Solid state television camera

© Start of patent duration: 1982 01 15 Longest possible duration:

© Issued on: 1982 10 11 © inventor:

© dependence:

© Pamphlets considered to delineate the prior art:

Nr.3S8335

The invention relates to a solid-state television camera, consisting of a solid state image sensor on which the image of an object is projected, and which generates a video output signal, a sample and hold circuit, to which the video output signal is passed, a level detector circuit, to which also Video output signal is passed, so that the level of the video signal is detected, one, at the output of the level detector circuit to a closed, gate circuit, to which a scanning signal is supplied, and which is connected to the Ahtast- and holding circuit, so that the operation of the sample and hold circuit is controlled by the output signal of the gate circuit.

There are already known television cameras which use charge-coupled devices as solid-state image sensors.

The solid-state image sensors or charge-coupled devices used in such television cameras are shown in principle in Fig.l and designated -1-. The charge coupled device (s) -1- consists of a photosensitive surface -1A- containing a plurality of picture elements -4- arranged on the surface of a semiconductor substrate in the vertical and horizontal directions and on which the object is projected. A buffer area -1B-, which is similar to the photosensitive area -1A-, except that it is optically shielded, serves as a charge storage for corresponding light information of an object. A register -IC- is used to read the charge carriers during each horizontal scanning period of the buffer area -1B-. In Fig.l from the read-out register -IC- outgoing output terminal -5- and the channel separation devices -6- is shown.

In the production of the charge coupled device, it is very difficult to make the semiconductor crystal uniform over a predetermined area. Local crystal defects can occur and it is possible that electrical charges for thermal causes cause such defects. As a result, noise will occur in the output signal when an object is imaged with a charge coupled device camera which is projected onto the charge coupled device. The noise will occur in unlit areas where the current is unusually high. In this case, the noise reaches the white level so that it appears in the reproduction and it becomes unusable.

In order to eliminate this noise, it is common to use a defect compensation circuit and to control a sample and hold circuit -7- to which the readout or video signal of the charge coupled device -1- of Fig. 2 passes. Suppose a trapezoidal signal S_A according to Fig.3A comes from the charge-coupled device via the output -5- to an amplifier -13-. Since the level of a noise signal corresponding to the crystal defects is higher than the white level -L ^ -, the noise signal S ^ in Fig. 3A can be represented as a dotted line. The noise elimination circuit -10- includes a level detector -8- for the video signal. The response value of the level detector -8- is eg is set to a desired value L ^, which is higher than the white level L ^, as shown in Fig. 3A. From the output of the level detector -8-, a signal Sp is obtained when the level of the supplied signal exceeds the value L ^. The signal Sp is fed to a gate -9-, which is fed to it a sampling signal S_G controls.

If the gate -9- is blocked by the output signal Sp from the level detector -8-, then the key signal S_G corresponding signal S_G^ (see Fig.3B) away from the gate -9- exit (the last output of the gate -9- is signal S)_glas shown in Fig. 3D). As a result, a signal Sp will appear at the output of the sample and hold circuit -7- of Fig. 3E.

When the above-described defect compensation by the gate -9- is not carried out, an output signal Sp is obtained based on the corresponding sampling signal S_gl from the hour-and-hold circuit -7- of Fig. 3C, with the result that the reproduced picture is disturbed in quality.

Since, in the above defect compensation circuit -10-, the output -S p - of the signal level detector -8- is applied to the gate -9- for controlling the sampling signal S_G is fed, there is a delay of the detection process due to the response time τ of the detector circuit -8-.

- 3 No. 368335

For this circumstance, in the noise elimination circuit -10- in front of the sample-and-hold circuit -7-, there is provided a delay circuit or a delay line -11- whose delay time is equal to the response time τ. This delay line -11- delays the video signal S ^ by a predetermined time and supplies the delayed video signal to the sample and hold circuit -7-.

As described above, a delay line -11- is necessary in the prior art noise eliminator circuit -7-, but the waveform of the video signal is distorted by the delay line -11-, and hence the fidelity of these known solid-state cameras is degraded.

