US3835247A

US3835247A - Field illumination for image analysis

Info

Publication number: US3835247A
Application number: US00252237A
Authority: US
Inventors: M Soames
Original assignee: Image Analysing Computers Ltd
Current assignee: Image Analysing Computers Ltd
Priority date: 1971-05-11
Filing date: 1972-05-11
Publication date: 1974-09-10
Anticipated expiration: 1991-09-10
Also published as: DE2222852A1; FR2139434A5

Abstract

The invention provides methods and apparatus whereby feature information in each of a number of successive images may be analysed. In particular continuous relative movement between the specimen and image device is provided for thereby eliminating the settlement times characteristic of systems employing intermittent relative movement between specimen and imaging device. The invention envisages the use of a gas discharge tube light source and proposes the insertion of a light diffusing device between the light source and specimen to reduce errors due to movement of the shading pattern caused by movement of the arc from flash to flash. The invention also provides means for sensing the light output and/or illumination level of the field to control the light output from the light source to maintain this substantially constant during an analysis. The high speed of operation achieved by the invention is due to the synchronism of a flashing light source with the continuous relative movement between specimen and imaging device. The specimen may for example be mounted on a microscope stage the latter being moved relative to the optics of the microscope to present different areas for analysis. Alternatively the specimen may comprise small objects located on a conveyor belt and presented in succession to an imaging device.

Description

United States Patent [191 Soames FIELD ILLUMINATION FOR IMAGE ANALYSIS [75] Inventor: I Michael Richard Soames, Fulbourn,

England [73] Assignee: Image Analysing Computers Limited, Melbourn, Royston, England [22] Filed: May 11, 1972 [21] Appl. No.: 252,237

[30] Foreign Application Priority Data May 11, 1971 Great Britain 14230/71 May 27, 1971 Great Britain 17687/71 [52] US. Cl l78/7.2, 178/DIG. l, 178/DIG. 37, 250/205, 250/563 [51] Int. Cl. H04n 1/02 [58] Field of Search ..178/DIG. 1, DIG. 28, 178/DIG. 29, DIG. 36, DIG. 37, 7.2;

[56] References Cited UNITED STATES PATENTS 3,111,555 11/1963 Dykeman et a1 l78/DIG. 1 3,243,509 3/1966 Stut l78/DIG. 1 3,275,744 /1966 Dietrich l78/DIG. 36 3,379,826 4/1968 Gray 178/72 3,517,167 6/1970 Bell 250/205 3,567,854 3/1971 Tschantz et al....... 178/7.2 3,577,153 /1971 Yagi et a1. l78/DIG. 1 3,612,947 10/1971 Dennewitz 250/205 3,624,291 11/1971 Miyata l78/DIG. 28 3,683,108 8/1972 Pieters l78/7.2

Primary Examiner-Robert L. Griffin Assistant Examiner-George G. Stellar Attorney, Agent, or Firm-Browne, Beveridge, De Grandi& Kline [57] ABSTRACT The invention 2 provides methods and apparatus whereby feature information in each of a number of successive images may be analysed. In particular continuous relative movement between the specimenand image device is provided for thereby eliminating the settlement times characteristic of systems employing intermittent relative movement between specimen and imaging device.

The invention envisages the use of a gas discharge tube light source and proposes the insertion of a light diffusing device between the light source and specimen toreduce errors due to movement of the shading pattern caused by'movement of the are from flash to flash;

The invention also provides means for sensing the light output and/or illumination level of the field to control the light output from the light source to maintain this substantially constant during an analysis.

The high speed of operation achieved by the invention is due to the synchronism of a flashing light source with the continuous relative movement between specimen and imaging device.

