RU2420018C1 - Television camera for viewing in conditions of low illumination and/or low brightness of objects - Google Patents

Abstract

FIELD: physics. ^ SUBSTANCE: in a television camera, having an objective lens, a matrix of charge coupled devices (CCD matrix), a control pulse generator and a signal processor, the surface area of the gate S1 of the field-effect transistor, which gathers charge carriers in the first charge-to-voltage converting unit of the CCD matrix, is based on the criterion of maximum control capacity of charge conversion, and the surface area S2 of the gate of a similar field-effect transistor in the second charge-to-voltage converting unit is based on the criterion of minimum inherent noise in the image signal, where S1>S2. ^ EFFECT: wide dynamic range of the image formed by the television camera through optimisation of charge-to-voltage conversion in the photodetector. ^ 5 dwg

Description

The present invention relates to cameras operating in conditions of complex lighting and / or complex brightness of objects, when in the field of view of the camera can be simultaneously highly and dimly lit objects and / or objects with a sharp difference in brightness.

The closest in technical essence to the claimed invention should be considered a camera [1], containing sequentially located and optically coupled lens and a matrix of devices with charge coupling (CCD matrix), consisting of successively connected charge connection storage section, the first storage section, the first output register, the first block of conversion of charge to voltage (BPS), a separation electrode, a second storage section, a second output register and a second BPS, and also a pulse control generator in, consisting of a pulse shaper (FI), a phase shaper (FF) and five level converters (PU), and the outputs of the clock pulses, horizontal blanking pulse, frame blanking pulse and control signals of the FI registers are connected to the first, second, third and fourth FF inputs respectively, and the first, second, third, fourth and fifth FF outputs are connected to the inputs of the first, second, third, fourth and fifth controllers, respectively, while the outputs of the first, second, third, fourth and fifth controllers are respectively the first, second, third, fourth and fifth outputs of the sweep signal of the control pulse generator, and the output of the PI synchronization pulses is the output of the synchronization signal of the control pulse generator, which is connected to the control input of the signal processor, consisting of the first and second video amplifiers, a comparator, the reference input of which connected to the reference voltage, and the video switching unit (BKV), and the output of the second video amplifier is connected to the information input of the comparator and to the first information input BKV, the second information input of which is connected to the output of the first video amplifier, the control input of which is combined with the control input of the first video amplifier and connected to the control input of the signal processor, while the information input of the first video amplifier is the first information input of the signal processor, the information input of the second video amplifier is the second information input of the signal processor, and the output of the BKV - the output of the signal processor, while the output of the first CPS of the CCD matrix n to the first information input of the signal processor, the output of the second BPZN - to the second information input of the signal processor, the output of which is the output of the "Video" camera.

When operating the prototype camera, conditions of complex lighting and / or complex brightness of the observed plots and objects are possible. In foreign literature, these observation conditions are called “Back light”.

Examples of complex lighting conditions include:

- observation of the launch of a rocket with a bright torch from running engines;

- observation through a window or against the background of open doors, when you need to simultaneously distinguish between objects on the street and in the room;

- observation against diffused sunlight;

- observation against the background of glare, lighting lamps and more.

A typical example of complex brightness conditions for industrial television is the monitoring of the hot-rolled zone in metallurgy, as well as the control of arc or electron-beam welding processes in automatic or semi-automatic mode.

The prototype camera provides an extension of the dynamic range of gradations of brightness of the images of the observed scene by forming a combined image, which is the result of the synthesis of video signals generated by the CCD with “long” - “long charge” and “short” - “short charge” in charge accumulation time.

The disadvantage of the prototype is the limitation of the dynamic range of the image of the camera due to the limited capabilities of the photodetector. When choosing the same gate area (S) for a field-effect transistor collecting photo charges in the first and second SPRs, only a compromise solution is possible. With a small shutter area, the introduction of intrinsic noise into the image signal is reduced and the necessary camera sensitivity is achieved. But at the same time, the control ability of the charge conversion is limited, which affects large-level charge packets at the input (at high illumination or brightness of the objects under control), causing a limitation of the upper boundary of the dynamic range. With a large shutter area, on the contrary, the sensitivity of the camera will be deteriorated and the lower limit of the dynamic range will be limited.

