Megapixel cameras with CMOS matrix.

Megapixel cameras with CMOS matrix.

The ever-growing interest of the security market in megapixel IP cameras stimulates the transition of CCTV to a fundamentally new level, which is high-definition television. Megapixel IP cameras, the advantages of which are high resolution, high speed of image formation, digital representation, processing and signal transmission, are truly the missing link in a real digital system.
Currently, CCD (charge-coupled device) and CMOS (complementary metal-oxide-semiconductor) light-sensitive matrices with resolutions from 1.3 to 8 MP are used as image generators in megapixel IP cameras.
An important advantage of the CMOS matrix is ​​the combination of the analog and digital parts of the device on one crystal, i.e., the light-sensitive elements and the signal processor. In fact, each pixel is equipped with its own amplifier, which allows you to precisely adjust its exposure time. The fundamental differences due to technological features are given in the table. They allow you to evaluate the advantages and disadvantages of each technology.

Sensitivity
The typical sensitivity index for megapixel CMOS cameras is lower compared to CCD cameras. Of course, if we are talking about matrices of the same format. This is due to the fact that the photodiode of the light-sensitive cell occupies a significantly smaller area of ​​the matrix element compared to CCD. It is also important to note that manufacturers of megapixel cameras are strictly bound by the format of lenses used in CCTV, and as a result use CMOS matrices of 1/3, 1/2, and less often 2/3 inch formats. With increasing resolution, the sensitivity of the matrix decreases due to a decrease in the pixel area. However, CMOS technology is steadily developing, and the sensitivity of cameras is growing. In addition, manufacturers use technologies for increasing sensitivity that are widely known in security surveillance. Thus, to improve the ability to «see in the dark,» all manufacturers of megapixel CMOS IP cameras implement a «slow shutter» mode. This technology is well known to anyone who has done night shooting with a digital camera. Unfortunately, the characteristic blurring of the image does not allow using this mode for any task. There are other methods, such as: turning off the color (electronic «day-night» mode), turning off the IR filter (mechanical «day-night» mode), or switching between matrices. In the latter case, the camera can use two matrices — one high-resolution CMOS matrix for daytime shooting, the second with a lower resolution (usually CCD) and, accordingly, higher sensitivity for night mode.

Dynamic range
Due to the above-mentioned CMOS structure, a random access mechanism is implemented in the matrix, which makes it possible to read selected groups of pixels. By adjusting the exposure time in this way, CMOS sensor manufacturers can create cameras with a wide dynamic range, capable of forming a high-quality image in high-contrast lighting conditions. The typical dynamic range for CMOS cameras is about 60 dB versus 50–54 dB for CCD-based models.

Image Formation Speed ​​and Compression
Despite the fact that CCD matrices have a higher sensitivity characteristic, the main factor limiting their use is the low charge reading speed and, as a result, the impossibility of ensuring a high image formation speed. The higher the matrix resolution, the lower the image formation speed. In turn, CMOS technology, which combines a light-sensitive element and a processing chip, allows for high frame formation speed even for 3 MP sensors.
However, the use of megapixel CMOS sensors for IP cameras of video surveillance systems requires effective compression of the data stream. The most common compression algorithms in IP CCTV are currently M-JPEG, MPEG4 and H.264. The first is often implemented directly on the CMOS sensor by the matrix manufacturer itself. MPEG4 and H.264 algorithms are more effective, but require a powerful processor. To form a real-time stream with a resolution of more than 2 megapixels, CMOS IP cameras use coprocessors that provide additional calculations.

Power consumption
An important advantage of megapixel CMOS matrices is lower power consumption. Unfortunately, power consumption is still not a fundamental characteristic in our country that allows us to evaluate technical security equipment. In fact, a megapixel IP camera for outdoor installation is actually an incandescent lamp that works 24 hours a day. For example, the power consumption of a 1.3 MP CCD IP camera is about 9 W, taking into account the thermal housing in the cold season, such a camera consumes 30-40 W. Thus, we can imagine the power needs for an object consisting of 100 cameras. The typical power consumption of a 1.3 megapixel camera with a CMOS matrix is ​​two times less.