The specimen may for example be mounted on a microscope stage the latter being moved relative to the optics of the microscope to present different areas for analysis. Alternatively the specimen may comprise small objects located on a conveyor belt and presented in succession to an imaging device.

gvl 3o 13 Claims, 5 Drawing Figures 24 2e 22 SENSE 28 32 l6 7 PEAK v HOLD 1, wane RESET V2 36 68 CLOSE I1 41 A OPEN 4% 40 I A LAMP J CONTROL 1 POWER SOURCE E SAMPLE 10 i Q p l lz A .SET I RESET INHIBIT 66 RESET 58 46 COUNT 56 F FRAME TO r SYNC FRAME 54 SCAN 6O COILS 52 COUNT 2X F 2 62 RESET 4 6 INHIBIT PAIENTE SEP I 0 I974 SHEET u 3o 24 22 SENSE 32 I6 PEAK v HOLD wIIITE RESET I 38 I I 4O LAMP Q IZ ICONTROLI {POWER 49 SI SAMPLE 10 SOURCE.

, RESET INHIBIT 66 RESET 5% COUNT 56 F FRAME To +2 SYNC FRAME 54 SCAN 60 COlLS 52 COlNT 2X F REsET 62 64 INHIBIT.

Fig.1

MOUNTING sTAes g XV-DRIVE 72 4 Tb Y-DRIVE SYNC Fig.2

. '1 (d) pr PAIENTEBSEP] 01914 3.835.247 5 SHEET 2 UF 4 1 FIELD ILLUMINATION FOR IMAGE ANALYSIS The term specimen is intended to mean any single item such as a microscope slide or polished steel sample (for inclusion content analysis) or a film in which each frame is presented in turn for analysis or a series of similar objects (for example on a conveyor belt) which are successively moved into the field of view of the imaging device.

Where different parts of the specimen must be analysed separately either because the whole specimen cannot be analysed simultaneously or the separate parts are presented in succession, relative movement is necessary between the specimen and imaging device. The specimen may be moved relative to a fixed imaging device or vice versa.

Hitherto it has been proposed to effect the movement in small discreet steps in one direction (as with a film) or in two directions at right angles (as with a microscope slide) so that the whole of a specimen can be covered. With such intermittent movement of either the specimen or the imaging device, a settlement time must be allowed at the end of each step before a useful signal can be obtained from the imaging device. This is particularly so where continuous illumination is also employed and the video signal is obtained from a television camera. In practice a single frame scan of the camera is often sufficient to derive all the information from each of the regions presented for analysis. A time period equal to one or more frame scans is usually allowed between read scans to allow for complete erasure of unwanted residual charge patterns on the camera target. However the time required for these is usually much less than the settlement time referred to above and the rate of analysis is therefore governed by the settlement time delays.

It is an object of the present invention to reduce the delays between each read scan (or series of read scans) in the analysis of a succession of different regions of a specimen.

According to one aspect of the present invention a method of analysing optically distinguishable feature content in a specimen comprises the steps of effecting continuous relative movement between the specimen and an imaging device in a plane perpendicular to the optical axis of the imaging device, flash illuminating at least a region of the specimen at intervals of time synchronised with said movement, scanning each image of an illuminated region of the specimen to produce a video signal corresponding thereto, detected relative to a reference voltage those amplitude excursions which relate to features and making measurements thereon.

Preferably constant amplitude pulses are generated from the detected amplitude excursions, of duration equal to the detected excursions, and the measurements are made thereon.

Preferably the speed of the relative movement and the duration of each flash of illumination are selected so that the movement of the specimen relative to the imaging device during an illumination interval is less than the resolution of the system in the direction of relative movement.

Where the images are formed on a television camera tube both the relative movement between the specimen and imaging device and the occurrence of the illumination flashes are synchronised to the camera tube scanning.

Where the camera target is scanned to produce the video signal in a single read scan after each flash of illumination, the latter is preferably accommodated during the frame flyback period just before and read scan.

To produce good charge pattern erasure where a Plumbicon (Reg. Trade Mark) tube is employed, n scans of the camera target are made between each read scan to erase the residual charge pattern on the camera tube target. In this event the illumination flash is applied during the frame flyback period of the last of the n erase'scans prior to each read scan.-

To further reduce the erase time interval the scanning rate for the erase scans may be a multiple of the scan rate for a read scan as provided in co-pending U.S.