In addition, the delay line is expensive and the cameras are thereby costly to manufacture.

The object of the invention is to provide a solid-state television camera, which is free from the defects and disadvantages described above.

Another object of the invention is to provide a solid-state television camera which does not require a delay circuit or line, so that there are no disadvantages caused by the use of delay lines of known type.

Another object of the invention is to provide a solid-state television camera which does not use a delay line but which still causes a defect compensation.

This is achieved in a camera of the type mentioned in the present invention, that a logic circuit in the form of a D-type flip-flop is provided with a data input terminal, a reference signal input terminal, and an output terminal, wherein the level of the output signal of the level detector is supplied to the data input terminal to be stored at each of the periodically occurring reference signals which are supplied to the reference signal input terminal, and in that a delay circuit is supplied with an output signal of the time shift logic circuit (τ) necessary for the response of the level detector circuit and for supplying a time-shifted signal to the gate circuit, the level detector circuit having a detection level for the video signal occurring at a reference time, which differs from the time of occurrence of the scanning signal by a predetermined period of time.

The other objects, features and advantages of the invention will be described below with reference to the Fig.l to 13. 2 shows a block diagram of a solid-state camera of the known type, FIGS. 3A to 3E show the signal curve for describing the mode of operation of the camera according to FIG. 2, Fig. 4 is a block diagram of a solid-state camera according to the invention, Fig. 5 and Fig. 11 show the circuit diagram of a part of Fig. 4, Figures 6 to 10 and 12 show the waveform for explaining the operation of the examples of Figures 4, 5 and 11 and Figure 13 is a block diagram of another example according to the invention.

With reference to Figure 4, an embodiment of the invention will be described below. In this example of a color television camera, three charge-coupled devices -1, 2 and 3 are used which are similar to the charge-coupled devices shown in FIG.

In the solid-state television camera of the invention of Figure 4, the image of an object is projected over an optical system consisting of a lens system -15-, and a pair of half-mirrors -16a, 16b- and a pair of mirrors -16c and 16d- exist, which are arranged in front of the corresponding charge-coupled devices. In this example, color filters are -17R, 17G and 17B- which are monochromatic light, eg Red, green and blue, between the mirror -16c and the charge-coupled device -1-, between the half-mirrors -16a, 16b- and the charge-coupled device -2- and between the mirror -16d and charge-coupled device -3- arranged so that color-separated images are projected on the charge-coupled devices -1 to 3-.

Because of the spatial arrangement of the three charge-coupled devices - 1 to 3, it is possible that when the spacing of the picture elements -4- from each charge-coupled device (see Fig. 4) in the horizontal direction 1_R which is proportional to the sampling period τ _R is

- 4 Nr.368335 the respective charge-coupled devices -1 to 3- are shifted by 1/3 lg in the horizontal direction in succession in mutual relation. Therefore, the image of the object -12- is projected onto the charge coupled devices -1 to 3 in such a manner that the image of the object -12- is shifted by 1/3 lg to the adjacent charge coupled device.

The reason why this above relationship has been chosen is described in U.S. Patent No. 3,975,760, and accomplishes an improvement in resolution without having to increase the number of pixels in the horizontal scanning direction. By this relationship, sideband components superimposed on the luminance component are effectively inhibited by the combination of the video signals derived from the respective charge coupled devices -1 to 3, so that the picture quality does not suffer from noise.

The video signals Sg, Sg and Sg are derived from the respective charge-coupled devices -1 to 3, and via a defect compensation circuit -10- (specifically, -10R, 10G and 10B-) which suppresses the noise caused by defects of the semiconductor body, and which will be described later, to a matrix 18- whose output signals are supplied to an encoder -19-. Thus, a composite color video signal, for example in the NTSC system, is obtained at the output 190a of the coder 19.

An embodiment of each of the defect compensation circuits -10- (or -10R, 10G and 10B-) will be described with reference to FIG. Each of the defect compensation circuits -10- includes a level detector -8- and a control circuit -9- for the sample signal Sg. The level detector -8- includes an operational amplifier whose (+) input terminal is connected to a reference voltage source -21- which sets the response level -Lg-; the other (-) input terminal is connected to the input terminal -10a to which the video signal S ^ is supplied from the charge coupled devices -1 to 3-. The setting of the response level -Lg- will be described later.

