JPH04338445A - Electronic endoscope - Google Patents

Abstract

PURPOSE:To eliminate the need for a variations correction adjusting circuit of a plurality of CCUs in a processor by incorporating a scan timing generator corresponding to the type of image sensors and a color separation circuit or the like CCU (camera control unit) containing it. CONSTITUTION:Electronic endoscopes 20a-20c incorporate different CCUs 23a-23c separately and are provided with a CPU(control processing unit) 24 to control the operation of the CCUs. The CCU 23a obtains an external synchronous signal to generate a synchronous signal. The CCU also has a scan timing generator 30 to generate transfer clocks of CCDs 22a-22c, a drive circuit 31 which generates a reading clock signal by timing signals therefrom to drive a CCD transfer clock terminal and a color separation circuit 32 which processes signals (OS signals) read from the CCDs 22a-22c to separate a color signal as main section.

Description

[Detailed description of the invention]

[Object of the invention]

0002

[Industrial Field of Application] The present invention enables the signal processing device of an electronic endoscope to have multiple CCUs and The present invention relates to an electronic endoscope that does not require a CCD variation correction adjustment circuit.

0003

[Prior Art] CCD (Charge) is used as an image sensor.
There are several types of electronic endoscopes that have a Coupled Device (charge-coupled device) built into the tip, depending on the site of use, such as for the upper gastrointestinal tract and lower gastrointestinal tract. Different types are used depending on the direction and purpose of use, such as for detailed examinations, routine examinations, group examinations, etc.

[0004] This is because it is extremely difficult to use only one type of CCD with a fixed number of pixels and color separation method to handle examinations and medical examinations with different areas of use, viewing directions, and purposes of use. This is because defects such as not being able to obtain the required image quality are likely to occur during the inspection. Also, if you try to handle all inspections with just one type of CCD,
In order to meet diverse demands, the number of pixels and various functions of the CCD must be maximized, and as a result, the CCD becomes larger and the electronic endoscope becomes thicker, which may pose a problem when inserted into a body cavity. It also depends on the fact that there is.

[0005] Regarding CCDs, for example, endoscopes for routine examinations do not require much precision in image quality, and CCDs with thin tips are preferred based on subjects with narrow body cavities. It is desirable to use a CCD chip (made of semiconductor) 1 with a small external shape, even if the number of pixels is small (approximately 300,000 pixels; 603×493 pixels), as shown in (A). On the other hand, in endoscopes for detailed examinations, image quality is given top priority, so a CCD with a large number of pixels (approximately 400,000 pixels; 801 x 508 pixels) is used, for example, as shown in FIG. Preferably, tip 2 is used.

[0006] Also, CCD colors G (green) and R (red)
・The mosaic filter system in which B (blue) filters are arranged in a checkered pattern includes cyan (Cy), magenta (Mg),
Frequency of yellow (Ye) and green (G) color filters arranged in an array as shown in FIG. There are several types depending on the arrangement, such as an interleaved method and a color difference line sequential method.Furthermore, the color difference line sequential method has two types of filter colors: a primary color method and a primary color method as shown in FIG. There is a complementary color system shown in (same).

[0007] Primary color type filters have very good color reproducibility because the RGB primary color filters are attached to the CCD, but they only use about 1/3 of the amount of visible light, so they are not very bright. It is inferior in terms of sensitivity. On the other hand, since the complementary color filter utilizes two-thirds or more of the amount of visible light, its sensitivity is good. However, it is difficult to obtain materials that achieve desired properties, and there are also problems in color reproducibility.

[0008] As described above, there are many types of CCDs, even when considering only the number of pixels and color system. Color filters are often made of organic materials, but in this case, even if the color filter is of the same type, there will be variations from production lot to production lot.