\ Pat. No. 3,683,108.

Preferably a gas discharge tube light source is employed, e.g. a Xenon Arc Lamp.

Non-uniformity of illumination characteristic of gas discharge tube light sources and movement of the shading pattern produced by movement of the arc is preferably reduced by diffusing the light from the source prior to it impinging on the specimen.

Preferably overall sensitivity of the imaging'device is controlled by a feedback loop in which the peak white level of the video signal is sensed during each read scan and the sensed value held for a time sufficient to allow a lamp brightness control voltage to be generated therefrom. Preferably the peak white value is compared with upper and lower preset voltages and a warning signal is generated if it exceeds or goes below the two preset voltages.

Alternatively the light output level of the light source is sensed by a photo cell the output of which constitutes a lamp control voltage and this latter is compared with upper and lower coltages and a warning signal generated if the photo cell output voltage exceeds the upper or drops below the lower of the two voltages.

Where a gas discharge tube light source is employed the power supply therefor preferably provides two pulses a firing pulse to initiate each flash and a second pulse of controllable magnitude to sustain the flash for a given duration. Conveniently the firing pulses are controlled by the frame synchronising pulses and the magnitude of the second pulses is controlled by the value of the lamp brightness control voltage. Preferably the lamp brightness control voltage is averaged so that abrupt changes do not occur in the light output.

Preferably a limit is imposed on the maximum energy which can be imparted to the light source to prevent the latter being damaged.

The light output from a flash tube is not always constant and may vary from flash to flash. According therefore to a further feature of the present invention the value of the reference voltage with which the video signal amplitude excursions are compared for the purpose of detection, is varied in sympathy with variations in the light output per flash. Preferably the reference voltage is itself derived from the peak white video signal amplitude as for example described in co-pending U.S. Patent application Ser. No. 84,72I or U.S. Patent application Ser. No. 227,160. Typically, the voltage so derived is increased or decreased by an appropriate amount to compensate for increased or decreased light output.

The specimen may be reflective or a transparent or semitransparent or may comprise a plurality of similar objects presented to the imaging device in a sequence such as mass produced articles on a conveyor system. The movement of the latter may be continuous and an image of each article obtained by flash illuminating the article when situated in the field of view of the imaging device. Preferably the movement and flash producing apparatus are synchronized and presence detecting means are provided to sense when an article occupies a given position in the field of view.

The invention will now be described by way of example with reference to the accompanying drawings, in which:

F [6.71 is a block circuit diagram of part of an image analysis system embodying the invention,

FIG. 2 is a part diagrammatical illustration and a part block circuit diagram of another part of the image analysis system of FIG. 1,

FIG. 3 illustrates waveforms of signals obtainable at different points in the circuit of FIG. 1,

FIG. 4 is a block circuit diagram of a power supply for a flash tube, and

FIG. 5 is a block circuit diagram of a modification which may be included in an illumination system to compensate for variation in light output of successive flashes from a flash tube.

In FIG. 1 a specimen is carried on a movable stage 12 forming part of a microscope whose objective is shown at 14. The enlarged image of the field of view is seen by a television camera 16 and the light from the objective 14 passes through a semi-reflecting mirror 18 by which light from a gas discharge tube 20 is supplied to illuminate the specimen 10.

The arrangement shown provides incident light for a reflecting specimen 10. Where the specimen is partly transparent and is to be viewed by transmitted light, the semi-reflecting mirror is unnecessary and the light source 20 instead supplies light toa conventional condenser lens system located beneath the specimen and stage 10, 12.

Any non-uniformity of illumination of the specimen may be compensated for by a shading correction circuit (not shown) of .the type described in co-pending U.S. Pat. No. 3,743,772. The correction would be applied to the signals from camera 16 before'being applied to junction 22.

Movement of the arc from flash to flash causes the shading pattern produced by the non-uniform illumination to move relative to the camera target. To overcome this, a diffuser 21 is interposed between the discharge tube and the optics of the microscope.