The inverted output of the level detector -8- is applied to the D input terminal of a D-type flip-flop -9A-, which forms another component of the control circuit -9-. The flip-flop -9A- is supplied at its T input terminal, a reference signal.

As already described in detail, a time delay τ is necessary for the response of the level detector -8-, which delay time is set by a replacement delay circuit -9C-. In this case, the actual delay resulting from resetting the flip-flop -9A- and the delay circuit -9C- is shown for explanation only. The output signal Sq of the flip-flop -9A- is delayed by the delay line -9C- as a signal Sq, an input terminal of an AND circuit -9B- supplied to the second input terminal, the sampling signal Sg is supplied. Thus, the sampling signal Sg is controlled by the output signal Sq of the delay line -9C-.

Before describing the operation of the defect compensation circuit -10-, the derived output or video signal S ^ (or Sg, Sg, Sg) will be explained with reference to FIG. In Fig. 6, Τθ represents the period of time necessary to read out the carriers included in a pixel -4-, Tg a pre-charge period, and Tg a period of time required to transfer the carriers. Ideally, the transmission time is zero, as shown in Figure 6 as a dashed line. Practically, the transmission time is different from zero, but has a defined value, which depends on the characteristic data of the charge-coupled device. Thus, the carrier transfer signal receives a trapezoidal course, which is shown in Figure 6 as a full line. The time duration Τθ from the beginning of the charge carrier transfer to the end of the same thus becomes the actual charge carrier transfer duration. In the example described, the duration Τθ is referred to as carrier transmission time (time).

The determination of the response level -Lp-of the level detector -8- of the invention will be explained below with reference to FIGS. 7A and 7B. Referring to Fig. 7A, a transmission waveform of an output signal Sg (broken line in Fig. 7A) is assumed to be a reference output whose maximum value at the end of transmission is higher than the white level -Lyj but lower than the level.^Lpj ~ ^^it Noise signal is.

Nr.368335

If the response time of the level detector -8- is fixed with τ, with the task the sampling signal S_G to control with the detected output signal, the time period from the time of arrival of the scanning signal must be at least τ. Will the time at which the sample arrives as t₃ Assuming, as shown in Fig. 7B, the time t₀ before the time t₃ selected as the reference time. A level which is higher than the signal (which, for example, has just the value in the size of the white level -L ^ -), is at the time t "of the reference signal S_ß as the response level "Lp-- set in the meaning of the invention.

The reference signal (pulse) S ^ is supplied to the flip-flop -9A- and the frequency of the pulse S according to the invention_M is equal to that of the sampling signal S_G chosen to be equal, wherein the pulse S ^ to the sampling signal S_G has a phase or time difference τ. That is, the time at which the reference signal S ^ arrives identical to the reference time t₀ is.

With reference to Figs. 5 and 8 and the above assumptions, the operation of the defect compensation circuit 10 of the invention will be described below.

First, the video signal S_A> which contains the noise signal according to Fig. 8A. If the video signal is fed to the level detector -8-, then an output signal according to Fig.SC is generated. In the example according to FIG. 5, the inverted signal according to FIG. 8D is obtained and fed to the D input terminal of the flip-flop -9A-. Since the flip-flop -9A- receives the above-described reference signal at its T input terminal, occurs at the Q output terminal of the flip-flop -9A- the signal Sq, which the level of the inverted signal S ^ at the time of arrival of the reference signal S ^, as shown in Fig. 8E.

The AND circuit -9B-, which receives the signal Sq delayed by τ (FIG. 8F) and the sampling signal S_G (Fig. 8G), generates an output signal S _G, from which the sampling signal S corresponding to the noise duration_G^ was removed, as shown in FigH. If the output signal S_gl of the TDM circuit -9B- is supplied to the sampling and half circuit -7- as a sampling signal, no sampling is carried out during the noise generation period, and the output value is maintained until the next video signal is received. The output 3θ (Fig. 81) thus held constant by the sample-and-hold circuit -7- is supplied to the output terminal -10b-.

