VisionLabs Luna Cars

The main articles are:

• Video Analytics (Russian Market)
• Video Analytics (Global Market)
• Neural networks (neural networks)

2024

Introduction of fire and smoke recognition technology

VisionLabs on July 22, 2024 introduced real-time fire and smoke recognition technology based on video data analysis. The solution will help ensure a prompt response to the occurrence of fire and stop the spread of fire in open spaces.

The system works with the video stream coming from the cameras in real time. To do this, it breaks personnel into the number of zones specified by the settings, in each of which an AIalgorithms analysis of the presence of signs of ignition is carried out. They are classified and prioritized by the following types: fire, black smoke and white smoke. When one of the attributes is detected, the system displays the incident in the event list and alerts the operator.

At the heart of the technology are neural networks trained on many thousands of different images collected by the VisionLabs team. The system is designed to operate in conditions where classic fire safety solutions are not used or ineffective - for example, on forest belts or in rooms with significant volumes and area, which means active air circulation, which makes it difficult for the usual physical sensors to quickly operate.

VisionLabs video analytics allows you to identify fires in the early stages, which is almost impossible when viewing images from dozens of cameras only by the operator. Artificial intelligence is able to continuously process a large stream of data, while working with constant accuracy without losing concentration. For comparison, according to studies, the operator cannot actively recognize objects present in video streams after 20-40 minutes. In addition, the level of misclassification increases to 95% after 22 minutes.

Early detection of a fire and the initiation of extinguishing allows faster localization of combustion and minimizes its impact and possible damage. This makes the technology especially in demand for large warehouses and distribution centers, landfills, agricultural land. It can also be used at transport hubs, construction sites and facilities of the oil refining industry, where it is important to warn people in danger zones and begin evacuation.

This technology has become part of the LUNA CARS product - an intelligent video analytics system for transport. The comprehensive solution is designed for use within intelligent transport systems and is already used in a number of regions. The system allows you to control traffic and in cases of emergency situations - an accident, fire or animal on the road - automatically notify the operator to call the emergency team, as well as display a warning for drivers on information boards on the highway.

Recognizing potentially dangerous situations such as fights, the presence of weapons, fire or smoke is one of the most sought-after areas of computer vision for smart and safe cities, and the VisionLabs team is actively developing these types of behavioral and situational video analytics. Our technology significantly increases the level of security at the facilities. With its help, you can quickly identify fires and prevent catastrophic consequences, - said Dmitry Markov, CEO of VisionLabs.

Astra Linux Special Edition Compatibility

VisionLabs and Astra Group on March 20, 2024 announced the completion of tests under the Ready for Astra technology partnership program. Test results showed that the LUNA CARS transport video analytics is compatible with the Astra Linux Special Edition operating system. The correct operation of the software in the protected OS environment is confirmed by the certificate. Both products are listed in the "Register of Russian Software" Ministry of Digital Development of the Russian Federation.

The use of LUNA CARS in a secure Astra Linux environment will allow the implementation of complex traffic management solutions: registration of transit traffic, navigation through the territory without the use of global positioning systems, electronic queues, road traffic monitoring and perimeter protection. The software stack will find its application in such areas as intelligent transport systems, security, traffic control.

The partnership between VisionLabs and Astra Group is aimed at creating fully domestic solutions that can be used in import substitution projects. VisionLabs' LUNA CARS video analytics system improves road comfort and safety, and automates traffic management. Its compatibility with Astra Linux makes it possible to provide fully domestic solutions for the digitalization of transport systems, including at critical infrastructure facilities, "said Dmitry Markov, CEO of VisionLabs.

Astra Linux allows you to close the infrastructure needs of everyone, industries industries including. transport This indicates a high level of manufacturability and versatility of our operating system. We have already worked with similar projects and are pleased that the boundaries of use of Astra Group products continue to expand. This means that even more end customers from the transport network can gain user experience when working with the VisionLabs solution by IT infrastructure implementing LUNA CARS based on Astra Linux in their video analytics, "said Alexey Trubochev, director of the Astra Group support department.

2022

At the heart of the VisionLabs suite of solutions for medical institutions

On December 26, 2022, VisionLabs presented a set of solutions for medical institutions based on computer vision.

As specified TAdviser in VisionLabs, the complex for medical institutions includes existing solutions that have been finalized taking into account the specifics of the industry. The complex is based on the flagship VisionLabs products - computer the vision platform LUNA intellectual video analysts and the LUNA transport CARS recognition system. Solutions have already been implemented in more than 15 clinics and medical centers throughout. Russia More. here

Integration with the Electronika Security Manager platform

On July 18, 2022, VisionLabs announced that using AI technologies from SOFT DIVISION and VisionLabs, it became possible to automate the process of passing vehicles and employees. The joint solution helps to reduce the burden on personnel while maintaining a high level of safety. The solution is based on the Electronika Security Manager (ESM) PSIM platform, which integrates a number of systems, including VisionLabs' LUNA CARS transport video analytics. Read more here.

Inclusion in the Unified Register of Russian Programs

More than fifteen VisionLabs products are included in the Unified Register of Russian Programs for Electronic Computers and Databases. The developer announced this on February 22, 2022. One of the products added to the register of domestic software is the LUNA CARS intelligent video analytics system for vehicle recognition. Read more here.

2020: LUNA Cars release

On October 15, 2020, VisionLabs, a resident of the Skolkovo Foundation's IT cluster, announced the release of a smart video analytics system for transport recognition - LUNA Cars. This product is able to determine most of the attributes of the vehicle: state number, brand, model, color, type and category.

VisionLabs works in the field of creating systems based on computer vision. Products based on this technology are in demand in many sectors of the economy and, in particular, in the field of transport, allow you to solve a wide range of problems. LUNA Cars - has every chance of taking a leading position in the market due to high recognition accuracy, versatility, high speed of the algorithm, noted Arkady Dvorkovich, Chairman of the Skolkovo Foundation

As of October 2020, the system recognizes more than 140 brands, 800 models and 5 categories of vehicles (A, B, C, D, E). Recognition accuracy is 95% and higher. LUNA Cars also defines all characters on license plates: from the series to the registration region code with an accuracy of more than 99%. In addition, the technology allows you to classify cars into several types: taxis, public and special transport.

LUNA Cars breaks the video stream into separate personnel. The neural network then detects the vehicle and its registration mark on them. However, it does not watch every frame - part of the path is completed using a tracking algorithm that predicts the position of selected objects in the following personnel. This saves computing resources and ensures high speed of the algorithm. Of the entire set of personnel, the two best are selected - separately by vehicle and separately by number, so that they are clearly visible.

The market demand has shifted from highly specialized systems to versatile and flexible solutions. Therefore, when creating LUNA Cars, our main task was to present a complex product. A wide range of potential applications makes it easy to scale it for specific business cases: road traffic monitoring, traffic management, automated access system for paid parking or protected areas, marketing analytics at gas stations, narrated by Dmitry Markov, CEO of VisionLabs

Due to high optimization and flexible algorithm settings, the system does not require large capacities and can be launched even on mobile devices.

An additional function of LUNA Cars is to determine the quality of images transmitted from the camera. This allows you not to start the recognition mechanism for images with overshooting, darkening or lack of sharpness, thereby reducing the load on the processor.