A CCD sequentially transfers charges photoelectrically converted by photocells arranged vertically and horizontally through a vertical transfer path and a horizontal transfer path, and reads out an image by scanning the charges of the photocells row by row. Therefore, charge transfer from the photocell to the vertical transfer path, vertical charge transfer through the vertical transfer path (i.e., transfer to the horizontal transfer path), and horizontal charge transfer through the horizontal transfer path (i.e., the output ), as well as for electronic shutter operation that shortens charge accumulation time by intentionally discarding accumulated charges, etc., but this transfer clock depends on the type of CCD, especially the number of pixels, and the semiconductor manufacturing process. (This determines the thickness of the n-site and p-site in the semiconductor substrate [which affects the amount of charge storage]), different types are required.

[0010]Also, the signal output from the CCD (OS signal) includes color information created by the mosaic filter, which is divided into a luminance signal (Y signal) and a color (C) signal.
It must be separated into carriers and further color-separated into R, G, and B. This color separation method includes methods corresponding to the frequency interleave method and color difference line sequential method described above, but in the case of the frequency interleave method, R and B components are separated by addition and subtraction of two horizontal lines adjacent to each other. To separate. On the other hand, in the case of the color difference line sequential method, R and B components are obtained alternately for each horizontal line by subtracting two adjacent pixels, but the color components obtained are different between the primary color method and the complementary color method described above. Therefore, depending on the arrangement method of the color filters, the signal processing section for the OS signal, particularly the color separation processor, differs greatly.

[0011]

[Problems to be Solved by the Invention] Conventionally, electronic endoscopes have only housed circuits equivalent to the endoscope's own driver (drive circuit); No circuits were included; these were housed in the CCU (Camera Control Unit) within the signal processing device. The CCU includes a CCD-specific driver having a specific number of pixels and color scheme (color separation scheme, color filter scheme), a timing generator, and a processor for color processing.

[0012] Therefore, when using an electronic endoscope equipped with a CCD suitable for the nature of the examination/diagnosis, the number of pixels of each CCD, semiconductor process, color separation method, and color filter should be selected as described above. Since the systems and the like are different, a signal processing device equipped with only a single CCU cannot handle the different uses of these endoscopes. Therefore, as shown in FIG. 16, on the signal processing device 4 side, drivers 6a, 6b, 6c, which can be switched by switches 5a, 5b, 5c, and a timing generator (T .GEN) 7a, 7b, 7c, and multiple types (here, 3 types) containing processors 8a, 8b, 8c.
It could be used for an electronic endoscope 10 equipped with CCUs (species) and equipped with a CCD 9 corresponding to any one of these CCUs.

However, even if the CCDs are of the same type,
Note that due to variations between manufacturing lots, etc., variations occur in color signals. Therefore, as shown in FIG.
11, for accurate color reproduction, a color carrier variable gain amplifier 13 is installed after the color carrier separation circuit 12.
Further, an R/B separation circuit 14 and a matrix circuit 1 are provided.
An adjustment circuit 16 such as HUE or CROMA is required to correct the dispersion of the R-Y signal and B-Y signal that have passed through 5. The variable gain amplifier 13 and adjustment circuit 16 are C
It is activated by PU17. Note that in FIG. 17, the same reference numerals are used for the same components as in FIG. 16.

[0014] Therefore, in a conventional signal processing device for an electronic endoscope, first, a plurality of CCUs corresponding to each CCD are provided, which increases cost.
If you only use CCs other than those that correspond to that CCD,
U will be wasted. Furthermore, even if a new CCD other than that envisaged at the time of manufacturing the signal processing device appears, it cannot be applied to an electronic endoscope equipped with that CCD. Furthermore, the above-mentioned adjustment circuit for correcting CCD variations has the drawback that it reduces the S/N ratio, increases the circuit scale and increases costs, and cannot completely correct the variations.

The present invention has been made in view of the above circumstances, and even when electronic endoscopes equipped with CCDs with different numbers of pixels and color systems are used depending on the type of examination/diagnosis,
It is an object of the present invention to provide an electronic endoscope that does not require a processor to include a plurality of CCUs and CCD variation correction adjustment circuits. [Structure of the invention]

[0016]

[Means for Solving the Problems] In order to solve the above problems, the present invention provides an electronic endoscope with a built-in image sensor.
The present invention provides an electronic endoscope further incorporating a camera control unit including a scan timing generator corresponding to the type of image sensor and a color separation circuit or a color processing circuit.