Cost of IP cameras with a CMOS matrix
Currently, IP cameras based on CMOS sensors are becoming increasingly popular primarily due to the support of the technology by IP CCTV leaders. At the same time, their cost is higher than that of similar cameras on CCD. And this is despite the fact that CMOS technology, which combines the analog and digital parts of the device, allows for the creation of cheaper cameras. The situation is such that today the cost of an IP camera is determined by its capabilities and characteristics. The matrix type is not fundamental, but the software implemented by the camera processor.

3-megapixel IP camera AV3100 “day-night” with M-JPEG, HDTV and PoE (Arecont Vision)
The AV3100 network camera transmits video with a 3-megapixel resolution and is designed for use in IP video systems or process television that require a high degree of image detail. The AV3100 is equipped with a 1/2-inch CMOS sensor with progressive scan, uses Arecont Vision MegaVideo® technology, has a sensitivity of 0.2 lux and can work with lenses with ARD. It is capable of broadcasting a video stream over an IP network in M-JPEG format with a resolution of up to 2048 x 1536 pixels at 15 fps or 1920 x 1080 pixels at 22 fps with a data transfer rate of up to 55 Mbps, supports TFTP and HTTP protocols, electronic PTZ and ROI functions and meets the requirements of the HDTV standard. The AV3100 features an alarm input/output, a 64-zone motion detector, can be powered by a 15-48 V DC source or via PoE technology, and comes with free AV100 software.

New megapixel IP camera TCM-4301 (ACTi )
The camera supports the modern H.264 data transmission and recording protocol at a speed of up to 18 fps (frames per second). The camera has a 1/3” Micron Progressive Scan CMOS matrix, a triple H.264/MPEG-4/MJPEG codec with two independent transmission channels. Resolution: SXGA (1280 x 1024); HD720(1280 x 720); VGA (640 x 480), QVGA(320х240), QQVGA(160×112). Video transmission speed up to 18 fps at SXGA, up to 24 fps at HD720, up to 30 fps at VGA and lower. 0.5 Lux at F 1.0. Bidirectional audio channel, built-in microphone. Built-in motion detector. Power supply DC 12 V and PoE.

Small-sized IP camera M1011 with H.264/M-JPEG/MPEG-4 and VGA (AXIS)
The M1011 color network camera from AXIS Communications is designed for indoor video surveillance. It supports H.264, M-JPEG and MPEG-4 Part 2 compression formats and can transmit multi-stream video with a resolution of up to VGA (640 x 480 pixels) at a rate of up to 30 fps. The M1011 is equipped with a 1/4-inch progressive scan CMOS sensor, electronic shutter, 4.4 mm lens with a fixed iris and a 47° viewing angle and transmits stable and high-quality video streams in illumination down to 1 lux. It supports full image quality adjustment and exposure control, has a motion detector and a 16 MB video buffer. Video can be accessed from any network PC via any web browser or via the AXIS Camera Station software interfaces.

Vandal-proof outdoor IP camera «day-night» 2530V with MPEG-4 and M-JPEG, D1 at 25 fps and IP66 protection (Cisco)
The 2530V day/night network camera has a vandal-proof thermal housing and is designed for outdoor use at temperatures from -30°C to +55°C. It is equipped with a 1/3°C CMOS sensor with progressive scan and WDR, a movable IR filter, a 3.5x varifocal lens with ARD and a swivel bracket for selecting the viewing direction. Thanks to this configuration, the IP camera is capable of generating color/black-and-white video at illumination of up to 0.4/0.04 lux and transmitting 1 or 2 video streams in MPEG-4 and/or M-JPEG with a resolution of up to D1 (720 x 576 pixels) and a frame rate of up to 25 fps. In addition, the Cisco 2530V provides bidirectional audio, motion detection and operator alerting, as well as access to its video streams through a web browser interface or Cisco CVSM software.

