#Special Equipment
Method for minimizing distortions of a television camera when operating under conditions of light overload.
Vyacheslav Mikhailovich Smelkov,
Candidate of Technical Sciences
METHOD OF MINIMIZING DISTORTIONS OF A TELEVISION CAMERA WHEN OPERATING UNDER CONDITIONS OF LIGHT OVERLOAD
Source: magazine «Special Equipment»
The first issue of the journal for 2001 published an article by A.N. Kulikov, “Television Surveillance in Bright Sunlight,” which examined the problems of operating a CCD-matrix television camera under excessive illumination at the site and provided practical recommendations for reducing the video signal distortions associated with these conditions.
The author of this article proposes to supplement the “arsenal” of known methods for combating light overload (or protecting against it) with another effective technical solution.
In the aforementioned work, the upper limit of the working illumination range for a security television camera is estimated at 100,000 lux, which is considered a reliable characteristic of observation conditions in bright sunlight.
However, it should be acknowledged that this result was obtained using a modern Japanese Sony CCD matrix in the camera, manufactured using EXWAVEHAD technology, in the electronic shutter mode of the photodetector, when the exposure time (accumulation time) is 1/1000 000 s.
In this case, the integral signal of the “blur” of the television image is 10%, which corresponds to the “borderline” transition from acceptable to low quality for visual perception of the image.
A significant reduction in the “smear” signal at an illumination of 100,000 lux can be achieved in a television camera using two CCD matrices.
The structural diagram of the camera is shown in Fig. 1.
Fig. 1. Structural diagram of the television camera
The television camera comprises a lens (1), a first television signal sensor (2), a power supply unit (3), a beam splitter (4), a second television signal sensor (5), a sync pulse selector (6), a test line former (7), a peak detector (8), a comparator (9), and a switching unit (10).
The beam splitter (4), when the photo targets of the sensors (2) and (5) are mutually perpendicularly arranged, comprises a sequentially arranged and optically connected semitransparent mirror and a neutral light filter, and when the photo targets are mutually parallel, it comprises a semitransparent mirror, a neutral light filter, and a reflective mirror (not shown in the figure).
The peculiarity of the sensor (2) is the application of an opaque mask on the upper or lower line of the CCD matrix photo target to isolate the “clean” “smear” signal – interference from the “smearing” of the image.
In extreme cases, if during the operation of the camera it is possible to avoid strong localized illumination of the upper or lower part of the projection of the optical image on the sensor’s photo target (2), then the application of the corresponding opaque mask can be avoided.
Let's distinguish two sections in the range of working illumination of the camera:
- section 1 — corresponds to low and medium illumination on the object;
- section 2 — corresponds to illumination on the object from above average to high.
Regardless of the illumination on the object, the input optical image through the lens (1) and the beam splitter (4) is projected simultaneously onto the photo targets of the sensors (2) and (5).
Regardless of the input illumination, the selector (6) selects line and frame sync pulses, the generator (7) produces the control signals “Strobe” and “Reset” at the output for the peak detector (8), and the detector (8) itself, in the interval of the first or last active line of the half-frame, registers the current level of “smear” of the photodetector of the television signal sensor (2).
Let us assume that the illumination on the object is within the limits of section 1. Then the voltage at the output of the peak detector (8) is necessarily less than the threshold voltage Up of the comparator (9), and the logical level “0” is set at the output of the latter.
The zero level of the control signal in the switching unit (10) ensures the transmission to the “video” output of the television camera of the full video signal from the sensor (2) and the disconnection of the supply voltage from the sensor (5).
Let the illumination on the object increase and move to section 2. In this case, the output voltage of the peak detector (8) will exceed the threshold voltage Uп of the comparator (9), and the output of the latter will set the logical level “1”. The unit level of the control signal in the switching unit (10) will lead to the supply voltage being supplied to the sensor (5) and the transmission of its full video signal to the “video” output of the television camera.
The neutral filter of the beam splitter (4) reduces the illumination, due to which the sensor (5) does not experience light overload on the photo target. Therefore, in the proposed camera for the upper section of the range, the signal/noise ratio from image “blurring” is maintained at a level no higher than the permissible level and for the maximum limit values of illumination on the object.
If the illumination on the object decreases and moves to section 1, then the camera will again perform automatic high-speed switching via video from sensors (2) and (5), and the supply voltage will be supplied only to sensor (2).
Thanks to the latter, economical power consumption of the camera is maintained in section 1, and additional power consumption occurs only in section 2.
It should be noted that the proposed solution does not lead to a loss of sensitivity of the camera in section 1, and as a sensor (2) it is recommended to use a camera module in which a CCD matrix of the EXWAVEHAD model from Sony with improved sensitivity by 3…4 times is used.
On the other hand, a cheaper camera module can be used as a sensor (5), where a CCD matrix with normal sensitivity (standard CCD) is used.
Let's perform an engineering calculation of the technical result of the proposed solution.
Let the permissible value of the integral signal of image «blurring» be 1%, i.e. the signal/noise ratio from image «blurring» in the camera output signal be 40 dB.
Let us use the graph of the dependence of the “smear” signal on the accumulation time for television cameras on a CCD matrix in the electronic shutter mode (see Fig. 3 of the article by A. N. Kulikov).
Then for 1% “smear” we have the accumulation time of sensor (2) – 2 10-5 s (see the dependence “EXWAVEHAD CCD”), and the accumulation time of sensor (5) – 2 10-4 s (see the dependence “standard CCD”).
Taking into account the linearity of the exchange of an increase in illumination on the CCD photodetector target for a decrease in its exposure time, we obtain for the accumulation interval Tн = 2 10-4 s of sensor (5) the accompanying value of illumination at the input E = 500 lux.
Then the attenuation D of the input illumination, performed by the translucent mirror and the neutral filter of the beam splitter (4), will be 200 times.
If the attenuation D1 of the translucent mirror is 2 times (transmittance coefficient 0.5), then the required attenuation D2 of the neutral filter is 100 times.
In conclusion, it should be noted that the proposed technical solution for minimizing distortions of a television camera when working under conditions of light overload is more economically advantageous than the classic solution — using a lens with an automatic diaphragm (ARD) in the camera.
Calculations show that the cost of a camera with an ARD lens is three times higher than the cost of a camera with a constant diaphragm.
On the other hand, the disadvantage of cameras with ARD lenses is their low mechanical strength and, therefore, their use in conditions of strong vibrations.
The proposed solution avoids this drawback.