Contact Us




Add:5 Floor, Building D, Wuwu Village Industrial Park, Longhua Street, Longhua District, Shenzhen, China

Home > News > Content
What Are The Mainstream Technologies Of HD Surveillance Cameras? Mar 12, 2018

With the evolution of DSP and CMOS technology, DPS uses a separate exposure and control technique for each pixel. In addition, a multi-frame image acquired by a CMOS sensor is used to synthesize a linear image of a complete image. Compared to the CCD, the double exposure image has Higher dynamic range. From a numerical point of view, the current processing technology of the CMOS camera using DPS technology can reach 120dB or even 140dB. Wide dynamic technology has become an important indicator to measure the performance of a camera. For now, the standard wide dynamic function has become the consensus of all IPC vendors.


    HD network cameras have developed two or three intelligent analysis functions, such as motion detection and video occlusion, launched in 2011. Today, almost all mainstream security network manufacturers have more than 10 intelligent functions as standard. Of course, these intelligent functions are currently available. The vast majority of the standard is limited to mid-to-high-end industry products. According to the market business applications, these intelligent analysis functions can be divided into the following points:

    1, diagnostic intelligence analysis. The diagnostic intelligent analysis of HD network cameras is mainly aimed at the common camera failures such as black screen, blur, pan/tilt control, and picture freeze, video signal interference, such as scene change, item left/disappearance, etc., for accurate analysis, judgment, and alarm. . Diagnostic intelligent analysis technology is relatively simple to implement, usually these intelligent functions are integrated in the front-end, of course, the back-end such as NVR also do similar diagnostic intelligence.

    2, identification intelligence analysis. This technology of HD network cameras is biased towards the analysis and processing of static scenes. Through the core technologies such as image recognition, image matching, and pattern matching, the extraction and analysis of relevant characteristics information such as people, vehicles, and objects can be achieved. In the application of vehicle identification and analysis is mainly license plate recognition technology. License plate recognition technology is widely used in parking lots entrances and exits, highway toll gates and other places. In recent years, it has developed rapidly. In conjunction with the traffic electronic bayonet system, license plate recognition technology has been widely used to capture vehicle traffic violations, effectively reducing The number of vehicle traffic violations has greatly reduced the occurrence of traffic accidents.

    3, behavioral intelligence analysis. HD Network Camera This technology focuses on the analysis and processing of dynamic scenes. Typical features are: vehicle retrograde, zone intrusion detection, personnel focus detection, squall line crossing detection, rapid movement, personnel detection and passenger flow statistics. Motion Detection (VMD) is the "early intelligence" in intelligent analysis of this kind. VMD discriminates based on changes in the motion of pixel blocks in video frames. Because it is a two-dimensional image intelligence analysis, false positives are high, and motion cannot be recognized. Whether the pixel block is an interference or a target, coupled with the difference in the algorithm technology between security vendors, the accuracy of the behavioral and identification intelligence analysis is generally not high.

        Industry-recognized 0.001Lux and below are called Starlight cameras. The most representative starlight camera is the TIDM8127/Anba S2+ Sony IMX185 hardware solution, which is widely used in safe cities, finance, hotel buildings, Ping An Village, and ports. In projects such as highways and highways, there is no need to install large-scale lighting installations on a large scale, and a good nighttime high-definition color monitor screen can be obtained. Starlight-level illumination monitoring technology is mainly affected by factors such as the lens, image sensor, and back-end image processing technology. Each of the security manufacturers is also upgrading from the following aspects:

    1, the use of large aperture lens: the lens is an important part of the camera component, its role in low-light monitoring application technology is to focus the camera's target camera's light, where the low-light application and technology is the key to the lens's caliber The larger the amount of light it enters, the larger the aperture of the lens can effectively increase the amount of light entering the camera, so that the camera can obtain the ideal low illumination effect.

    2, choose a large target sensor: the essence of the camera is to convert light energy into electrical energy, and the core component of the quantification is the sensor, the sensor's role is to pass to the body of different intensity of light photoelectric conversion, converted into voltage information eventually Generate digital image information. The part of the sensor that receives light is naturally the core of the core. If the camera with the same resolution, the larger the target area of the image sensor, the greater the amount of light per unit pixel, the stronger the ability to suppress noise, and the better the image quality when shooting with low illumination.

    3. Good image processing technology: In the past, cameras used traditional 2D algorithms to achieve noise reduction, but now the 3D noise reduction technology used in the original frame based on noise reduction, through the comparison of the image before and after the two frames Filter processing to find out the noise position and gain control of it. The 3D digital noise reduction function can reduce the noise interference of the weak signal image. Because the appearance of image noise is random, the noise appearing in each frame image is not the same. 3D digital noise reduction automatically filters out non-overlapping information (ie, noise) by comparing adjacent frames of images. Using 3D noise reduction cameras, the image noise will be significantly reduced, and the images will be clearer and more transparent, thus showing a comparison. Pure and delicate picture.