Pelco Sarix™ IXE20 1/3-inch Megapixel IP Camera with H.264 and M-JPEG and HD Resolution at 30 fps (Pelco by Schneider Electric)
The 2.1-megapixel IXE20 day/night network camera uses a powerful Extended Platform processor, proprietary Sarix™ technology and is designed to operate as part of IP video surveillance systems at illumination levels of up to 0.2/0.05 lux. It is equipped with a CMOS sensor with progressive scan, a mechanical IR filter, supports H.264 and M-JPEG compression algorithms and can broadcast up to 2 video streams simultaneously with a maximum resolution of 1920 x 1080 pixels. At the same time, the camera is capable of providing a speed of 30 fps with HD resolution. In addition, the Pelco IXE20 has a WDR and autofocus (ABF) function, an alarm input/output, a MiniSD card slot, supports PoE technology and can be controlled by a web browser or Pelco Endura and Digital Sentry software.

1.3-megapixel day-night IP camera STC-IPM3090A with M-JPEG/MPEG-4, 4.2 mm lens and audio channel (Smartec)
The universal network camera STC-IPM3090A «day/night» of Smartec brand is equipped with a 1.3-megapixel CMOS-sensor with progressive scanning, a movable IR-filter and forms high-quality color/black-and-white video at illumination up to 0.5/0.05 lux, having IR-sensitivity in the range from 700 to 1100 nm. It can simultaneously broadcast four video streams over the network with individually configured parameters in MPEG-4 or M-JPEG formats with a resolution of up to 1280 x 1024 pixels and a frequency of up to 30 fps. STC-IPM3090A is equipped with a megapixel lens with a focal length of 4.2 mm, supports PoE technology and has an audio input/output, a motion detector and an alarm input/output. In addition, the IP camera comes complete with Russified NVR software for 32 channels and can operate under Smartec NetStation or XProtect software from Milestone.

Network mini-dome camera SNC-DM160 (Sony)
The vandal-proof camera has a 1/3-inch CCD matrix with progressive scanning, ExwavePRO technology and a resolution of 1.3 megapixels. The camera has a high-resolution megapixel matrix, Light Funnel function. The quality of JPEG images is set using an algorithm with a constant digital flow. Variable gamma settings, the possibility of wall or ceiling mounting, simple adjustment of the viewing angle make the camera convenient to install and operate. The camera has a powerful varifocal lens (3.6x zoom, viewing angle over 100 degrees) with zoom and viewing angle adjustment, fast focusing. Selectable compression format — JPEG or MPEG-4, dual encoding is possible. Day-night function. Bidirectional sound transmission and voice warnings.SNC-DM160 has a DEPA platform — intelligent video analysis function, intelligent motion detection. Equipped with «sensor input/alarm output» ports. The camera complies with the IEEE802.1X standard. It has SolidPTZ/Cropping and privacy masking functions.

HD megapixel high-resolution IP camera (IndigoVision)
The camera implements the best-in-class compression standard H.264 (confirmed by the ISO 14496-10 standard), with a minimum resolution of 15 fps. Camera resolution is 1280 x 720 pixels. CMOS matrix 1/3″ Progressive Scan.
Power supply via PoE protocol via a network cable or an external 24 VAC source.
All IndigoVision cameras are equipped with lenses.
Built-in analytical algorithms (both in cameras and in the recording system) allow processing all images, increasing the efficiency of control and the volume of monitoring tasks performed without increasing the number of operators.
Analytical algorithms allow detecting clusters of objects, movements, abandoned objects, etc.
The variable frame rate (ACF) function allows, using built-in analytics, to constantly track the number of changes between frames and, in the absence of movement, the frame rate is reduced to a minimum of one per second. When movement occurs, the frame rate is instantly restored to the value specified by the user. This allows for multiple savings in recording capacity and transmission channel load.