If the scanning operation is carried out without suppression of the scanning signal during the noise generation period, an output signal S_QD obtained, which is in no relation to the object -12- (Fig.81, dashed line).

Since the response level -Lp- is higher than the level of the video signal S as described above_Aat time t₀ has been selected, and the reference signal and the detected signal S_Q the D-type flip-flop -9A- is supplied, which then outputs the signal Sq, which the delay line -9C- supplied the signal Sq 'results, it can be absolutely determined whether a noise signal is present or not, and the noise can be suppressed inevitably without using a delay line -11- of the known type.

The defect compensation process can be carried out according to the invention similar to the known type, but without the use of the delay line -11-, so that the distortion of the held output signal βθ caused by the delay circuit -11-, are forcibly eliminated and thus the manufacturing cost of the camera can be kept low.

When a noise elimination circuit -10- of Fig. 5 is executed, it operates satisfactorily when used with a charge coupled device with a short carrier transfer time.

In general, there are various television cameras with charge-coupled devices which have either a small (see Fig. 9) or a large (see Fig. 10) carrier transmission time. For cameras with a short carrier transmission time, there are no problems. In the case of the cameras with large carrier transmission time, if the response level -L ^ - is selected to be higher than the white level -L ^ -, similar to the known manner, the time at which the noise signal is detected in the time period between the times t₀ and t₃ fall.

Therefore, the sampling signal S_G can not be controlled when a noise signal is detected and no noise elimination is achieved. In this case it is necessary, so

Nr.368335

6 similarly as in the known embodiments, to shift the sampling signal Sg in advance in advance and also to delay the video signal to be sampled by the delay circuit 11.

However, if the reference time t "according to the invention is selected so that the response level is below the white level, then the noise elimination process may be performed forcibly by using the reference signal Sp and the D-type flip-flop -9A-.

For example, in the case of small transmission times, the response level -Lp- is set higher than the white level -L ^ -. Fig. 11 is a circuit diagram and Fig. 12 is a signal waveform for explaining the circuit of Fig. 11; As can be seen from Figure 11, this circuit does not contain a flip-flop.

If the video signal (Fig.l2A) is supplied to the level detector -8-, then the detected signal S is at the output_Q (Fig.l2B) generated. In this example, the inverse output signal Sp (Fig.120) is applied to the delay line -9C-, and then delayed by the time τ as the output signal Sp '(Fig.l2D), the output signal Sp' and the input signal S_G (Fig. 12E) are supplied to the AND circuit -9B- which generates the output signal Sg 'which corresponds to the noise duration while it does not shine, as shown in Fig. 12F. The output signal Sg * of the AND circuit -9B- is supplied to the sample and hold circuit -7- as a sampling signal. Therefore, no sampling is performed during the noise generation period, and the signal obtained from the previous frame period is used as an output signal. during the noise generation period. The thus held by the sample and hold circuit -7- constant output signal S_Q (Fig. 12G) is taken from the output terminal -10h-- of the sample and hold circuit -7-.

According to the embodiment of the invention of Figure 4, the levels of the video signals Sp, S_G and Sg are detected separately from the charge coupled devices -1 to 3, and the noise squelching operation is performed separately for each video signal. It is therefore also possible that the noise detection and control for the respective signals are performed by a common circuit.

Fig. 13 shows a block diagram of another embodiment according to the invention, wherein the same reference numerals for the individual functional elements as in Fig.4 are used.

The common level detector -8- of Fig. 13, the video signals S_D , S "and S_Dnb D supplied from the charge-coupled devices -1 to 3-. The detected output signal Sp of the circuit -8- is led to the common control circuit -9-, to which also the sampling signal Sg is supplied. The output S g 'of the control control circuit -9- is supplied in common to the sample and hold circuits -7R, 7G and 7B- which are inserted in the signal transmission path of the video signals Sg, Sp and Sg.

The embodiment according to the invention of FIG. 13 is simple in construction and gives the same results as in the example of FIG.