How to choose a low-tech engineering HD network camera?

The designers of video surveillance systems should consider the use requirements of the project itself, including camera parameters, equipment procurement costs, equipment stability, and scalability.

It is not the higher the resolution of the design, the better the monitoring project. High resolution can result in a significant increase in the cost of front-end cameras, in-transit switches, and back-end storage devices. Often, a project designed by a mature monitoring designer will select cameras of different resolutions according to actual needs. Meet the needs of the use while controlling the cost of the project.

It seems that a 2 million and 3 million camera can not be a few dollars, H.264 and H.265 compression methods can not be a few dollars, but there is a big difference in data transmission and data storage.

Common monitoring system requirements for clarity

1, general monitoring

Can have a general understanding of the monitored area, can identify the basic judgment of animals, people, vehicles, colors, behavior, direction and so on. There is no face recognition requirement and no license plate recognition requirements. For example, look at the square, see the scenic area, and see the environment.

2, can be identified

Can accurately judge the target being monitored, for example, can identify someone, license plate number, and so on.

3. Detail monitoring (can be confirmed)

The details of the monitoring target can be confirmed, such as bank counter personnel billing action, cashiers in various industries, and ATM machine face monitoring.

Pixel density

Pixel density refers to the number of pixels within a certain range. The pixel density determines what we can see by the target we monitor. Pixel Density = Camera Horizontal Resolution ÷ Field of View (Left to Right).

For example, there are cameras 1920*1080 pixels, field of view coverage 1920cm, then 1920 (camera pixels) ÷ 1920cm (scene) = 1 pixel / cm = 100 pixels / meter, that is, 100PPM.

The pixel density decreases as the monitoring distance increases. To maintain the same pixel density at different monitoring distances, the long focal length lens should be replaced.

You can use the online pixel calculator software to see which pixel density is related to the parameters. Through the following figure, we clearly see the relationship between camera resolution, object distance and focal length. Select a 1080P camera, the fixed focal length value drags the object distance, and its resolution will decrease as the object distance increases; Fix the object distance value, drag the focal length, and the resolution increases with the increase of the focal length.

We confirm the required resolution according to the three levels of monitoring. Due to different needs, different departments will have different standards for this. In practical applications, subjective judgments will also be made, the environment of the scene, the camera. The quality of the image with the lens is affected by this. SKL (Swedish National Judicial Science Laboratory) recommends that the resolution for the logo should be at least 500 pixels/meter. The bottom right corner is the detected image of the scene, and the upper left corner is the enlarged recognition effect.

To give another practical example: assuming a face width of 16 cm, the recommended value for face confirmation is 80 pixels, then how many pixels do you need for a field of view width of 1 meter?

100 cm / 16 cm = 6.25, then 6.25 × 80 = 500 pixels / m = 5 pixels / cm, according to which can be derived as follows:

Scene coverage

So what is the range of a 1280x720 camera? Confirmation is required to be 500 pixels/meter and confirmed at 1280×720 resolution. 1280 pixels / 500 pixels per meter allows us to confirm at a distance of 2.56 meters. Therefore, the maximum scene width we can cover is 2.56 meters, meeting the 500 pixel/meter requirement. In summary, the higher the resolution of the camera, the wider the width of the scene at the same pixel density.