JPS5970086A - Image correcting method and device - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method and apparatus for image correction, and in particular to a method and apparatus for image correction in a television camera.

A television camera has an optical system that directs incident light to an image pickup tube from which an electronic signal is obtained. In order to obtain a preview image of satisfactory quality, several phenomena must be corrected. The number of phenomena that require correction is considerably greater in color telephoto cameras.

Conventionally, such correction has been performed by manually adjusting related controls in consideration of the effect on the final image.

Manual adjustment has low reliability ↑1, is time consuming, and is generally unsatisfactory.

Automatic adjustment is used in which a correction signal is applied to the circuitry of the image tube or to the electronic signal of the image. Do these signals have a certain shape? It depends on the particular phenomenon being corrected.

Although this technique uses complex correction signals, the curve of A is determined primarily on theoretical grounds.

The applied correction is calculated for a standard image pickup tube, and if the image pickup tube deviates from the standard, sufficient correction cannot be made.

The present invention solves the above-mentioned volcano problem.

The present invention firstly provides a monitoring means for monitoring parameters from each pixel, and a means for calculating a correction signal from the parameter values.
Calculate Ill! ! The present invention provides an apparatus for correcting defects in an image that affects a plurality of pixels, which includes a control means for correcting image defects, and a control means for applying a correction signal.

Second, the present invention measures the parameter values in each of the plurality of pixels, generates a correction signal from the parameter value, and applies the correction signal to correct the image defect. A method is provided for correcting existing image defects.

One of the corrections mentioned above relates to focus correction.

The image carried by the image carrier tube is focused on a single quadrant by generating an axial magnetic field within the image pickup tube in cooperation with an electrostatic field. The magnetic field is created by passing current through magnetic coils called deflectors, and the electrostatic field is created by applying a voltage to the focusing electrodes.

Due to a well-known phenomenon, the electron image produced by the image tube corresponds to a curved object surface, so that the image tube image of a flat object is not uniformly focused. It is known that the lack of uniformity of focusing can be corrected by applying a correction signal to the electrostatic field of the image pickup tube (1). A predetermined unidirectional correction signal is applied to each of the four elements.

As a result, the focus correction is often inferior to the best focus correction because the combination of each picture tube and deflection yoke causes a slight deviation in the + and two focal planes of the electron beam sleeve 11.

A preferred embodiment of the present invention provides for the construction of an electro-optical tube, such as monitoring means for measuring the best focus parameter fiQ at each of a plurality of pixels, and processing means for measuring a correction signal from measurements 1+i'i. A focus correction device for an electron-optical tube is provided, which comprises a control means for correcting the focusing of an image and a control means for applying a correction signal.

Embodiments of the present invention include the steps of measuring the value of the best focus parameter at each of a plurality of pixels, calculating a correction signal from the measured values, and correcting the focusing of the image formed by the electron-optical tube. A focus correction method for an electron-optical tube is provided, which comprises the step of applying a correction signal so as to

For each pixel, the maximum focus parameter value intensity is 1'
It is preferable to test a range of focusing voltages to confirm the The voltage identified represents the best focus of the pixel.

Preferably, the pixel signal is passed through a bandpass filter so that the high frequency energy originating from the peak of the pixel is used to determine the maximum peak amplitude.

If the camera has several image tubes, the tubes need to be spatially positioned according to JL. For example, a color television busy camera requires three color image tubes to have a common spatial positioning. This positioning must be the best so that the final image is not blurred. A similar need exists for color television picture tubes in that the electron guns must be positioned relative to each other.

Another embodiment of the invention provides a reference generator for generating a plurality of reference pixels, a monitor means for monitoring image signals of the image tube originating from the reference pixels, and a combination of the pixels and the reference pixels. a comparator means for each pixel; moving the pixel in a first direction to a position with a relatively large misalignment; a moving means for moving the pixel in the opposite direction to the first direction to a position l<l; and a 4 error signal representing the best spatial positioning position 16 from the movement by the moving means and the initial error in the spatial positioning of the pixel. A relatively small error correction device of 63 tJ is provided in the spatial positioning of an electron-optical tube, comprising processing means for calculating .

Corresponding embodiments of the present invention are more than 4 i%! A step of generating one pixel, a step of monitoring the pixel signal of the image pickup tube originating from the pixel, a step of comparing the pixel signal with the gL i9- pixel signal, and a step of comparing each pixel with the pixel signal. moving a pixel in a first direction to a position with a relatively large misregistration in a first direction, and moving pixels in a direction opposite to the first direction to a position with a relatively large misregistration; Provides a method for determining the relatively small spatial positioning of an electron-optical tube consisting of the step of measuring the spatial positioning position and the initial error of the position of the pixel. Ru.

In other configurations, no reverse movement is used and the movement is performed continuously in small steps until the error is minimized.

The invention also provides a device for spatial positioning correction as described above, and processing means for calculating a correction signal from the measured pixel error signal.
1. An apparatus for compensating for spatial positional deviation of an electron-optical tube is provided, which comprises a processing means for applying a correction signal to compensate for spatial positional deviation of the image pickup tube in an image produced by the image tube.