The objective of the invention is the expansion of the dynamic range of gradations of brightness generated by the camera image by optimizing the charge-voltage conversion in the photodetector.

The problem is solved in that in the inventive camera, which contains sequentially located and optically coupled lens and a CCD matrix, consisting of sequentially charge-coupled sections of the accumulation, the first storage section, the first output register, the first SPS, the separation electrode, the second storage section, the second the output register and the second overhead voltage protection device, as well as a control pulse generator and a signal processor, the control inputs of the accumulation section, the first storage section and combined at the control inputs of the first and second output registers of the CCD matrix are connected respectively to the first, second and third outputs of the scan signal of the control pulse generator, the fourth and fifth outputs of the scan signal of which are connected respectively to the control input of the second storage section and to the control input of the separation electrode of the CCD matrix, the output of the first BPSN which is connected to the first information input of the signal processor, the second information input of which is connected to the output of the second BPS, and yayuschy input - to the output of the generator control pulse synchronizing signal, the output of the signal processor is output "Video" camera, the shutter S ₁ FET area, performing the collection of the charge carriers in the first BPZN is formed by the criterion of maximum control capacity charge conversion, and the area the gate S _{2 of a} similar field-effect transistor in the second SPR - according to the criterion of the minimum introduction of intrinsic noise into the image signal, while S ₁ > S ₂ .

Comparative analysis with the prototype shows that in the device of the inventive camera there is a constructive difference between the first and second overvoltage arresters in terms of the gate area of the field effect transistor, which performs element-by-element collection of charge image carriers.

The combination of known and new features is not known from the prior art, therefore, the claimed solution meets the requirement of novelty.

In the proposed solution, at the first and second outputs of the photodetector, video signals are generated that are generated at two different durations of accumulation of the CCD matrix. Thanks to the optimization of the charge-voltage conversion for light and dark fragments of the scene in the photodetector for two outputs and the synthesis of the combined image in the output video signal of the camera, an additional expansion of the dynamic range is provided. Achievable technical result is a beneficial advantage of the proposed camera when monitoring the observed scene, for which in one field of view there is a sharp difference between objects in terms of illumination and / or brightness. According to the technical result and methods for its achievement, the claimed solution meets the requirement of inventive step.

Figure 1 shows the structural diagram of the inventive camera; figure 2 - light characteristic of the photoelectric conversion carried out in a television camera; figure 3 is a possible structural diagram of a generator of control pulses; figure 4 is a possible structural diagram of a signal processor; 5 is a timing diagram explaining the operation of the camera.

The inventive camera (see figure 1) contains sequentially located and optically coupled lens 1 and a CCD matrix 2, consisting of sequentially charge-coupled accumulation sections 2-1, the first storage section 2-2, the first output register 2-3, the first EPR 2-4, a separation electrode 2-5, a second storage section 2-6, a second output register 2-7, and a second BPZN 2-8; a control pulse generator 3 and a signal processor 4, the first output of a scan signal of the control pulse generator 3 being connected to the control inputs of the accumulation section 2-1, the second output of the scan signal to the control inputs of the storage section 2-2, and the third output of the scan signal to combined between the control inputs of the output registers 2-3 and 2-7, the fourth output of the scan signal to the control inputs of the storage section 2-6, the fifth output of the scan signal to the control input of the separation electrode 2-5, and the signal output onizatsii - to the control input of the signal processor 4, a first information input connected to the output BPZN 2-4, second information input - to the output BPZN 2-8, and the signal processor 4 output is the output "Video" cameras.

Matrix 2 of the CCD, as in the prototype, has the organization of “personnel transfer” and contains three sections on one chip, two output registers, a separation electrode, and two overvoltage protection devices. Note that the domestic device A-1131 with an n-type channel and three-phase section and register control, designed specifically for television motion detectors, in order to isolate the difference between frames, potentially has the necessary circuit organization and can be considered a technological groundwork for the implementation of the invention. Information about the A-1131 device was published in [2].