[0017]

[Operation] The present invention provides an electronic endoscope that incorporates different types of image pickup devices such as CCD and MOS depending on the type of examination and medical examination.
The camera control unit (CCU) is affected by the specifications depending on the color system (color method), and has to use its own unique one. (In case of multi-line readout type CCD). Therefore, by using a pair of an image sensor and a corresponding CCU, it is possible to adjust and standardize the output according to the unique characteristics of the built-in image sensor within the electronic endoscope, and to correct variations in color reproduction. ,
It becomes possible to standardize the output from the electronic endoscope to the TV system or the like while correcting distortions in color reproducibility. Furthermore, the optimum filter constant for the built-in CCD can be obtained, and the resolution of the electronic endoscope can also be improved.

[0018] In a signal processing device that processes signals from an electronic endoscope as an interface with the outside, it is necessary to provide a plurality of CCUs corresponding to each assumed CCD and a switching circuit therefor. This reduces the size, increases economic efficiency, and improves the S/N ratio. Furthermore, even if a new image sensor appears in the future, the existing signal processing device can be used as is.

[0019]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to FIGS. 1 to 13.

FIG. 1 shows electronic endoscopes 20a and 20a each incorporating CCDs having different numbers of pixels according to a first embodiment of the present invention.
20b, 20c, and a configuration diagram of a signal processing device 21 connectable to these electronic endoscopes.

The CCDs 22a, 22b, 22c built in the electronic endoscopes 20a, 20b, 20c each have 12
First, the electronic endoscope 20a has 250,000, 250,000, and 400,000 pixels, so the diameter of the electronic endoscope 20a can be reduced by taking advantage of the small external size of the CCD 22a, which has a small number of pixels. Turn. Further, the electronic endoscope 20b is C
Since the CD22b has a large number of pixels, its outer diameter is also large, but it can be used for detailed inspections. The electronic endoscope 20c is intended to be equipped with a 400,000 pixel CCD when semiconductor processes advance and a CCD with 400,000 pixels appears in the future.

[0022]The above three types of CCDs 22a, 2
2b and 22c use the same complementary color line sequential color system, but the number of pixels is different, so the electronic endoscopes 20a and 2
0b and 20c are different CCUs 23a and 23b, respectively.
, 23c, and further includes a CPU (central processing unit) 24 that controls the operation of these CCUs.

Therefore, when using the electronic endoscope of this embodiment, the signal processing device 21 uses a light source 25 for a light (not shown) attached to the tip of the electronic endoscope, and A pump 26 used to pump air and water in and out, and a display device 27 that processes the output of the electronic endoscope.
a process circuit 28 that sends data to the
It is only necessary to have the PU29, and there is no need to equip it with the CCU function.

Next, FIG. 2 shows the configuration of the CCU 23a as an example. The other CCUs 23b and 23c also have similar configurations.

The CCU 23a receives an external synchronization signal, emits the synchronization signal, and also uses a scan timing generator 30 that generates a CCD transfer clock, and generates a read clock signal using the timing signal from the scan timing generator 30, and outputs a read clock signal to the CCD transfer clock terminal. The main parts are a drive circuit 31 that drives the CCD, and a color separation circuit 32 that processes the signal read out from the CCD (OS signal) and separates color signals. Among these, the scan timing generator 30 is related to the number of pixels of the CCD.
It is.

On the other hand, the OS signal from the CCD 22a passes through a buffer amplifier 33 to a CDS (Correlated
A double sampling (correlated double sampling) circuit 34 is reached, and a CDS circuit 36 samples the field-through period and the optical signal output period based on the CDS pulse from the scan timing generator 30 to obtain the output of each pixel. This output is then sent to the color separation circuit 32 via the amplifier 35. In the color separation circuit 32, a scan timing generator 30
Based on the color separation pulses and clamp pulses from , every other pixel is sampled to extract color signals (Y, RY, B-Y).