Similarly, the present invention comprises the steps of the method for correcting spatial position deviations as described above, the step of calculating a correction signal from the measured pixel error signal, and the steps of calculating the correction signal from the measured pixel error signal, A method for compensating for spatial misalignment of an electron-optical tube is provided, comprising applying a correction signal to compensate for misalignment.

Image blurring produced by electro-optical tubes and its correction is well known. Image blurring also occurs in solid-state imaging devices. The problem of blur is particularly pronounced in images produced by devices such as color television broadcast cameras. Each of the three color image tubes produces a slightly different blur so the final image has color shading. The compensation signal for each image pickup tube is conventionally compensated for by converting the image pickup tube output into a correction signal.The correction signal usually has a simple parabolic waveform, and the result of the correction is Although satisfactory, this conventional method was difficult and required a skilled technician to perform.

Another embodiment of the present invention includes a monitoring means for monitoring (8) from each of a plurality of pixels, a processing means for calculating a correction signal from the monitored signal, and a processing means for calculating a correction signal to correct image distortion. An apparatus for correcting image blur is provided, which comprises an application control means.

Similarly, the present invention includes means for monitoring signals from each of a plurality of pixels, a step of dividing a correction value of 15 yen from the monitored signal, and a step of applying a correction signal to correct image blur. The present invention provides a method for correcting tJ of an image consisting of

It is known that shading occurs in the image produced by the camera even when only bias light is present.

Bias light is added to the image tube to reduce image retention when the image on the image tube changes. Bias lights do not support flat field of view illumination, so the final image (
In order to prevent shading, it is necessary to correct the image pickup tube output.

Traditionally, such corrections have been made by adding parabolic and sawtooth waveforms to the image tube output.

However, this was difficult, and complete correction was sometimes impossible.

A preferred embodiment of the present invention includes monitoring means for monitoring the signal produced by the binoa slide from each of a plurality of pixels, and comparing each monitored signal with a reference to determine an error value for each pixel. 1-elephant black processing, which comprises a comparator means, a processing means for calculating a correction signal from the error value, and a control means for applying the correction signal to correct the image. l1
Provides iIE neck wear.

A corresponding fourth embodiment of the present invention includes the steps of monitoring a signal generated by bias light from each of a plurality of pixels, and comparing the monitored signal with a reference to calculate an error value for each pixel. A black shading correction method is provided, which comprises the steps of: calculating a correction signal from the error value; and applying the correction signal to correct the image.

Images produced by television picture tubes have errors due to the geometry of the picture tube. Errors arise from the arrangement of the image tube and/or scanner itself. Errors also occur due to nonlinearity of image scanning within one image tube. Generally, such errors become more noticeable toward the edges of the image. I
Geometric errors occur due to nonlinearity of the current waveform and barrel distortion in image scanning.

In the past, the electrical power to the image tube yoke was manually varied to obtain the best final image. This requires a skilled operator and often results in less than optimal results.

A preferred embodiment of the present invention includes monitoring means for monitoring the value of a parameter from each of a plurality of pixels, processing means for calculating a correction signal from the value of the parameter, and correction of a lζ image produced by an electron-optical tube. An image correction device is provided for correcting errors in □□□ made by a geometrically shaped electron-optical tube, comprising control υ11 means for applying a correction signal.

Similarly, a preferred embodiment of the invention includes the steps of: monitoring the value of a parameter from each of a plurality of pixels; calculating a correction signal from the value of the parameter; Correcting J: Applying a correction signal to the fl
This method provides a method for correcting errors that are added to the image.

Due to the optical arrangement of the camera lens, there may be defects in the image produced by the image pickup tube. Such defects are particularly problematic when the focal length of the camera lens changes due to lens replacement or the use of a zoom lens.

The focal length ratio of zoom lenses used in television broadcast cameras is as high as 42:1. Such a drastic change in focal length will naturally result in an equally drastic change in the angle of view, and such a change in angle of view will result in a serious aggravation of the image defects produced by the image pickup tube. Lateral chromatic aberration, which changes with focal length, is also a problem to be solved.

Another embodiment of the present invention includes monitoring means for monitoring the values of parameters of a plurality of pixels, processing means for calculating a correction signal from the parameter IC1, and applying the correction signal to correct image defects. 1. A device for correcting and covering up defects in the camera lens in the image formed by the image tube. Presented by IJt Rir.

In each embodiment, a first correction is applied and its effectiveness is monitored. If necessary, the correction is adjusted to obtain the best correction. Mar: repeated.

A corresponding overlapping example is one of the parameters of each of a plurality of pixels.
The stage of monitoring 1a and the parameters 110 to h
Imaging consists of a step of calculating a positive signal, a step of applying a correction signal to correct image defects, and a step of scaling the correction signal according to changes in the focal length of the camera lens. This is a method of eliminating defects caused by the camera lens in the image produced by the tube.

In each embodiment of the present invention, the four correction signals are multiplied by each of two adjacent pixel error curves in opposite phase to the frequency of the pixel having the difference signal. half of C
Preferably, it consists of multiplying by a certain reference frequency and adding the multiplied signals, thereby including a smooth interpolation of the correction signal.

Advantageously, this smooth interpolation is performed using the scaling 1b nature of the digital-to-analog converter.