The video signal output from camera 16 which appears at junction 22 is applied to a peak white sensing circuit 24 whichfor example comprises a peak rectifying circuit. The peak white value is sensed at least at the beginning of each read scan (hereinafter defined) and the value held for the duration of the read scan period by a HOLD circuit 26 so that the peak white voltage for the scan appears at junction 28.

The signal at 28 is compared with two reference voltages V1 and V2 respectively in two

comparators

30 and 32 respectively and the outputs from the two comparators are commoned to appear at junction 34 forming the set input to a bistable device 36. The

comparators

30 and 32 are arranged to provide an output signal if the voltage at 28 exceeds V1 or drops below V2. If the signal at 28 lies between V1 and V2 no signal appears at junction 34.

The signal at 28 is also supplied via an averaging or smoothing circuit (not shown) to a power amplifier 38 serving to control the power to the discharge tube 20.

' The averaging or smoothing circuit (not shown) may comprise an RC or LC (or combination thereof) circuit. Power from a source (not shown) is supplied to the amplifier 38 via a controlled switch 40. This latter is opened at appropriate intervals in the analysis sequence from signals serived from the frame synchronising pulses (see FIG. 3).

The signal at junction 28 serves to control the gain of amplifier 38 and the control of the gain follows an inverse law such that an increase in the peak white value reduces the power supplied to the lamp. If the voltage appearing at junction 28 is so small as to drop below V2 it is assumed that something is probably wrong with either the specimen or the discharge tube or the camera tube and the set output Q from bistable device 36 which appears at junction 42 serves as a warning signal G Likewise if the voltage at junction 28 exceeds V1, it is again assumed that there is something wrong due to the very high peak white value obtained and bistable device 36 is again set so that a warning signal again appears at junction 42.

In the system shown a single warning signal is obtained irrespective of whether the peak white value exceeds or is below the acceptable range determined by V1 and V2. It will be appreciated that two separate warning signals may be obtained by supplying the out put from comparator 30 to a second bistable device (not shown) instead of to device 36 and arranging that the set output from the second bistable device (not shown) constitutes the second warning signal.

Since something is probably wrong if a warning signal is generated, the signal at junction 42 serves to close a gate 44 in the signal path from junction 22. In this way the video signal from the camera 16 is inhibited in the event that the peak white value at 28 lies outside the range determined by V1 and V2.

The video signal released via gate 44 is compared with a reference voltage from e.g. a potentiometer 45 by comparator 47 to produce a constant amplitude pulse for each amplitude excursion of the video signal which exceeds the reference voltage, of duration equal to that of the detected excursion. The pulses are chopped by a gate 49 operating at a high-frequency relative to the average duration of a constant amplitude pulse and the pulse trains so produced fed to a computing circuit 51 by which measurements may be made on the signals relating to the detected parts of the image producing the video signal.

The control signals for gate 40 are obtained from a divide by two device 46 acting as a pulse switching circuit having as input the frame synchronising pulses normally employed to synchronise the frame scan deflection currents for the camera 16. conventionally these synchronising pulses occur during the frame flyback period of each normal frame scan period for the camera. The normal frame synchronising pulses appear at junction 48. However due to the action of the divide by two circuit 46, the signal appearing at junction 50 will correspond to that as shown in FIG. 3(0). The missing frame synchronising pulses will appear in the other output of device 46 at junction 52. These pulses are shown in FIG. 3(b).

The signal at junction 50 triggers a first frame scan deflection current generator 54 which produces a normal frequency frame scan deflection current at junction 56 which is supplied to frame scan coils (not shown) of the camera 16. A counter 58 set to count 1, counts the number of deflection current pulses from generator 54 and inhibits the generation of any subsequent pulses after the first 1.

The signal at junction 52 drives a second frame scan deflection current generator 60 which operates at twice the frequency of generator 54. These double frequency deflection currents appear at junction 62 and are conveyed to the frame scan deflection coils of camera 16 at the end of each single frame scan deflection current from generator 54. A second counter 64 which is set to count 2 counts the deflection current signals at junction 62 and inhibits the generation of any subsequent deflection signals by generator 60 after two deflection current signals have appeared at junction 62.

