In the last few years, the popularity of video analytics also known as intelligent video surveillance is growing remarkably.
That doesn’t only come from the increase of security threats. But it has also its origins from the need to manage the growing hours of recording that nobody has time to watch. And the need of automation task as well.
In this article, you will learn what is video analytics how it works and what is its real use
This video is a demo of a BriefCam. That helps for rapid video review and search
What is video analytics ?
Video analytics is the using software to detect, classify, and track objects or persons in video surveillance.
The software is either built in a camera, or in the NVR or it can be implemented in the server.
No matter where you install it. It will perform the same functions. When the server performs analytics. It must have extensive processing capabilities. So processing performed by the cameras reduces servers activity. Otherwise it could require high capacity as the number of cameras increases.
How video analytics works ?
First, it is essential to determine the event to focus on. In other words, we should determine which scenario the analytics function will detect.
Then the software captures the relevant actions, examines them and turns them into actions.
The recommended actions are transmitted to the surveillance systems or to the people who monitor them.
The actions that the system sends could be a pop up on a monitoring screen to alert the operator. It could also be a beginning of a recording. In other words a system registers only over movement. For example or certain act reducing network traffic and the need for storage.
Video intelligence includes the following functions:
- Video motion detection: It detects the movement in areas of the camera scene. It analyzes the image and detects visual changes in it
- Dynamic masking: It is a practice that excludes parts of a camera scene for privacy reasons. And also to ignore unwanted areas. We use dynamic masking especially in outdoor applications.
- Object identification: It is the function of locating and identifying objects in the scene
- Object recognition: Once a target is given, the function recognizes all the objects presence on the image
- Video tracking: track all the movements of a targeted objects or individuals in the scene
- Facial recognition: The software measures facial features from given image and recognizes it. So it detects it when it enters into the camera scene
- Automatic license plate reader: They are camera systems that scans autobile plates. They are mounted on street poles, streetlights, highway overpasses, mobile trailers, parking entrances, or attached to police squad cars. They capture license plate numbers that come into view and take a photograph of the vehicle. And sometimes the photograph includs its driver and passengers. Then upload it to a data base server.
- People counter: it counts people and detects over – crowding
Why using Video analytics ?
- Efficiency of the system:
video analytics makes the system smarter and more efficient. It helps to reduce the storage needs by reducing the records footage. In addition, the system analyzes the images and gives valuable output data to make decisions.
For example: Heat mapping can help retailer stores to understand their costumers’ movements. In fact, this function colors the image to detect movements and human heat throughout the store. So this can help for optimal placement of products
- Video analytics helps to manage the growing number of cameras:
Not only buildings and proprieties are protected by video surveillance, but even cities tend to be smart as well. Police departements and authorities watch streets and public areas. More than that, cities are pooling their own cameras with privately owned cameras. With so many cameras to watch, only analytics can help.
For example The detroit police department implemented a public-private-community partnership project. It consists of connecting private installed cameras with police headquarters to fight crime. it is clear that video analytics is the only way to process that large number of cameras data.
Click here to read more about Detroit green light project
- More opportunities for installers to expand their services, solutions and revenues
While video analytics is a save time and costs for an end user, it also offers big opportunities for an installer to expand his revenues and opportunities. In fact, by selling video analytics, security cosystem installer, provide himself with additional revenue by increasing the value and return on investment in the surveillance system,
Real use of video analytics
- Automated license plate readers (ALPRs)
Parkings can use this function. Highways toll stations and police cars are equiped with ALPRs to track cars
- Transportation: tracking
tracking of live traffic conditions at key locations in cities and highways.
Example New York city hypertext
- Heat detection
To detect feverish costumers in the corvid 19 pandemic
Read our post on Thermal imaging cameras
Conclusion
Video surveillance systems are not just a source of video. They are smarter thanks to video analytics and artificial intelligence. The challenge in this field is to make the computer understands the situations. Then be able to transform it into data to help make a decision.