Many features of images produced by television cameras deviate from ideal conditions. Because of this deviation, the quality of the images often becomes unwatchable. The present invention seeks to provide automatic correction of such deviations.

Certain features of the corrected 1-power image are selected. The correction device is adapted to measure the value of the associated correction parameter.

The value of the correction parameter is measured from each of the plurality of pixels. The exact method of measuring the correction parameters depends on the correction being performed, but typically involves comparing the signal from each pixel to each reference signal.

Many features of television cameras require correction as to the precise elements of the image passing through the camera's optics. If this is the case, the display plate 17-1 is placed in front of the camera. The destination is a regular 71-rick of rectangular image areas with 15 fi 5 columns. The image area is the same, a small white rectangular area on a black background.

The associated parameters 1iftIJ, each of the image areas 1 through 10, are measured from the generated signals. .

The measured values are converted into digital form to facilitate further processing. The circuitry that monitors the values from each pixel includes analog processing of the gated image, analog-to-digital conversion, and storage in RAM under CPU control. The images are interpolated in real time to provide a correction signal that is applied for real time correction of image defects.

The correction signal is subtracted from the correction parameter 1Ilj by 7.

The quadrupling of the necessary correction signals requires simultaneous processing of line and field data.

Typically, the line speed is on the order of a few -1kl-12, and the field speed is rarely in excess of a few hundred l-1l (!).Low frequency waves can be applied to line data. is not possible at the same time as field data because the information from line data is lost, so it is necessary to use interpolation.

Interpolation methods traditionally used in data processing have used complex software. Such software is expensive, difficult to maintain, and can cause discontinuities in the interpolation, resulting in unsightly positional changes in the image.

The interpolation of the correction parameter values of the present invention is performed using the scaling function of the digital-to-analog converter □.

Other interpolation methods can also be used.

The value of each correction parameter in digital form from a pair of adjacent field data elements is multiplied by a respective reference signal. The reference signals are equal to half the frequency of the pixels of the field and are in antiphase with each other.

The multiplied values are added to form a smooth interpolation.

FIG. 1 shows the interpolation of information at field rate.

Two criteria (iQV REF, 1 and V REF.

2 is shown. The frequency of the reference signal is half the frequency of the pixel whose value is being multiplied. This example - IV R
Er−,1 and V REF,2+li people,! = minimum, and then there is an inverse use C'.

Values of two adjacent field correction parameters.

Explain the interpolation between value 1 and value 2. The base ring signal varies between a minimum of O and a maximum of V. ] Vessel 1, which has a correction parameter value of 1, has VR “l”.

Multiplied by 1, the second? df is the value of the positive parameter (
iQ 2 is multiplied by VREF, 2. Initially, VREF
, 1 sets the value V to the right L/V , RF 1-, 2 High ii
Q O is right. Are you at time Δ in Figure 1?
+Ii i)-The sum of each reference signal including the value of the parameter is vXllllll. When interpolating between values 1 and 2, V RE 1:, 1 decreases to O and VREF,
2 increases to V. Therefore, when the interpolation is between two values, it is indicated at time B in FIG. In other words, the output signal is V
3, which is equal to /2X (value 1+value 2), and the signal when the first pixel has completely transitioned from the first pixel to the second pixel is shown in Figure 1, as marked C. The output signal at this time is equal to the Vx value 2. This process is then repeated between value 2 and the next value, value 3, with the temporal position of 1 value 3 being represented by D in FIG.

A circuit for performing the interpolation shown in FIG. 1 is shown in FIG. Interpolation uses the scaling function of the digital-to-analog converter. Two digital to analog converters 2 and 4 are installed. VREF, 1 is applied to the input of converter 2 and VI'IF, 2 is applied to the input of converter 4.

Initially, a value of 1 is applied to the input of converter 2 and a value of 2 is applied to the input of converter 4. The converter outputs of converters 2 and 4 are applied to a summing circuit 6. The values applied to each digital-to-analog converter 2 and 4 are the respective reference signals VREF,
1 or V REF.

When 2 becomes 0, it is changed. Referring to Figure 1,
At time C, the value applied to converter 2 is changed from value 1 to value 3. Similarly, the value applied to the transducer 4 at the time of day is changed to the value 4, which is the value of the next correction parameter from the ship 2.

Vertical interpolation is performed by mixing the outputs of two multiplying D/A converters. D/AI has a value corresponding to the first column of the ripple, and D/A2 has a value corresponding to 1-5 in the second column. Scan or move the screen downwards/v to move forward D/A
The output from I decreases and the output from L)/Δ2 increases. For this purpose, a cross field is created between the two converter outputs, and vertical interpolation is performed. When the second signal reaches 11, the D/AI inputs the complementary data in the third column to the [coded, 17'Δ2 output decreases and the D/Δ1 output increases, I'' space. The process is repeated. - Influence of combinations on the final analog waveform (J
l (Complement 1 of the above is processed by routine.Ano screen software 1-UL) that interpolates at the edges and corners of the screen (in the necessary waveforms and into blanking areas)' Interpolation of the horizontal ripple region is performed by applying one wave of the analog output of the D/Δ output.

The application of the method described above to the correction of focus related errors will now be described.