A reset signal is applied to each of the

counters

58 and 64 to reset each counter to zero by the arrival of a pulse at

junction

50 or 52 serving the generator associated with the other counter. In this way each of

counters

58 and 64 is ready to count I or 2 respectively at the appropriate intervals in the sequence.

The two signals at

junction

50 and 52 also serve as set and reset signals for a bistable device 66 whose set output 0 serves as an open signal for gate 68 in series with gate 44. In this way video signal output from camera 16 is inhibited during all but the read scans formed by the scan deflection currents at junction 56.

A reset signal is shown for bistable device 36 by which the device may be reset to inhibit the Q output after a warning signal has been generated and some action has been taken to correct the fault producing the warning signal. Conveniently the reset signal is applied by manual operation of a switch or the like.

The two frame scans produced by the double frame scan rate generator 60 serve as erase scans for the camera 16 and since twice the normal frame scan rate is employed, two complete erase scans are obtained in the time normally required for one frame scan. This produces more complete erasure of the residual charge pattern on the camera target at the end of a read scan. Additionally the beam may be defocussed and the beam current increased to more effectively remove the charge pattern, as described in U.S. Pat. No. 3,683,108.

As shown in FIG. 2 the stage 12 on which the specimen is mounted is moveable in two directions at right angles by a first drive mechanism 70 which moves the stage in a so-called X direction and a second drive mechanism 72 which moves the stage 12 in a so-called Y direction perpendicular to the X direction.

Conveniently one of the

drives

70, 72 is a continuously running and reversing motor and the other is a stepping electric motor. The speed and stepdistance of the two motors is controlled by a synchronising control system 74 which conveniently drives its control signal from junction 48 in FIG. 1. The continuous reversing motor operates to cause the specimen to move completely across the field of view in the particular X or Y direction selected for the continuous movement and to reverse at the end of a complete traverse and the stepping motor operates to shift the specimen to produce I a fresh line or column (as the case may be) of specimen surface to be presented to the optical system of the microscope.

FIG. 3 illustrates waveforms obtainable at certain points in the embodiment of FIG. 1. Thus at FIG. 3(a) the frame deflection signals applied to the scan coils appear as alternate slow ramps 76 separated by pairs of fast ramps 78 whose rise and decay time is equal to half that of the ramps 76.

The frame deflection currents are synchronised from a master oscillator (not shown) which produces a pulse signal having a repetition frequency equal to twice that of a signal containing only the slow ramp waveforms 76. This signal (not shown) is applied to junction 48 in FIG. 1 and by the action of the divider network 46, the successive pulses appear alternately at outputs 50. and 52. The signal at junction 52 is shown in FIG. 3(b) and that of junction 50 in FIG. 3(e).

The pulses at junction 50 are stretched (or shrunk) as is required to produce a pulse of appropriate duration to open the gate 40 and allow power to the control 38 for discharge tube 20. In this way a flash is produced during the flyback interval of the second of each fast ramp deflection voltages 78. As shown at FIG. 3(d) these pulses are reduced in width so as to produce a flash which in a preferred arrangement lasts only for 50 microseconds.

Thesignals at jucntion 50 also serve to set bistable 66 and conveniently a short delay (not shown) is provided between junction 50 and the set input of bistable 66 to ensure that the latter operates at the instant of the beginning of the slow ramp 76. The set (Q) output of bistable 66 opens gate 68 to release video from camera 16 and a reset signal for bistable 66 is derived from junction 52. The form of the set (Q) output from bistable 66 is shown in FIG. 3(e).

FIG. 4 illustrates one form of flash tube power supply and control circuit and replaces the

circuit elements

38 and 40 of FIG. 1. The flash tube 20 is supplied with trigger pulses to initiate a flas'h from a trigger pulse generator generally designated 102 and each flash is maintained for a given duration by a sustain circuit generally designated 104.