FIGS. 3A and 3B show the CCD 22, respectively.
The pixel arrays of a (120,000 pixels) and CCD 22b (250,000 pixels) are shown. Among these pixels, effective pixels 36a, 3
6b has a color filter pasted thereon, and can output a signal effective as an image (effective signal). On the other hand, optical black (black reference pixels) 37a and 37b are
On top of that is an aluminum light-shielding film that outputs a signal when no light enters. This signal is used as a reference for the useful signal.

FIG. 4 shows how to read out the CCD charges in the horizontal direction.
This is done using the H1 and H2 clocks as shown in FIG. In other words, during the period when H1 = H (high) and H2 = L (low),
Charge is being transferred from the vertical charge transfer path to the horizontal charge transfer path, and during this time, horizontal reading of the CCD is stopped. Then, when H1=L and H2=H, the first pixel in the horizontal direction is read out as an OS signal, and thereafter each pixel is sequentially read out.

As mentioned above, this OS signal is processed by the CDS circuit 36 by the CP1 pulse and the CP2 pulse, respectively.
The output of each pixel is obtained by sampling during the field-through period and the optical signal output period. Note that the RS pulse in the figure is a reset pulse necessary to convert the charge generated in the photoelectric cell into voltage.

By the way, the television monitor (display device 27)
Since the length of one horizontal line is fixed, if the number of pixels in the horizontal direction of the CCDs differs, the horizontal readout frequency must be changed for each CCD. Furthermore, since the horizontal/vertical size ratio of the screen of a TV monitor is fixed, in order to reproduce the image formed on the CCD surface on the TV monitor without distortion, it is necessary to A horizontal clock is required.

CCD 22a shown in FIGS. 3(A) and 3(B)
, 22b, the horizontal readout frequencies are 9.53 MHz and 12.28 MHz, respectively, but of these, 12.28 MHz is not 1/N (N is an integer) of the horizontal frequency of the TV, so phase matching is required. A circuit, etc. is required, and a 120,000 pixel CCD22 is installed on the signal processing device side.
It is difficult to provide a common scan timing generator for the CCD 23b and the 250,000 pixel CCD 23b. If the signal processing device is to be equipped with a scan timing generator, the only option is to provide a separate scan timing generator on the assumption that the signal processing device will be larger.

However, in this embodiment, the electronic endoscopes 20a and 20b are CCUs equipped with scan timing generators corresponding to the CCDs 22a and 22b, respectively.
Since 23a and 23b are built-in, there is no difficulty in sharing the scan timing generator.

In addition, the color separation circuit converts the signal from the CDS circuit into the SP1 pulse (for odd-numbered pixels) and SP1 pulse shown in FIG.
Since color signals are extracted by sampling every other pixel using the P2 pulse (for even-numbered pixels), it is necessary to adjust the frequency of these color separation pulses in the same way according to the difference in the number of pixels. In this embodiment, the electronic endoscope 2
0a and 20b incorporate the CUs 23a and 23b corresponding to the CCDs 22a and 22b, respectively, so that there are no problems related to sharing the CCUs or an increase in the size of the signal processing device.

Furthermore, inside the color separation circuit, the DC level is determined based on the optical black signal mentioned above. That is, in the clamp circuit shown in FIG. 5, the DC component is once cut from the OBCLP (optical black clamp pulse) by the capacitor 38, and as shown in FIG. The signal level is clamped to the DC level during the corresponding clamp period (optical black signal period).

Therefore, as shown in FIGS. 3A and 3B, the position of the optical black differs between CCDs with different numbers of pixels, so for example, as shown in FIG.
For the 250,000 pixel CCD and the 120,000 pixel CCD, different optical blank clamp pulses (OBCLP1 and OBC
LP2) and change the position of the clamp pulse.

