What Is an Infrared Surveillance System?

Infrared surveillance is a technology that uses infrared imaging systems to monitor targets based on infrared technology. The infrared sensor converts the infrared signal from the target into an electric signal, amplifies the electric signal, and then performs three major processes of graphics, data, and control, and transmits the processed signal. We can use the background difference method, projection method and template matching method to process the moving target image.

The infrared imaging system can use the eye to see the heat emitted by the target on the field of view screen. This is because the target is always a little warmer or colder than the background, so the target is more prominent and easily distinguishable from the background. This contrast technology is also the thermal imaging seen at the exhibition (infrared imaging is often referred to as thermal imaging, thermal imaging, etc.). When using infrared imaging systems for surveillance, the observed objects may include people and things that are of interest to us, such as airplanes and ground vehicles. For more than two decades, infrared imagers have played a very important role in completing both military tasks and legal tasks. In terms of reducing environmental conditions, accurate, real-time information and 24-hour monitoring are required. These are extremely important for modern military command and decision-making, and they are related to the success or failure of military missions. ^[1]

Infrared radiation is due to the inside of the object

CCD camera can receive both visible and visible light

The monitoring system is mainly composed of monitoring software, TV wall, video capture card, twisted pair transceiver, Bosca, monitoring decoder, monitoring head and power control board.

Image segmentation and detection (recognition) is actually a very difficult task. It's hard to say why the image should be segmented this way instead of that. The human visual system is very superior. It includes not only the eyes, but also the brain. It can separate and recognize each object from very complex scenes, and it can effortlessly keep up with changes of dozens of frames per second image. The application of image detection in this system is not required to distinguish the detailed information of the moving object, it only needs to monitor the moving object and track its movement direction.

Because the surveillance system runs multiple channels simultaneously, the general video is about eight channels, and some can reach a dozen channels. If the calculation of the motion detection is complicated, the computer configuration requirements will be higher. The following are three common image processing methods.

Infrared surveillance background difference method

The principle of the background difference method is very simple: the two images before and after are subtracted, and the resulting difference is used as the resulting image. The picture can explain the principle of the background difference method.

Figure 4-1 is the foreground image (cat) plus background image (Jupiter). Figure 4-2 is the background image. The result of Figure 4-1 minus Figure 4-2 is shown in Figure 4-3, so that the foreground is obtained.

The background difference method is an important method of motion detection, and the difficulty lies in how to update the background.

Infrared surveillance projection

Projection is a natural idea, a bit like a grayscale histogram. For better results, projection is often used with thresholding. Take the following example: Figure 4-4 is the famous Washington Monument. The projection method is used to automatically detect the horizontal position of the monument. Observing carefully, it is not difficult to find that the gray levels of the pixels on the monument are similar and different. If we choose a suitable threshold and do clipping processing (here, 175 to 220), we can binarize the figure, as shown in Figure 4. -5 as shown:

Because the columns where the monument is located have more white points than the other columns, if the figure is projected vertically, as shown in Figure 4-6.

Infrared surveillance template matching method

Use template matching to find known objects in an image. For example, you captured a photo of a shot, and you need to find the position of the football in the photo. Then you can use the template matching method. The so-called template matching, in fact, the idea is very simple: take a known template (in this example, the image of a football) and match it with an area of the same size as the original image. At the beginning, the upper left corner point of the template and the upper left corner point of the image are coincident. Take the template and the same area of the original image for comparison, and then pan to the next pixel, still perform the same operation, all positions After everything is correct, the piece with the smallest difference is the object we are looking for. ^[2]