A feature of the CCD matrix 2 in comparison with the prototype photodetector is the difference in the design of blocks 2-4 and 2-8.

Block 2-4, like block 2-8, is designed to convert the charge signal of the image into the voltage of the video signal. Their principal difference is the different level of charge packets at the input, which is taken into account in the design of the output field effect transistor in terms of the capacitance of its gate. For BPS 2-4, a high level of useful charge signal is expected, therefore, it is necessary to increase the control ability of the unit by increasing the shutter area (S ₁ ). On the contrary, for BPS 2-8, a low level of the useful charge signal is assumed, therefore, the shutter capacity should be extremely small, which is achieved by choosing a geometry of its size that provides a small area (S ₂ ). So the condition for the design of load transistors is: S ₁ > S ₂ .

Here it is necessary to provide a technical justification for this proposal. The fact is that in each element of the video signal output from the CCD matrix, in addition to the photon noise, an intrinsic noise source, called read noise, appears. The standard deviation (RMS) of the read noise is determined by the gate area of the output field effect transistor. The typical value of the standard deviation of reading noise is 20 electrons [3, p.52], and it is the result of designing an output transistor, the gate area (S) of which contains the maximum value of the expected charge, i.e. meets the criterion of maximum control ability of the charge photoelectronic conversion carried out in the CCD matrix.

According to experts, theoretically, the standard deviation of read noise can be reduced by an order of magnitude [3, p. 52]. Therefore, the reason for limiting the dynamic range from below is already inherent in the typical organization of the CCD matrix, and to expand the dynamic range it is advisable to improve the matrix itself.

On the light characteristic of the claimed camera (see Fig. 2), point B corresponds to the minimum (threshold) illumination of the camera E _min2 for the threshold signal-to-noise ratio Ψ _pore = 6, when the shutter area S _{2 is} selected by the criterion of the minimum inclusion of the matrix intrinsic noise into the video signal. Point A on the same characteristic corresponds to the minimum illumination E _min1 for the same threshold signal-to-noise ratio Ψ _pore = 6, and the area S _{2 is} selected according to the criterion of the maximum control ability of the charge conversion. Therefore, the ratio of threshold illuminances can be:

The generator 3 of the control pulses is designed to perform a scan in the matrix 2 of the CCD and generate a synchronization signal for the signal processor 4. A possible structural diagram of the generator 3 of the control pulses, which coincides with the solution of the prototype [1], (see figure 3) contains FI 3-1, FF 3-2, the first launcher 3-3, the second launcher 3-4, the third launcher 3-5, the fourth launcher 3-6 and the fifth launcher 3-7, while the outputs of the clock pulses, horizontal blanking pulse, frame blanking pulse and register control signals FI 3-1 connected to the first, second, third and fourth to the inputs of FF 3-2, respectively, and the first, second, third, fourth and fifth outputs of FF 3-2 are connected to the inputs of PU 3-3, PU 3-4, PU 3-5, PU 3-6 and PU 3-7 respectively, while the outputs of PU 3-3, PU 3-4, PU 3-5, PU 3-6 and PU 3-7 are respectively the first, second, third, fourth and fifth outputs of the sweep signal of the generator 3, and the output of the synchronization pulses FI 3-1 - its synchronization signal output.

The FI 3-1 and FF 3-2 blocks can be made in the form of a large integrated circuit (LSI), for example, using Sony CXD2463R chip technology [4]. The remaining blocks of the generator 3 control pulses, namely: PU 3-3, PU 3-4, PU 3-5, PU 3-6 and PU 3-7, can be implemented in the form of a second LSI necessary set.

The signal processor 4 is intended for two-channel amplification and processing of the image signal from the outputs of the CCD matrix and the formation of the combined video signal at the output. A possible structural diagram of the signal processor 4, which coincides with the solution of the prototype [1] (see Fig. 4), contains a first video amplifier 4-1, a second video amplifier 4-2, a comparator 4-3, the reference input of which is connected to the reference voltage U _n , and BKV 4-4, while the output of the video amplifier 4-2 is connected to the information input of the comparator 4-3 and to the first information input of the BKV 4-4, the second information input of which is connected to the output of the video amplifier 4-1, the control input of which is combined with the control 4-2 video amplifier input, is the control input processor house 4, the information input of video amplifier 4-1 is the first information input of processor 4, the information input of video amplifier 4-2 is the second information input of processor 4, and the output BKV 4-4 is the output of processor 4.