For focus correction, a matrix of identical rectangular white areas of 15 rows and 15 columns on a black background is used.

Focus correction is performed for each of several image pickup tubes in, for example, a color television camera.

Since the image of the image pickup tube is uniformly focused on a curved object plane, correction is required for the focusing of the image tube.

Defects due to well-known phenomena also require correction. Conventionally, a predetermined correction signal has been applied to the electrostatic field of the image pickup tube, but the first correction signal has been theoretically obtained or determined by applying the correction signal by trial and error. In the present invention, a 1 DEG carrier tube is manually focused onto a destination 17 which provides a correction signal divided from the value of the best focus parameter for each of a plurality of pixels.

The pixel signals generated by the Vri picture tube are applied to the focus correction device 10 shown in FIG.

The focus correction device 10 includes a monitor means 12 for measuring the value of the best focus parameter at each pixel, and a correction signal δ from the measured value.
The control means 1G includes a processing means 14 which performs a calculation, and a control means 1G which applies a correction signal to correct the electron beam focusing of the R picture tube image.

The image tube image signal is applied to an input 18 of the measurement circuit 12.

The circuit 12 includes a gate circuit 20 responsive to the video signal of the pixels detected by the camera and thus to the image area of the test image I'-i-. The value of the pixel E-BII Kt with the image area precedes the start of the measurement sequence. The gated video signal is output from units 1-20 to bandpass units 1-22. Bandpass Units 1
-22 is the image area under investigation. A single wave of pixel blindness is performed so that only the high frequency energy present in the video information is used for the processing of (1) below.

- The waveformed pixel signal is output to the peak detector 24.

The peak detector 24 detects the maximum and minimum amplitudes of the pixel signals converted into ψ waves. Lean pull unmil hold unit 2
6 receives maximum and minimum amplitude data from a detector 24, and units 20 to 26 constitute a measuring circuit 12.

After the group 1-20 locates the image tube pixel with the image area, the voltage applied to the image carrying focusing electrode 28 is increased in steps over a predetermined range. For each focusing electrode voltage, the highest and lowest amplitudes are detected by the detector 24.
Detected by sample non-hold unit [to 2
6 and further to the CPU which keeps track of which focal electrode voltages do not result in the highest and lowest amplitudes of the pixel signals. After determining and maintaining a record of the focus, the gate control advances to match the next image area of the image area, thus testing all 225 areas. The values for focus are compared to previous passes and corrections are output to increase the focus at each location. This is repeated until the best focusing is obtained at all locations.

The processing means 14 includes an analog-to-digital converter 30, C
PU32. Includes RAM 34 and digital-to-analog converter 36. The signal from the ripple and hold unit 26 of the measurement circuit 12 is in analog form. This signal is converted by analog-to-digital converter 30 and the digital information is stored in RAM 34 under the control of CPU 32.

Under the control of CPU 32, RAM 34 stores information regarding the signal from each pixel. When the measurement operation for all pixels of the image pickup tube is completed, the RAM 34 sets a flat object surface with a uniform focus to Simulation 1! substantially stores the correction No. 4g necessary to compensate for the a fit tube image.

It is unfortunate that the supplement i1E l+6 stored in RAM 34 is applied to the focusing electrode voltage in real time. Therefore,
RA to correct the focusing electrode voltage in real time.
It is necessary to perform an interpolation of the 11θ stored in M34. Is it smooth using the method explained above with reference to Figures 1 and 2? A digital-to-analog converter 36 is used to perform between +Ii and Ii.

The 1=+= signal interpolated by the unit 36 is a modulator 16 interposed between the focal pressure power source 38 and the focusing electrode 28.
The voltage received by the focusing electrode 28 is sufficiently corrected so that the best focusing of the tube image is 1q.
The supplied focus voltage is modulated in response to the interpolated signal from unit 36 so that

The interpolation of the values stored in the RAM 34 utilizes the scaling function of the digital-to-analog converter. The values from two adjacent pixels are then multiplied by their respective reference signals and the multiplied values by 1 are added. Multiplication of values from RAM 34 is performed in the manner described above with reference to FIG. Interpolation is performed by the digital-to-analog unit 36 in Figure 3.
In the circuit shown in FIG. 2 corresponding to t:J',

Color preview cameras must ensure common positioning of all image pickup tubes to prevent defects in the final television image. This is done by placing a test chart within the camera's prism system. The positioning of each image pickup tube is measured with respect to test charts 1 to 1, and a correction signal is calculated by 4 for each image pickup tube and applied to correct the output.

The feet 1 to 1 consist of a regular matrix of 15 rows and 15 columns of identical white rectangular areas on a black background.

Although the correction signal is provided for each image pickup tube, the method is the same for each 1III image tube, so a detailed explanation will be given for only one image pickup tube.

The image produced by the image pickup tube is divided into 225 pixels, each of which is positioned in each image area of the test chart. A pixel signal is measured for each pixel and compared to a reference pixel. The reference pixel is the output of the reference decoy 1 quadrant.

It is usually preferred to be placed in the green 11th channel.

The error signal is generated in response to the I1''! positioning error between the pixel signal and the reference pixel. FIG. 4 shows the variation of the error signal 40.