However, in this embodiment, since the electronic endoscope incorporates a CCU containing a scan timing generator that emits clamp pulses corresponding to each CCD, the horizontal readout frequency and color separation pulse The difficult problem of adjusting the frequency of the clamp pulse as well as the frequency of the clamp pulse does not arise.

Now, FIG. 8 shows an example of a color separation circuit. CDS
The OS signal output from the circuit 40 first passes through the gain control amplifier GCA1. This is to correct variations in sensitivity of the CCD, and the gain is changed by controlling the control voltage CV1. Its output is a low pass filter LP
It passes through FF1 and becomes YH (luminance signal).

Gain control amplifier GCA2, GC
A3 is for correcting variations in the position of the CCD color filter, and control voltage CV2. CV3
You can change the gain by Similarly, gain control amplifiers GCA4 and GCA5 also operate at control voltage CV4. Change the gain with CV5, R and B
Correct the variation between the two.

As described above, various variations must be adjusted even for CCDs of the same color system, but the electronic endoscope of this embodiment has a one-to-one correspondence between the CCD and the color separation circuit, so it is completely Adjustment is possible. In the electronic endoscope of this embodiment, gain control amplifiers GCA1 to GCA5 that can be controlled by voltage are used in accordance with the waterproof structure.

The above-mentioned control voltages CV1 to CV5 are generated by an analog switch AS that operates two resistance-divided reference voltages SV1 and SV2 using an analog switch control signal in the control voltage generation circuit shown in FIG.
It can be created by switching. The analog switch AS is controlled by a CPU within the electronic endoscope, and the CPU can communicate with a signal processing device via electrical contacts of the electronic endoscope. At this time, the analog switch AS can also be automatically adjusted by combining it with means for checking whether adjustment has been made.

In the YH signal of FIG. 8, as shown in FIG. 10, G+Cy and Mg+Ye appear for each pixel in the m line, and Mg+Cy and G+Ye appear in the m+1 line. Therefore, due to SP1 and SP2, on the m line, G+Cy is applied to the SH1 output, Mg+Ye is applied to the SH2 output, and on the m+1 line, Mg+Cy is applied to the SH1 output, and G+Y is applied to the SH2 output.
Separate as e. Then, by subtracting at each line, the m line output is (Mg+Y
e)-(G+Cy)=2R-G, for m+1 line output (G+Ye)-(Mg+Cy)=-(2B-G)
can be obtained (after all, Mg=R+B, Ye=R+
G, Cy=B+G).

Next, a second embodiment of the present invention will be explained. The second embodiment is an electronic endoscope incorporating a CCD manufactured by a different manufacturing process.

FIGS. 11 and 12 show examples of two CCD drive circuits. In other words, even if the number of pixels is the same,
The voltage of the CCD transfer clock varies depending on the manufacturing process. In the drive circuit of FIG. 11, the V direction is -
There are three values: 8V, 0V, and +8V. That is, the V clock takes three values, and field shift (transfer of charge from the photoelectric cell to the vertical transfer path) is performed at +18V, and line shift is performed between 0V and -8V. Also, in the H direction, there are 0V and 9V clocks (that is, the amplitude of the H clock is 9V), and the OS
OFD (Over Fl) determines the saturation voltage of the signal.
ow drain), but here a DC voltage is applied to the OFD.

On the other hand, in the drive circuit of FIG.
The clock has three values (-9V, 0V, +9V), but two clocks (XSG1 and XS
G3) is required, and the horizontal system is H1 and H2 for CCD.
0 to 5V, and finally 9V of LH1 and 9V of RS are required. The drive circuit is of a type in which the 24V pulse for the electronic shutter is replaced by an OFD. As described above, in this embodiment, even if the voltage of the transfer clock is completely different due to a difference in the semiconductor manufacturing process of the CCD, this is handled within the electronic endoscope, so there is no problem in dealing with it on the signal processing device side.

