CMOS photosensitive matrices in CCTV cameras.

Currently, the market offers a fairly wide range of cameras for video surveillance systems with photosensitive elements made on the basis of CMOS matrices. However, it should be remembered that any photosensitive elements (photoresistors, photodiodes, phototransistors), as well as microcircuits made on the basis of both CCD and MOS structures, have not only their own technological features, but also quite significantly different parameters. Therefore, the fact that the photosensitive element of the camera is made on the basis of a CMOS matrix, in fact, does not yet carry the necessary information to the user about its technical characteristics.

Let's start with a description of CMOS microcircuits. They found more or less significant application in the 80s of the last century. At the same time, their serial production and active improvement began. CMOS microcircuits are improved microcircuits based on MOS (metal-oxide-semiconductor) transistors. Complementary microcircuits are created on the basis of CMOS cells, that is, two complementary MOS transistors, and all of them operate in the enrichment mode.

A complementary CMOS cell consists of two parts, one of which is built on the n-MOS structure, and the other on the p-MOS structure. These structures are complementary. The cell works in such a way that when one transistor is open, the other is closed.

The basic electrical circuit diagram of this cell is shown in Fig. 1. The p-MOS transistor is connected between the positive pole of the power supply Ec and the circuit output, and the n-MOS transistor is connected between the circuit output and the negative pole of the power supply Eu. The X input is connected to the gates of the n-MOS and p-MOS transistors. When a high voltage is applied to the X input, the p-MOS transistor is closed, and the n-MOS transistor is open. When a low voltage is applied to the X input, the p-MOS transistor is open, and the n-MOS transistor is closed.

All the advantages in the electrical characteristics of these circuits follow from the described circuit features. Since one of the transistors is always in the closed state, there is no current from Ec to Eu. Therefore, if the input signal does not change, the power consumption is equal to the product of the power source and the very small current flowing through the MOSFET. This determines the extremely low power consumption of the cameras.

The described circuit design features also determine such unique properties of the CMOS cell as independence of parameters from power supply fluctuations, noise and temperature variations. A typical relationship between the output and input voltages of a CMOS cell (its transfer characteristic) at different values ​​of temperature and source voltage is shown in Fig. 2. It is clearly seen that the curves have very minor differences over a wide temperature range.

In addition, changing the supply voltage only shifts the curve without significantly changing its nature, the input and output voltages change proportionally to the change in supply voltage, and the CMOS cell correctly performs logical operations in a wide range of supply voltages. This range is much wider than that of other types of integrated circuits.

The properties of the transfer characteristic, namely its steep drop, explain the high noise immunity of CMOS structures.

That is, in fact, the CMOS microcircuit provides the output signal value of 1 or 0, — there are no intermediate values. This principle, which is known to be used in digital signal transmission, has become one of the main ones in the use of CMOS microcircuits for creating video cameras with a digital output. After all, in this case, no additional and intermediate transformations of the received video signal are required as a result of photo transformations and reading for digital processing (BLC functions, etc.) and providing a digital output.

The first network cameras with IP output had low image quality and, accordingly, low prices. This was due to the fact that MOS transistors were used as photosensitive elements of CMOS matrices, which were manufactured in a single technological cycle with transistors of digital processing circuits. MOS transistors are characterized by low sensitivity to light. At the same time, the circuit organization of the matrix as a whole and the number of its sensitive elements were low (the resolution of these cameras was 320×240 pixels). The main companies that supplied cameras of this class, called web cameras, to Russia were Agilent Technologies (USA), Mitsubishi Electric (Japan), OmniVision Technologies, Photon Vision Systems, STMicroelectronics, Semiconductor Insights, Symagery Microsystems, Toshiba, VLSI Vision.

To solve systemic problems of organizing video surveillance using video cameras with digital output, higher quality cameras were, of course, required. And, naturally, manufacturers responded to market demand.

Today, video cameras based on high-resolution CMOS matrices are supplied by, for example, AXIS, Arecont Vision, Hunt (Taiwan). These are, as a rule, cameras with a large number of sensitive elements, better optics and advanced network capabilities. However, the matrices of such cameras were manufactured according to the previously described principle. Therefore, along with the advantages of CMOS technology, they also have such significant disadvantages as low noise characteristics and low sensitivity. These cameras rather belong to the class of web cameras — I mean, first of all, the technology for manufacturing their photosensitive matrix. The photosensitive element in it in the form of MOS transistors and reading microcircuits are also manufactured in a single technological cycle.

A separate niche is occupied by high-quality video surveillance cameras. The high-resolution photosensitive CMOS matrices used in them usually contain photodiodes integrated with a CMOS microcircuit in each pixel. Almost all problems with sensitivity, noise characteristics, etc. for photodiodes were solved back in the 70s and 80s of the last century. So, there are no problems with sensitivity here. But, it should be noted, such integration is a very expensive technology.

Partly for this reason, today this is a fairly narrow segment of the video surveillance technical equipment market. One of its undisputed leaders is Mobotix AG. For example, one of the cameras of this company, created specifically for work on alpine ski slopes, has a resolution of 1280×960 px, and its 1/2″ CMOS matrix has a high enough sensitivity for successful operation in day/night mode. Thus, as we have seen, CMOS matrices are different. The names of various photoelectronic devices are the same, but the architecture of construction is different. Accordingly, the characteristics of video cameras can differ by orders of magnitude.