The above is repeated for all 225 pixels.

Next, one pixel error is interpolated to overturn the image tube correction signal by 34 times.

The circuit that monitors the value from the pixel is the gate part 9 of each pixel.
The position shift detector includes digit-to-digital conversion and storage in RAM under CPU control. The values stored in the RAM are interpolated in real-time with a correction signal 11 applied for real-time correction of visual defects.

Interpolation methods traditionally used in data processing use complex software. Such software is expensive, difficult to maintain, and requires a discontinuous t! 1 may occur, making the final television image unsightly. The interpolation in this embodiment of the invention takes advantage of the scaling functionality of the digital-to-analog converter. Such interpolation was described above with reference to FIG. 1.2.

In color television cameras, each image pickup tube is used to prevent color shading that appears in the final image.
There is a particular need to correct. The blur due to the three color tubes is −C specific for each color.

The software generates a flat field standard for each ml
It is compared with the actual video signal from the picture tube, and a correction signal necessary for the picture tube is calculated. Next, the correction signal g lhl,
It is applied to the output of the image tube so that all images by the image tube are corrected for blur.

The circuit that monitors the value from each pixel is a sample-and-hold circuit at the video level for each pixel.

Analog to digital conversion and RA under CPU control
Contains memory to M. The value 1 stored in RAM is interpolated in real time to provide a correction signal that is applied to correct image defects in real time.

FIG. 5 is a diagram showing the position of the snow image and necessary correction signals. The uncorrected image from the imager tube is represented by a curved line 50 versus the desired linear output 52. The curved shape of signal 50 indicates that it becomes darker towards the edges of the image.

This is a common phenomenon associated with color television cameras, caused by a defect in the image pickup tube and prism called white and black color kneading. From FIG. 5, to generate the desired output signal 52, the Vri picture tube output 5 is
It can be seen that at j, a correction signal indicated by line 54 must be added to 0. Correction signal 14 complements signal 50.

The correction signal 54 is lighted (L) by the following method. The video signal is monitored at each of the plurality of pixels of the snow image. These pixel signals are is compared with each reference pixel signal forming a portion of the flat field I'IQ'- generated by the pixel.The error signal obtained from the comparison of each pixel with the reference is used to calculate a correction signal 54.

A feedback arrangement is used in which the error signal is calculated iteratively until the comparison yields no errors.

For the white 1-lens, 12 J pixel signals are provided to the diascope in the picture tube, so no external image is required to generate the pixels.

Separate detectors are used to examine the movement of black shinning and white shingling. To improve low noise, a black shading detector integrates the entire sample area and outputs a DC potential corresponding to the local video level to the A/D converter. Prior to multiple passes, the image is updated to match the analog correction waveform applied to the video processing amplifier prior to the sample point, so that the effects of the correction are monitored in subsequent passes. By repeating the process, the generated correction waveform is gradually transformed, so that the correction waveform can accurately offset black shading.

In this method, black shading is corrected to 0.5% or less of peak white both absolutely and differentially in measurements after comma correction.

The diascope pattern is again used to measure the white aging error and the video level is measured with a peak detector. This DC level is converted to Δ/1 for use in the Mitaro processor. The exact fit is 4" to the nearest increment of the value stored in the memory.
ClFQ and multiple passes are used with a steady stream of correction waveforms. The white aging error is reduced to less than 1% after gamma correction, both absolute and differential, due to the present method. To overcome the shortcomings of the J3 in the uniformity of the diascope's illumination, the camera uses a flat field of illumination of the external illumination seen through the lens and is memorized by the diascope to later correct the results obtained. It has means to allow for differences between the Tie A7 scope field used in

A simplified block diagram showing an apparatus for monitoring pixel signals and making corrections is shown in FIG.

Video signals from each pixel are applied to inputs 60+,1.

Each video signal is processed by a sample-and-bold unit 1-62, and then an analog-to-digital converter 6
4 is output. The digital card is output to the CPU 66 and stored in the AM 68 under the control of the CPU C36. The values stored in RAM 68 are interpolated to form the correction signal 5/1. Interpolation is digital to analog [J
This is done by utilizing the scaling function of the digital-to-alog converter in the programming unit 70.

Interpolation involves interpolating the error signal from each of two neighboring pixels in digital form with each reference signal. The multiplied value is tightened to perform smooth interpolation. The 1-nit 70 is made from the circuit shown in FIG.
This is done as described with reference to FIG.

In the case of black shading, color preview camera dance @
The tube is equipped with a bias light to reduce afterimages that occur as the image of the camera changes. However, bias lie 1 does not represent a flat field, but rather undergoes correction and shading in the final image.

The color television camera is operated, but the lens is 1X.
Seven dots are left on the camera lens. Therefore, only the bias light is used to generate the output signal from the image tube.

For each imager tube a correction factor is calculated to compensate for the fact that the bias light does not show a flat field.

These corrections (No. 5 are stored and applied to the output from the imager tube during normal operation of the camera and bias line 1 to J
, to prevent shading of the final image. Camera C is only required to high 7 slide 1 and the diascope is switched off.

The circuit that monitors the value from the pixel is a gate section of each pixel, and a single non-pull and small gate at the video level.

