FAQ

Well, you probably expected to hear this: “It depends!” A business innovation of Meleagros is that supports various pricing models depending on the set of products and services of interest. This stems from the modularity of the platform and its ability to integrate with heterogeneous equipment and platforms.

Many factors affect the cost of the system:

  • Whether it is considered for OEM use 
  • Whether there is equipment (processing units, cameras) that will be assessed for reusability
  • What the size and accessibility of the monitored forest area is (for MTower)
  • The quantity of components required (measured in terms of processing units and cameras deployed)
  • Whether there should be integration with other external systems (e.g., C&C, civil protection services, etc.)

Our representative gets in contact with you in order to define the area to be monitored. Several things must be considered, such as size and morphology of the area, availability of energy sources, existing equipment and/or facilities that should be re-used etc. If it is necessary we can visit your area and conduct on-site measurements. However, in many cases blueprints or annotated maps will suffice.

In principle and in most cases yes. In practice, it depends on your equipment characteristics and the area to be monitored. Meleagros video detection software is made so that it can work with any type of camera having some minimum characteristics: can provide IP stream, VGA resolution, 15 FPS.

Another point to consider, when re-using already installed cameras is whether they can cover the required area: fire monitoring usually has different coverage requirements than surveillance or security applications.

In any case, you should keep in mind that although we work with some camera vendors, we always help our clients to minimize costs through reusing equipment so that the system's Total Cost of Ownership (TCO) is reduced and the Return On Investment (ROI) is increased.

Not at present. In fact, there are not well established standards in the video-based fire detection domain yet. Some organizations (UL Listings and FM Approvals) provide some certification processes, however, they mostly refer to the hardware and system-level characteristics and not to the detection method per se. You may read this article for more information on this.

Our present approach is to use certified hardware for the installations where possible.

Such technology is mature and in most cases mandatory by laws and regulations. So you cannot and should not avoid it. However, in many cases it has been proved ineffective due to its “operation principles” (i.e., even if product quality is perfect and guaranteed). Meleagros can monitor large and open areas and also provide visual verification of the incident. We should not forget that early-warning is not always the sole objective of a safety and security system: early assessment of the situation is equally important.

We are open to discuss the installation of a small-scale pilot system in two cases:

  • It concerns an application area or environment with particular research interest
  • There are prospects for more extensive deployment, if the pilot is successful 

A pilot installation is based on the following main terms and conditions:

  • It has a specified time duration
  • The client will allow and help our team to perform tests
  • The client covers the inelastic expenses (e.g., equipment, travel)
  • Mobics delivers a report after trial end with the main findings and proposals for further steps
  • Mobics allows the client to use the deployed pilot system even after the pilot and free of charge or with a preferential pricing license

It depends on the cameras used and the monitored area. We have successful tests with NIR cameras that detect smoke at night. Of course flame can be detected.

Of course. All of the Meleagros software is developed by Mobics and is based on open technologies and interfaces. Moreover, in many cases well-established standards have been adopted such as: OGC SOS, Common Alerting Protocol (CAP). The Application Programming Interfaces (APIs) are well-structured and can be further extended in order to allow for wider integration.

If you have to protect areas, assets or facilities of particular importance, every additional measure and system is worth (especially if it is as cost effective as Meleagros is).

However, if you want to enjoy more functionality and cover multiple needs with a simple system, then Meleagros is a good choice, too. Mobics can implement and design event and object detection algorithms and add them to the software of a Meleagros product in order to cover special needs. Some examples are: human detection, perimeter intrusion detection, vehicle classification and counting, illegal forest activity through audio processing, etc. Mobics researchers have the know-how and experience to work with you and customize the system functionality to your needs. Besides, smoke detection is one of the most challenging tasks in video analytics, so we are ready to work on any other problem!

After all, we hope that you never have a situation which will trigger our system!

The probability that a fire is not detected (aka a “false negative” case) is really small. If detection conditions are difficult in the region (very far incident, bad lighting conditions, dense fog and bad visual conditions in general), the system will just delay to trigger an alert. A careful installation study will ensure that the system does not delay.

It is almost impossible for any video analytics solution provider to promise exact false alarm numbers or rates. This happens because the monitored scene conditions can never be fully known or predicted. This is the case for Meleagros, too. Here is a brief list of what we do to improve system accuracy and reduce false alarms:

  • Scene characterization: the user can easily characterize what a camera sees in specific classes, such as forest, water, sky, built environment, road, etc.
  • Zone exclusion: the user can exclude zones where fire is not expected and may produce false alarms (e.g., chimneys).
  • Spatiotemporal self-learning algorithms: when the user explicitly indicates false alarms, the system searches for any spatiotemporal patterns in these (e.g., smoke in a specific tile every afternoon) in order to self-adjust its sensitivity.
  • Custom adjustment of specific detection parameters by Meleagros support staff, if required

We know that an alert itself does not constitute all necessary information for taking actions. For that reason we have included an event localization algorithm. The algorithm can work with a single camera or with multiple cameras (stereo vision approach). The localization accuracy depends on the size of the monitored area but in general the error is less than 100m. Such accuracy is quite sufficient as an initial estimate for fire events, since visual localization by first responders is very easy when they are close to the fire incident.

Want to learn more about the localization technology? Visit the technology section.

Common maintenance tasks are: 

  • Cleaning of cameras for better visual conditions
  • Upgrading of detection algorithms 
  • Other equipment maintenance tasks

In particular for the second task, remote administration is available. The owner of Meleagros products can connect to them locally (similarly to connecting to a DSL router) and open a secure connection with the Meleagros team servers. Through this connection the team can log into the device and perform maintenance or parameter adjustments operations. Moreover, the owner of the system can automatically download and install the latest software upgrades from a central distribution repository hosted on a Mobics server. 

Important note: For privacy and security reasons, only the owner of the products can initiate a remote assistance session. We have no other means to connect to an installed system.