All blocks of the signal processor 4 can be made in the form of a single LSI, for example, using the technology of the Sony CXA1310AQ microcircuit [5].

The camera operates as follows.

Suppose that the CCD matrix 2, like the A-1131 sensor [2], is a three-phase device with an n-type channel. Therefore, the first, second, third and fourth outputs of the scan signal of the generator 3 have three control phases for each of the outputs, and the fifth output is single-phase.

We use the time diagram shown in figure 5, which shows the diagrams of the output signals of the generator 3, including:

- figb - the first output, the first phase,

- figv - the first output, the second phase,

- fig.5g - the first exit, the third phase,

- fig.5d - the second output, the first phase,

- fig.5e - the second output, the second phase,

- fig. 5g - second exit, third phase,

- figi - the third output, the first phase,

- fig.5k - the third output, the second phase,

- fig.5l - the third exit, the third phase,

- fig. 5m - fourth exit, first phase,

- fig.5n - the fourth output, the second phase,

- figo - the fourth output, the third phase,

- figs - fifth exit.

Note that the diagrams of the output signals are presented in the time diagram with respect to the frame quenching pulse with a period of T _to and a duration of t _o.h.c. (see figa). The level of the control potential, which ensures the accumulation and transfer of charge packets for the n-channel matrix, relative to the substrate ¹ (the substrate level of the photodetector is taken equal to the potential of the "common wire" (housing) and coincides in the diagram drawing with the position of the time axis) is high.

Consider the operation of the camera, starting from the end of the reverse frame scan indicated on the diagram of figa moment t ₀ . From this moment, during the forward frame scan T _n1 , a “long” charge accumulation of the frame, “long charge”, is _performed on the photo target 2-1 of the CCD matrix 2. Then, in the frame transfer interval T _{p1, the} charge picture of the “long charge” frame is rewritten from section 2-1 through section 2-2 and register 2-3 into section 2-6. This becomes possible due to the high level of potential present at the time T _p1 on the separation electrode 2-5 (see fig.5z).

At the end of the interval T _p1 in section 2-1 during the time T _n2 is a "short" accumulation of charges - "short charge".

In parallel with the accumulation of the "short charge" of the frame in section 2-2, the array is cleaned of stray charges by applying three-phase pulse sequences to its phase electrodes, which have a frequency equal to the frame removal frequency F = 1 / T _in (see fig. g). The maximum duration of the cleaning interval T _about equal to the interval T _n2 .

The pulse control signals of the register 2-4 (see Fig. 5i-l) provide such a movement of carriers that during the horizontal line reverse (t _oxc ) the charge strings of the spurious signal are added (enlarged) in it under the second phase, and then during forward line scan these parasitic charges are element-wise transferred to BPS 2-4.

At the end of the accumulation of the "short charge" of the frame, the charge pattern in the interval T _{p2 is} rewritten from section 2-1 to section 2-2 previously cleared of parasitic charges (see fig. 5b-d and fig. 5d-g).

The minimum cleaning interval T _about is:

T _a = N _a T _c (T _a / t _o.h.s)

where N _with - the number of lines in section 2-1;

T _with - period of the line.

Take typical parameter values; N _c = 290; T _s = 64 μs; T _at = 0.6 μs; t _oxc = 12 μs. As a result, the required value of Т _о is equal to 928 μs, and in total with the personnel transfer interval T _p1 equal to 2N _c • T _in and amounting to 348 μs, and with the personnel transfer interval T _p2 equal to N _c • T _in and amounting to 174 μs, does not exceed 1450 μs. The obtained time period completely "fits" into the reverse course of the personnel sweep t _o.x.k. since it takes up an interval of less than 1600 μs.