RA on analog-to-digital conversion and CPU control
Contains memory to M. The values stored in the RAM are interpolated in real time to provide a correction signal that is applied to correct image defects in real time.

I will explain the correction system for one image pickup tube.

To correct camera tube black shading, monitor the signal generated by the bias light from each of multiple pixels and convert the error signal to each pixel! This is accomplished by comparing each monitored signal with a reference, calculating a correction signal based on the error value, and applying a correction signal that corrects the image produced by the image retraction tube.

The feedback operation is repeated until no error h\t is detected by comparison.

17 tp- is generated by software and interpolated [i (required 'C' in the stem).

The software traditionally used for data processing (interpolation U Atsu Iko.

This embodiment of the invention takes advantage of the scaling capabilities of a digital-to-log converter, as described above with reference to FIGS.

Errors occur in the images produced by television cameras due to camera geometry. Such phenomena include the shape of the image retraction tube itself and the nonlinearity of image scanning within the shaking tube.

The present invention provides automatic correction for such errors.

Image correction monitors pixel signals from the target, generates the basic signal of the desired image, compares the corresponding pixel A word and the basic signal, and from this comparison, 4 correction words are multiplied by R. , 1st!
J, who corrects the image created by the tube, huh? 1n correct i 5; ″I issue・
By applying J, go to line 45.

Rule of rectangular image area with 15 rows and 15 columns 111111'J7
Test chart containing 1 to 1 nocs to 7J Meu's iC
Go to j.

The image area is the same, 11 pure white on a black background.

Geometric correction is usually done with consideration given to positioning the green image in accordance with electronic 10 standards. The pixels are scanned and each pixel is stored for processing 1. Including how to place the sink.

The circuitry that monitors values from the pixels includes gating for each pixel, shape/position detection, log-to-digital conversion, and storage in RAM under CPU control. RAM
The stored values are interpolated in real time to provide a value of 10+<-a which is applied to correct image defects in real time.

The reference generator generates a fundamental signal of the desired image, which is of course an exact replica of the test boot 7-. Pairs of 16 pixels and the base signal are compared by a comparator which generates a difference signal for each pixel.

The difference signal represents the accumulated geometric error.

A correction signal is calculated from the difference signal and applied to correct the image produced by the image retraction tube. The feedback operation is repeated until the minimum error is measured over the entire area. Calculation of the correction signal is performed by a processing means that utilizes interpolation between adjacent different signals.

Control means are provided for storing correction signals and applying correction signals to correct images produced during normal operation of the R picture tube.

The interpolation of different signals by the processing means takes advantage of the scaling function of the digital-to-analog converter. The difference values from neighboring pixels are multiplied by each reference signal, and the multiplied values are added. The interpolation used was described above with reference to Figure 1.2. The measurement and correction method is the same for both shape and positional deviations. 1. The output of the green channel and the reference electronically generated test pattern are first measured for vertical error detection and correction, then for horizontal error detection and correction. Compare for location. The reference itself is switched through the detector so that errors in the detector are automatically compensated for.

Errors are detected by sequentially addressing the detector at each location to determine how much polarity correction is required. Starting from the value of the intermediate point, the memory location of each non-pull is increased 1j11 or decreased depending on the required correction value n1. Death 1-
After each complete pass of the pattern, the values are read out and applied to three scan amplifiers which modify the scan current waveforms and are converted to analog waveforms such as J, . This process is repeated until the accuracy is no longer improved due to the correction of all confidence points: b<(i', N It can be made smaller as the processing progresses.This method corrects the shape of the green chief funnel by 0.1%.Of course, this value exceeds the shape of the lens.

The same method is used for misalignment correction, except that instead of an electronic reference, the reference is the green channel and the correction waveform is applied only to the red or blue scan amplifier, as appropriate. A positioning better than 0.05% of the image height over the entire image area is obtained.To handle the difference between the positioning obtained using the diascope and that made by the lens, the camera is forced to adjust the error difference. Once positioned on the diascope, the camera is aimed at an optical date having the same pattern as the diascope slide. and the distance to the chart are used. The automatic process is repeated to position the camera under these conditions. Differences between the two sets of correction data are stored as correction factors and Used for all subsequent line-up corrections; correction coefficients for several lens types can be stored in memory at the same time. can identify which type of lens it is.

The optical arrangement of the camera lens causes defects in the image produced by the image pickup tube.1 Such defects are particularly problematic when the focal length of the lens changes due to lens replacement or the use of a zoom lens. In color array vision cameras, the ratio of focal lengths of zoom lenses tends to be 42:1. Naturally, the large +1J change in focal length will affect the visual field.
It is accompanied by a large number of changes. Such changes significantly worsen the defects in the Fri tube image. Lateral chromatic aberration is also a particularly important problem for color television and broadcast cameras that use zoom lenses with modest focal length ratios.

These drawbacks are alleviated by generating a correction signal that corrects for defects due to the camera lens in the imager tube image, and by slewing the scanning sclerosis relative to the focal length of the lens -1.

This shape of the lens is particularly effective as the correction signal is scaled to 11 for the focal length of the lens.The shape of the lens is 111 for wide angle applications.
changes dramatically. This is illustrated in FIG.

