Both LiDAR and camera-based people counting systems have their advantages and limitations. But which should you choose for your project?
People counting has become a critical need for businesses. Various research agencies have forecast the people-counting solutions market to grow at solid rates of up to 10 percent CAGR over the next 4-5 years.
Video surveillance camera-based people counting systems that leverage the power of analytics have become a popular choice over recent years. But on the other hand, solutions that use LiDAR technology provide certain unique advantages. This article explores the pros and cons of both LiDAR-based and camera-based people counting.
Also read: Is radar a better choice for perimeter protection?
Advantages of LiDAR-based people counting
Gerald Becker, VP of Market Development and Alliance at Quanergy, listed several key factors that make LiDAR technology a better choice for counting:
- When using LiDAR-based solutions, you don’t have to worry about environmental issues that are often part of camera deployments in outdoor environments. The sun and shadows don’t affect this solution.
- They are also not dependent on pixel movement for motion detection in a direction for a typical line-cross deployment for count with cameras. “We also see in 3D which inherently provides better coverage than most sensors like beam cross sensors at door entry/exits,” Becker pointed out.
- These don’t require high-end GPU processing for computer vision algorithms to run VA video analytics, offering an appealing TCO.
- One of the most significant advantages is installation location. Unlike cameras that need to be mounted overhead at pinch points as people enter and exit, LiDAR solutions have more flexibility. “With our MQ8 sensor, you can have the sensor mounted on the opposite side of the venue and still be able to configure virtual line cross zones that will provide a high degree of accuracy for people counting,” Becker added.
Advantages of camera-based people counting
Gunnar Erlandsson, Global Product Manager for Analytics Solutions at Axis Communications, points out that camera-based people counting had replaced thermal counters as the preferred method to do people counting. LiDAR is still a new technology not widely used for people counting.
“Camera-based people counting systems can better identify persons walking closely together and better manage people loitering in the counting zone, thus offering higher accuracy,” Erlandsson said. “Although the video is not the primary output from a camera-based counter, it offers the possibility to verify the counting accuracy and fine-tune the installation. Something which is not possible with non-camera-based systems.”
Limitations
Both LiDAR, as well as camera-based people counting solutions have their limitations. Erlandsson said that the challenges for any people counting system are to distinguish people from other objects, for instance, shopping carts, baby strollers, animals, or shadows on the floor, and to correctly identify people walking closely together, walking very fast or carrying large objects.
“Using cameras for people counting offers the best opportunities to manage these situations. Stereo camera technology and modern AI-based algorithms trained with relevant video material further increase the accuracy,” Erlandsson added. “Often there is a need to separately count adults and children and to separate staff from customers. Again, stereo camera technology together in combination with well-trained AI-based algorithms is a method ideal for correct identification of adults and children. By having staff wearing special badges, special clothes or RFID tags, they can be identified separately to members of the public.”
Becker added that the limitations of a LiDAR-based people counting system are similar to a camera in the sense of line-of-sight obstacles.
“But we have much more flexibility in the installation location of the LiDAR sensor,” Baker continued. “The lasers shoot into an open space and can see with a wider field of view. This, of course, is a bigger limitation of other sensors because they can only see what’s in front of the lens.”
Conclusion
Both LiDAR and camera-based people counting systems have their advantages and limitations. The final decision to choose either of them would depend on many factors, including the nature of the location, budget, processing power, etc. However, as a steadily growing market, we can see more technological developments in this area in the coming years.