Finally, the standard interface between the electronic endoscope and the signal processing device of the present invention will be explained.

[0046] Since an electronic endoscope can be regarded as a television camera, a normal NTSC system interface is also possible, but the circuit and CCD inside the electronic endoscope must be used to maximize performance according to its scale. In the sense of letting
A component type such as Y signal and R-Y, B-Y, or R, G, B, etc. is preferable. Also, if it is necessary to synchronize several electronic endoscopes, an external synchronization signal is required.

[0047] In a normal television camera, the output after γ correction is used, but in the case of an electronic endoscope, signal processing etc. are expected, so it is often desirable to make γ variable. For this reason, it is convenient to use a linear output of γ=1 for the electronic endoscope output (it may be multiplied by γ).

Furthermore, as shown in FIG. 13, if the communication protocol of the CPU within the electronic endoscope is determined, any electronic endoscope can be connected to one signal processing device.

Note that the electronic endoscope of the present invention can also use a three-wire readout type CCD that does not require color separation (in this case, a color processing circuit is built in instead of the color separation circuit). It is also possible to use an image sensor other than a CCD, such as a MOS type.

[0050]

[Effects of the Invention] As explained above, according to the electronic endoscope of the present invention, it is possible to maximize the performance of the image sensor, use a common external interface, and save money. (There is no need for multiple signal processing devices with built-in CCUs compatible with electronic endoscopes, or signal processing devices with built-in multiple CCUs and switching circuits. Also, when new image sensors appear in the future, it will be easy to install the signal processing devices. (no switching is required) and the S/N ratio can be improved by eliminating the switching.

[Brief explanation of drawings]

FIG. 1 is a configuration diagram of an electronic endoscope and a signal processing device according to a first embodiment of the present invention.

[Fig. 2] CC installed in the electronic endoscope according to the above embodiment
Configuration diagram of U.

FIGS. 3A and 3B are pixel layout diagrams of CCDs with 120,000 pixels and 250,000 pixels, respectively.

FIG. 4 is a diagram showing the read timing of a CCD.

FIG. 5 is a clamp circuit diagram.

FIG. 6 is a diagram showing an optical black clamp pulse.

FIG. 7 is a diagram showing optical black clamp pulses for a 250,000 pixel CCD and a 120,000 pixel CCD.

FIG. 8 is a color separation circuit diagram.

FIG. 9 is a control voltage generation circuit diagram.

FIG. 10 is a diagram showing color output by line of the CCD.

FIG. 11 is a diagram showing an example of a CCD drive circuit built into an electronic endoscope according to a second embodiment of the present invention.

FIG. 12 is a diagram showing another example of a CCD drive circuit built into the electronic endoscope according to the second embodiment of the present invention.

FIG. 13: C in an electronic endoscope according to another embodiment of the present invention.
A diagram showing a PU communication protocol.

[Figure 14] (A) and (B) are 300,000 pixels and 40,000 pixels, respectively.
FIG. 3 is a diagram showing the pixel size of a million-pixel CCD.

FIGS. 15A and 15B are diagrams showing the arrangement of color filters in a frequency interleave method and a complementary color difference line sequential method, respectively.

FIG. 16 is a configuration diagram of a conventional electronic endoscope and signal processing device.

FIG. 17 is a configuration diagram of a color separation circuit built into a conventional signal processing device.

[Explanation of symbols]

20a, 20b, 20c electronic endoscope 22a, 22b
, 22c CCD 23a, 23b, 23c CCU 24 CPU

Claims (3)

[Claims] 1. An electronic endoscope with a built-in image sensor, further comprising a camera control unit including a scan timing generator corresponding to the type of the image sensor and a color separation circuit or a color processing circuit. . 2. The electronic endoscope according to claim 1, further comprising means for controlling the camera control unit in response to a remote control signal from the outside. 3. The electronic endoscope according to claim 1, further comprising an external synchronization circuit that synchronizes the scan timing generator with an external signal.