Considering that the control ability of the CCD output register is several times higher than the control ability of the storage section, and the distribution of stray charge decreases in the “top to bottom” direction, the matrix can be completely cleared of stray carriers during the interval T _o .

Note that during the accumulation of T _n2 “short charge” of the frame and its transfer in the interval T _p2 , the storage of the “long charge” of the frame under the second phase electrodes of section 2-6 is carried out (see Fig. 5n).

Then, in the interval of the forward stroke of the current frame, parallel readout of the “short charge” packets in the register 2-3 and in the LPS 2-4 and of the “long charge” packets in the register 2-7 and in the LPS 2-8, respectively. Note that for register 2-3 this becomes possible due to the fact that the low potential level present at this time on the separation electrode 2-5 (see FIG. 5z) “isolates” it from section 2-6.

Let us designate, as in the prototype, the corresponding video signals of the frames as “short signal” and “long signal”.

In contrast to the prototype, where the gate area of the field-effect transistors in both SPLs is equal to S ₁ and performed according to the criterion of the maximum control ability of the charge conversion, the gate area S _{2 of the} field-effect transistor in BPZN 2-8 in the claimed solution is made according to the criterion of the minimum introduction of intrinsic noise into the image signal matrix 2 CCD, with S ₂ <S ₁ . Due to this, the “long signal” video signal at the output of the WPS2-8 with the same illumination of the observed scene will have an increase in the signal-to-noise ratio Ψ, and the threshold illumination of the scene will be guaranteed to be reduced (E _min2 <E _min1 , see _figure 2).

In the signal processor 4, a combined image signal is synthesized, which is implemented in exactly the same way as the prototype. But the synthesized video signal provides an additional extension of the dynamic range of the camera, since in comparison with the prototype it optimized the transformation of the "charge-voltage".

The expected gain in the dynamic range estimated by the ratio

in the claimed solution is at least 20 dB or 10 times.

Currently, all blocks of the proposed solution have been mastered or can be mastered by domestic industry, therefore, the proposed invention should be considered as meeting the requirement for industrial applicability.

INFORMATION SOURCES

1. Patent No. 2235443 of the Russian Federation, IPC ⁷ H04N 5/335, 3/14, 5/202. Television camera on a matrix of devices with charge coupling / V.M.Smelkov // B.I. - 2004. - No. 24.

2. Skrylev A.S., Starovoitov V.I., Frost N.I. Photosensitive device with charge coupling A-1131 // Electronic industry. - 1991. - No. 7. - p. 83.

3. Nikitin VV, Tsytsulin A.K. Television in physical protection systems. S-Pb .: SPbGETU "LETI", 2001.

4. Sony CXD2463R chip. Timing Controller for CCD Camera. User manual in English, p.1-12.

5. The CXA1310AQ chip from Sony. Single Chip Processing for CCD Camera. User manual in English, pp. 1-14.

Claims (1)

A television camera for observation in conditions of complex illumination and / or complex brightness of objects, containing sequentially located and optically coupled lens and a matrix of devices with a charge coupling (CCD matrix), consisting of successively connected charge communication storage section, the first storage section, the first output register, the first block of charge-to-voltage conversion (BPS), a separation electrode, a second storage section, a second output register and a second BPS, and also a generator of control pulses owl and a signal processor, the control inputs of the storage section, the first storage section and the combined control inputs of the first and second output registers of the CCD matrix are connected respectively to the first, second and third outputs of the scan signal of the control pulse generator, the fourth and fifth outputs of the scan signal of which are connected respectively to the control input of the second storage section and to the control input of the separation electrode of the CCD matrix, the output of the first over-voltage converter of which is connected to the first information the input of the signal processor, the second information input of which is connected to the output of the second SPS, and the control input is connected to the output of the synchronization signal of the control pulse generator, while the output of the signal processor is the Video output of the camera, characterized in that the gate area S _{1 of the} field-effect transistor, performing the charge carriers in the first charge BPZN is formed by the criterion of the maximum charge capacity control conversion, and the gate area S ₂ a similar FET in the second BPZN - criterion of the minimum entry in the intrinsic noise of the image signal, the S _1> S _2.

Images