To generate the correction signal, image signals from each of a plurality of pixels are measured. Each pixel signal is compared with a single or each base reference to provide an error signal for each pixel.
Ru. The error values are stored in digital form and interpolated to provide a correction signal.

Pixels are generated by placing a despot in front of the camera. Test chart i~ is 151:Jl on black background
Contains a regular matrix of 5 columns of rectangular areas of the same white shading.

The system for monitoring the iU from the pixel includes a gating section for each pixel, shape/position detection, analog-to-digital conversion, and storage in RAM under CPU control.
The values stored in M are interpolated in real time to provide a correction signal that is applied to correct image defects in real time 11.

Interpolation methods traditionally used in data processing use complex software. Such software was expensive, difficult to maintain, and could introduce discontinuities in interpolation.

The interpolation between errors in this embodiment of the invention utilizes the scaling function of the digital-to-alog converter.
There is C. Such interpolation is described above with reference to FIGS.

Automatic image correction is provided in each of the above embodiments. In each case, the correction is made using a feedback arrangement, and the adjustment is repeated until a correction is obtained, the effect is monitored, and the best result is obtained.

Although some embodiments of the invention have been described above, it will be readily apparent to those skilled in the art that many modifications can be made without departing from the duck meat of the invention. For example, ゛, reference signal V R
EF, 1 and V REF, 2+gl It may be right-handed to have a linear form.

[Brief explanation of the drawing]

Figure 1 shows the interpolation of different signals, Figure 2 is a block diagram of the circuit that performs the interpolation shown in Figure 1, Figure 3 shows the focus correction system, and Figure 4 is the standard. Figure 5 shows the spatial positioning of pixel signals relative to pixels; Figure 5 shows the desired image tube output, the actual output, and the correction signals needed to produce the desired output; Figure 6 shows the imaging The pipe is 1
FIG. 7 is a block diagram of a part of the device for correcting 2./I... Digital to analog converter, 6...
Addition circuit, 10... Focus correction device, 12... Monitor 1' stage, 14... Processing means, 16... Control means,
20... Gate unit, 22... Band pass unit, 24... Peak detector, 26... Sample and hold conite, 28... Focusing electrode, 30...
・Analog to digital converter, 32...CPU,
34...RAM. 36... Digital to analog converter, 38...
Focal voltage power supply, 40... Error signal, 50... Image pickup tube output, 52... Desired output, 54... Correction signal, 6
0...Input, 62...nucle and hold unit, 64...ana[1g-to-digital converter, 6
6...CPU, 68...RAM, 70...Digital to analog unit. Patent Applicant Link Electronics ABCD FIG, l. FIG, 2゜, positive + t' = bff position FI
G4゜Continued from page 1 Priority claim September 9, 1982 [phase] United Kingdom (G
B) ■8225704 September 9, 1982 [Share] UK (GB) ■ 8225705 ■ September 9, 1982 ■ United Kingdom (GB) ■ 8225706 ° September 9, 1982 [Share] United Kingdom (GB) ■ 8225707 ■ September 9, 1982 ■ United Kingdom (GB) [Yes] 8225709 ■ September 9, 1982 ■ United Kingdom (GB) [Yes] 8225710

Claims (1)