Each firing pulses is generated by the action of a pulse transformer 106 with a high step-up rate having a primary winding 108 connected in series with capacitor 110, the series circuit being connected in parallel with a silicon controlled rectifier 112. Signals from junction 50 (see'FIG. 1) are applied to the firing electrode of SCR 112 (if necessary after amplification by an amplifier, not shown) to fire the SCR 112.

Current for the SCR 112 is derived from the capacitor which is charged through a resistor 116 from a source of direct current 118.

The value of resistor 116 is selected so as to cause SCR 112 to become cut off after the capacitor 110 has become substantially discharged.

The high secondary voltage from the transformer 106 is applied to a spark gap 120 and the polarity of the connections of the transformer are selected so that the pulse which appears at junction 122 when the spark gap conducts, is of the correct polarity to fire the flash tube 20. When the latter fires, it acts as a low resistance discharge path for capacitor 124 and surge limiting inductance 126 forming a damped resonant circuit. The values of 124, 126 are chosen so that with capacitor 124 in a fully charged state, a single light pulse is obtained of a suitable duration,leaving 124 substantially discharged.

Capacitor 124 is charged via bridge rectifier 128 and diode 130 from a source of alternating current 132.

As shown the source and bridge rectifier are transformer coupled and a current limiting inductor 134 and power factor correction capacitor 136 are also shown. A bleeder resistor 138 is connected in parallel with capacitor 124 between

junctions

140 and 142 and a potentiometer resistor network formed from two

resistors

144 and 146 is connected between junction 148 and junction 150. Junction 150 is supplied with the voltage output of an amplifier 151which derives its input from junction 28 of the circuit of FIG. 1.

The junction of the two

resistors

144 and 146 junction 152 provides one input to a differential amplifier 154 whose other input is fixed for example at earth potential. The output from the differential amplifier 154 provides the voltage for the firing electrode of a second SCR 156. When the voltage across capacitor 124 reaches the required level, the output from amplifier 154 is arranged to be sufficient to cause SCR 156 to fire causing diode 130 to be cut-off. Capacitor 124 cannot charge any further. Thus if the control voltage applied to junction 150 rises, the voltage at junction 148 can rise to higher value before diode 130 is cut-off and vice versa.

The duration of the light flash is determined by the value of capacitor 124 and inductor 126 and the arc resistance of flash tube 20.

The inductor 126 serves to substantially prevent the transfer of a firing pulse from junction 122 to capacitor 124 in addition to controlling the surge of current from capacitor 124, when the discharge occurs in flash tube 20.

In FIG. the flash tube provides the illumination for a specimen (not shown) which is viewed by a television camera 156. The light from the flash tube 20 is partly reflected (to illuminate the specimen) and partly transmitted by a semi-reflecting mirror 158. That which is transmitted is focussed onto a light sensitive device such as a photodiode or a photocell 160 and the signal obtained therefrom is amplified by amplifier 162. The amplifier output signal is averaged over a period of time in an averaging (or smoothing) circuit 164 which conveniently comprises an RC network of long time constant and the instantaneous light level output from amplifier 162 is compared with the averaged value in a differential amplifier 166 serving asa comparator.

The system of FIG. 5 is designed to be included in an image analysis system in which the camera 156 scans each of a succession of fields and provides a video signal which is processed and on which measurements may be made. The succession of fields may be obtained from a single specimen as by producing relative movement between the specimen and camera or may arise from locating a succession of similar objects in the field of view of the camera. In either event each field is illuminated by a flash of light from the flash tube 20 and the image so produced on the target of thecamera tube is subsequently scanned during a read scan in the manner hereinbefore described. To this end the averaging circuit 164 is set to average the light output value from amplifier 162 over a number of field scans.

The average value applied to differential amplifier 166 therefore comprises a signal representing the mean light output from the flash tube 20 and the positive or negative difference signal obtained in the output of differential amplifier 166 after any particular flash will represent the divergence of the light output of that flash from the mean. This difference signal is held for the duration of the subsequent read scan in a hold circuit 168 which is reset at the end of the read scan and the value so held may be employed to compensate for the variation in the light output of the flash tube 20.

