can affect the detection. Optimised night or special cameras can be useful depending on the area.
Yes. Events, cameras, rules and evaluations of multiple sites can be centrally combined depending on your infrastructure and authorisation concept.
No. It helps your team detect, review and document visible events. Prescribed fire detectors, leak sensors, emergency shutdowns, alarm systems and operational safety processes remain necessary.
Your specific application
We check which applications can be reliably implemented with your existing camera infrastructure and develop a suitable analysis, integration and operational architecture for your location or your entire network of filling stations.