[Claims] 1. Measuring the value of a parameter at each of the li pixels, calculating a correction signal from the parameter value, and applying the correction signal to correct image defects. A method for correcting defects in images with multiple pixels. 2. The correction method according to claim 1, wherein feedback is repeatedly performed when calculating the correction signal. 3. Calculation of the correction signal is performed by multiplying each of the values of two adjacent line or field parameters by each reference signal that has a frequency that is half the frequency of the pixel to which the value to be multiplied is opposite to each other and is in opposite phase to each other. , and the calculation includes smooth interpolation. 4. Measuring the value of the best focus parameter at each of the plurality of pixels; and adding a correction signal from the measured value;
A method of focus correction for an electron-optical tube, comprising the steps of: applying a correction signal to correct the focusing of an image produced by the electron-optical tube. 5. Testing a predetermined range of focus control values for each pixel, identifying and overriding the control 21I value to indicate the best focus for the pixel that has the maximum intensity at the value of the focus parameter. Focus correction method described. 6. providing a plurality of reference pixels; monitoring the pixel signal of the electron-optical tube derived from the reference pixel; and comparing the pixel signal for each pixel with the reference pixel signal; is moved in a first direction to a position with a relatively large misalignment. moving the pixel in a direction opposite to the first direction to a position with substantially the same relatively large misregistration, and determining the best spatial positioning position from the shifted movement and an initial error in the spatial positioning of the pixel; 7. providing a plurality of reference pixels at a relatively small spatial position of the electron-optical tube; , monitoring the pixel signal of the electron-optical tube derived from the reference pixel, comparing the pixel signal with the reference pixel signal for each pixel, and moving the pixel to a position with a relatively large misalignment. moving the pixel in a direction opposite to the first direction to substantially the same position with a relatively large misregistration, and determining the best spatial positioning position and the space of the pixel from the performed movement. Calculate the error signal representing the initial error of target positioning by h1,
The method is characterized by comprising the steps of calculating a correction signal from the calculated pixel error signal, and applying the correction signal so as to compensate for spatial misalignment of the electron-optical tube in an image formed by the electron-optical tube. A method for compensating for spatial misalignment of an electron-optical tube. 8. Monitoring signals from each of the plurality of pixels;
A method for correcting image blur characterized by comprising a step of calculating a correction signal from a monitored signal, and a step of applying a correction signal to correct .eta. of the image. 9. Monitoring the signals generated by the bias line 1~ from each of the plurality of pixels; Comparing each monitored signal with a reference to provide an error signal for each pixel; , a step of applying a control signal to correct the image, and a step of applying a control signal to correct the image. 10. From each of a plurality of pixels. An electronic device characterized by comprising a step of monitoring parameter values, a step of calculating a correction signal from the parameter values, and a step of applying a correction signal so as to correct an image formed by an electron-optical tube. A method of correcting errors caused by the geometrical phenomenon of fish produced by an optical tube. 11. Monitoring the value of the S = I L parameter for each of a plurality of pixels, calculating a correction signal from the parameter value,
Correcting image defects J: Applying a correction signal and scaling the correction signal according to changes in the focal length of the camera lens. 12. Monitoring means (12) for measuring parameter values from each pixel, processing means (14) for calculating a correction signal from the parameter values, and applying a correction signal to correct image defects. 1 characterized in that it consists of a control means (16)
A device (1) for correcting defects in an image having multiple pixels.
0). 13. The apparatus according to claim 1, wherein the processing means repeatedly performs feedback when calculating the correction signal. 14. The processing means sets one of the values of two adjacent line or field parameters (value 1. value 2) each reference having a frequency half the frequency of the pixel having the value to be multiplied and having antiphase with each other. Signal (V REF, 1.V
Two amplifiers (2, 4) multiplying REF, 2), respectively
> and an adder circuit (6) that adjusts the multiplied value, and the calculation of the correction signal includes smooth interpolation. The device described in. 158 an E-monitor means (12) for measuring the value of the best focus parameter at each of the plurality of pixels; a processing means (14) for calculating a correction signal from the measured values; A focus correction device (10) for an electron-optical tube, characterized in that it comprises a control means (1G) for applying a correction signal to perform the correction. 16. The measuring means (12) includes a means for testing the value of the focus parameter (C) that corresponds to a predetermined range of focusing control values for each pixel, and a pixel whose maximum intensity is the value of the focus position parameter for each pixel. A focusing system that expresses the best focus according to 3.1 (
16. The device (10) according to claim 15, characterized in that it comprises means (24) for identifying the a11 value. 11. a reference generator for generating a plurality of reference pixels; monitoring means for monitoring pixel signals of the electron-optical tube derived from the reference pixels; and comparator means for comparing the pixel signals with reference pixel signals; When the pixel is moved in a first direction to a position (△) with a relatively large displacement as measured by the comparator means, the pixel is moved in a first direction to a position (△) with a relatively large displacement as measured by the comparator means. a moving means for moving in a direction opposite to the first direction to a position (C2) and a difference signal representing the position of the best spatial positioning and the initial error of the spatial positioning of the pixel from the movement performed by the moving means; 1. An apparatus for calculating a relatively small spatial positional deviation of an electron-optical tube, characterized in that it comprises a processing means for dividing . 18. A group f that generates a plurality of reference pixels;
a monitor stage for monitoring a pixel signal of the image pickup tube derived from the reference pixel; comparator means for comparing the pixel signal with the reference pixel signal; movement in a first direction to a position of displacement (A>) and movement of the pixel in a direction opposite to the first direction to a position (C2) of substantially the same relatively large displacement as measured by the comparator means; and an error signal representing the position of the best spatial positioning from the movement made by the moving means and the initial error of the spatial positioning of the pixel, and both are calculated: +7'
I: 11j zukarasode iF 4'F. The 14th characteristic consists of a processing means for correcting the image, and a control means for applying a correction signal to compensate for the spatial positional deviation of the image tube with the image created by the image. Monitoring means (62) for monitoring signals from each of a plurality of pixels;
, a means (67-70) for calculating a correction signal (54) from the input signal, and a control means for applying a correction signal (54) to correct the image blur. An image blur correction device 120, characterized by a monitor means for monitoring signals generated from each of a plurality of pixels to a bias line 1, and each monitor for providing an error signal for each pixel! An image black screen characterized by comprising a comparator means for comparing the detected signal with a reference, a processing means for generating a correction signal from an error value, and a control means for applying a control signal to correct the image. I
-Singing correction device. 21. Monitoring means for monitoring parameter values from each of a plurality of pixels, processing means for calculating a correction signal from the parameter values, and correction means for correcting the image formed by the electron-optical tube.
An image correction device for correcting errors due to geometrical phenomena in an image formed by an electron-optical tube, characterized by comprising a control means for applying a correction signal. 22. Monitoring means for monitoring parameter values for each of a plurality of pixels, processing means for calculating a correction signal from the parameter values, control means for applying a correction signal to correct image defects, and a camera. An apparatus for correcting defects caused by a camera lens in an image formed by an imaging chamber, the apparatus comprising: a scaling means for scaling a correction signal according to a change in the focal length of the lens.