The output from camera 156 is shown applied to one input of a comparator 170 forming part of a detector for the image analysis system. The other input is derived from the tapping of a potentiometer 172 and the action of the comparator 170 is to provide a binary type signal at junction 174 which has a l-state when the instantaneous amplitude of the video signal from camera 156 exceeds the reference voltage set on the potentiometer 172 and a zero signal level when the instantaneous amplitude falls below this reference voltage. The term exceeds is intended to be a relative one and includes'both positive going excursions which go above a reference voltage and negative going excursions which go below a reference voltage.

The pulses v.from the comparator 170 are fed to a computing circuit 51 (see FIG. 1) for analysis. The pulses may be chopped electrically as by a gate 49 (FIG. -1) not shown in FIG. 5.

Compensation for varying light level is achieved by altering the potential applied to the potentiometer 172.

' To this end the power supply for the potentiometer 172 includes a variable gain amplifier 176 whose gain is controlled by the output from the difference amplifier 166 which is maintained for each read scan and whose input is provided with a suitable voltage from a source (not shown) applied to junction 177 equivalent to, for example the peak white amplitude level of the video signal from the camera 156. The amplifier 176 operates to increase the voltage across potentiometer 172 with an increase in the light level output from amplifier 162 and to decrease the voltage in the event of a drop in the light level output of amplifier 162 relative to the mean light output value obtained from average circuit 164.

It will be appreciated that an image analysis system including the modification shown in FIG. 5 will be capable of analysing a succession of fields in which each contains a similar but different article of a sequence of for example mass produced articles in which the reflected or transmitted light from the separate articles may vary from. field to field and in which the light output from the source 20 may also vary from field to field.

I claim:

1. A method of analyzing optically distinguishable feature content in a specimen, comprising the steps of: effecting continuous, relative movement between the specimen and an imaging device in a plane perpendicular to the optical axis of the. imaging device, flash'illuminating at least a region of the specimen at intervals of time synchronized with said movement with a gas discharge tube'light source, scanning each image of an illuminated region of the specimen to produce a video signal corresponding thereto, detecting relative to a reference voltage those amplitude excursions which relate to features, making measurements thereof, sensing the light output per flash of said gas discharge tube light source, generating a voltage whose magnitude is proportional to the light output of each flash, averaging the generated voltages so produced to produce an average value voltage, comparing the level of the generated voltage for each flash with the average value voltage, and generating a difference voltage, controlling the value of a further voltage in response to variation in the magnitude of the difference voltage and employing as the reference voltage for said video signal either said further voltage or a voltage derived therefrom.

2. The method as set forth in claim 1 wherein the imaging device forms the image of the illuminated region on a television camera tube of the Plumbicon type, the relative movement between said specimen and said imaging device and the occurrence of said illumination flashes being synchronized to the camera tube scanning, the video signal arising during the first frame scan of the camera tube target after each flash being supplied as said video signal, the specimen region being illuminated during at least part of the frame flyback period immediately preceding each said first frame scan and n frame scans of the camera tube target being performed between the end of each first frame scan and the next flash of illumination, thereby to more completely erase the residual charge pattern on the camera tube target.

3. The method as set forth in claim 1 wherein the illumination is obtained from a gas discharge tube light source, further comprising the step of diffusing the light before it is employed to illuminate a region of specimen.

4. The method as claimed in claim 1 wherein the specimen comprises a plurality of similar objects which by virtue of the relative movement are presented to the imaging device in turn, further comprising the step of sensing when each object in turn occupies a given position relative to the imaging device, generating a voltage pulse when each object in turn occupies the given position, and initiating a flash or illumination in response to the generation of each said voltage pulse.

5. The method as set forth in claim 1 further comprising the steps of generating a constant amplitude pulse from each detected amplitude excursion of duration equal to that of the detected excursion, and making measurements on said constant amplitude pulses.

6. The method as set forth in claim 5 further comprising the step of gating at a high frequency at least each constant amplitude pulse to replace it by a plurality of short pulses the number of which is proportional to the duration of the constant amplitude pulse.

7. A method of analyzing optically distinguishable feature content in a specimen, comprising the steps of: effecting continous relative movement between the specimen and an imaging device in a plane perpendicular to the optical axis of the imaging device, flash illuminating at least a region of the specimen at intervals of time synchronized with said movement, scanning each image of an illuminated region of the specimen to produce a video signal corresponding thereto, sensing the peak white amplitude level of the video signal during the beginning of each read scan, generating a voltage whose value is proportional to the sensed amplitude level, holding said voltage for the remainder of that read scan period, controlling the duration of the light output of the flash corresponding to said read scan in dependence on the value of the generated voltage, detecting relative to a reference voltage those amplitude excursions which relate to features, and making measurements thereon.

8. The method as set forth in claim 7 further comprising the steps of generating a firing pulse to initiate said flash illumination and generating a sustain pulse having a magnitude corresponding to said held voltage to sustain each illumination for a duration dependent on said magnitude.

9. The method as set forth in claim 7 further comprising the step of averaging or smoothing the generated voltage so as to remove abrupt changes in the magnitude thereof.

10. The method as set forth in claim 7 in which the imaging device forms the image of the illuminated region on a television camera tube of the Plumbicon type, the relative movement between the specimen and the imaging device and the. occurrence of illumination flashes being synchronized to the camera tube scanning, the video signal arising during the first frame scan of the camera tube target after each flash being supplied as the video signal, the specimen region being illuminated during at least part of the frame flyback period immediately preceding each said first frame scan and n frame scans of the camera tube target being performed between the end of each first frame scan and the next flash of illumination, thereby to more completely erase the residual charge pattern on the camera tube target.

11. The method as claimed in claim 7, wherein the specimen comprises a plurality of similar objects which by virtue of the relative movement presented to the imaging device in turn, further comprising the steps of sensing when each object in turn occupies a given position, and initiating a flash or illumination in response to the generation of each said voltage pulse.

12. The method as set forth in claim 7 further comprising the steps of generating a constant amplitude pulse from each detected amplitude excursion of duration equal to that of the detected excursion, and making measurements on the constant amplitude pulses.

13. The method as set forth in claim 12 further comprising the step of gating at a high frequency at least each constant amplitude pulse to replace it by a plurality of short pulses the number of which is proportional to the duration of the constant amplitude pulse.

Claims

1. A method of analyzing optically distinguishable feature content in a specimen, comprising the steps of: effecting continuous, relative movement between the specimen and an imaging device in a plane perpendicular to the optical axis of the imaging device, flash illuminating at least a region of the specimen at intervals of time synchronized with said movement with a gas discharge tube light source, scanning each image of an illuminated region of the specimen to produce a video signal corresponding thereto, detecting relative to a reference voltage those amplitude excursions which relate to features, making measurements thereof, sensing the light output per flash of said gas discharge tube light source, generating a voltage whose magnitude is proportional to the light output of each flash, averaging the generated voltages so produced to produce an average value voltage, comparing the level of the generated voltage for each flash with the average value voltage, and generating a difference voltage, controlling the value of a further voltage in response to variation in the magnitude of the difference voltage and employing as the reference voltage for said video signal either said further voltage or a voltage derived therefrom.

10. The method as set forth in claim 7 in which the imaging device forms the image of the illuminated region on a television camera tube of the Plumbicon type, the relative movement between the specimen and the imaging device and the occurrence of illumination flashes being synchronized to the camera tube scanning, the video signal arising during the first frame scan of the camera tube target after each flash being supplied as the video signal, the specimen region being illuminated during at least part of the frame flyback period immediately preceding each said first frame scan and n frame scans of the camera tube target being performed between the end of each first frame scan and the next flash of illumination, thereby to more completely erase the residual charge pattern on the camera tube target.