WO2009030923A1

WO2009030923A1 - Inspection system

Info

Publication number: WO2009030923A1
Application number: PCT/GB2008/003012
Authority: WO
Inventors: Steve Gusterson
Original assignee: Mettler-Toledo Safeline X-Ray Limited
Priority date: 2007-09-08
Filing date: 2008-09-08
Publication date: 2009-03-12
Also published as: GB0717520D0

Abstract

A method of controlling an inspection system (100) which generates radiation from a source (102) and allows the radiation to be detected by a detector (122) during an exposure period (C) of a control waveform used to drive the detector (122), the control waveform comprising a reset pulse (R) in addition to the exposure period (C), the reset pulse (R) having a nominal duration and the method comprising calibrating the inspection system (100) by increasing the relative length of the reset pulse (R), to a calibration duration, when compared to the exposure period (C) whilst maintaining a substantially constant period (T) of the control waveform.

Description

INSPECTION SYSTEM

Field of the invention

This invention relates to an inspection system and related methods of inspecting articles. Generally, the inspection system will be an X-ray inspection system but this need not be the case.

Background of the invention

There is an on-going need to inspect articles, whether this is the inspection of baggage in an airport, or other transport related situation, or in the output of a production process. For example, it is common in the food industry to inspect the actual content of the food in order to determine that the food content is as desired and does contain any foreign bodies such as stones, bone fragments, metal from the machines used in the production of the food, or the like.

A typical x-ray inspection apparatus comprises a conveyor arranged to carry objects or materials to be inspected through the apparatus. Within the apparatus there is an x-ray source, generally with a collimator associated therewith arranged to produce a narrow irradiation zone extending across the conveyor. Beneath the conveyor there is provided a detector arranged to detect x-rays which have passed through an object, on the conveyor, passing through the irradiation zone.

The detector generally comprises a linear array of photo-diodes, extending across the conveyor, adjacent the irradiation zone. The photo- diodes are generally provided in a series of modules, each of which contains a plurality of photo diodes. Often a phosphorescent strip is mounted above the photo-diodes within a module and x-rays which are incident upon the phosphorescent strip cause light to be emitted therefrom. The intensity of the light emitted from the phosphorescent strip is proportional to the amount of x-rays that are incident upon it and the light output is detected by the photo-diodes.

Thus, the output from the detector can be used to give an indication of the amount of x-rays which are reaching the phosphorescent strip (or bare diodes) through the irradiation zone. The amount of x-rays reaching the detector will be dependent upon the nature of the object which is passing through the irradiation zone; denser materials such as bone, metal, stone and the like will absorb more x-rays that material such as meat, or other foodstuffs. Likewise, the absence of material, such as due to a void, will absorb less x-rays than meat or other foodstuff. Therefore, the amount of x-ray reaching the detector can be used to determine whether there is foreign matter in the product, or indeed whether there is an absence of matter.

X-ray detectors, including diodes, are able to receive a finite level of X- rays before their output saturates. Such saturation may sometimes be a beneficial technique during the inspection process, but is to be avoided during the off-line calibration process.

It some modes of operation it can be desirable to set the power of the X- ray source such that with, no object present, the detector is taken into saturation but, when an object is present, a proportion of the X-rays is absorbed such that the signal reaching the detector no longer causes it to saturate. It is however, difficult to calibrate the detector for such modes of operation since the detector will always be in saturation during calibration thereby rendering traditional calibration techniques redundant. A conventional technique to avoid saturation during online inspection is to reduce the level of X-rays emitted during calibration and then to re- increase the level of X-rays during online inspection. A problem exists where the control system does not have the facility to reduce the X-ray dose from that encountered during online inspection. In this case it would not be possible to perform a valid calibration of the detector.

Summary of the invention

According to a first aspect of the invention there is provided a method of controlling an inspection system which generates radiation from a source and allows radiation to be detected by a detector during an exposure period of a control waveform used to drive the detector, the control waveform having a substantially constant period and comprising a reset pulse in addition to the exposure period, the reset pulse having a nominal duration and the method comprising calibrating the inspection system by increasing the relative length of the reset pulse, to a calibration duration, when compared to the exposure period whilst maintaining the substantially constant period of the control waveform,

The exposure period and reset pulse duration, that comprise the control waveform, are interdependent and maintaining a substantially constant period of the control waveform means that changing one changes the other. For example, if the exposure period were reduced then the duration of the reset pulse would be increased by the same amount, and vice versa.

Generally, the system is arranged to use X-rays as the radiation, but other forms of radiation might be used. For example, the energy might be any other electromagnetic wave such as light (whether, visible, infra red, ultra violet, etc.) , radio waves, microwaves, millimetre waves, and the like.

Such a method can lead to an improved output of the inspection system since it can help to prevent the system becoming saturated during calibration of the system thereby allowing the system to be correctly calibrated. Reducing the exposure period may be thought of as reducing the exposure of the system since it allows less time for dose to accrue on the detector.

In one use of the method, the reset pulse may be reduced in length (thereby increasing the exposure period) when compared to the calibration duration used to calibrate the system during online inspection of articles. In some embodiments the reset pulse may be reduced in length to have, or to be close to having the nominal duration.

Scanners are generally operated with the reset pulse at a minimum duration as this allows the exposure period to occupy as much of the period of the control waveform as possible, giving the longest period for radiation to accumulate on the detector. Decreasing the length of the reset pulse (and thereby increasing the length of the exposure period) should make the system more sensitive as there is further time for radiation to accumulate. The nominal duration may be the minimum possible duration for a particular scanning apparatus but, in other embodiments, it is conceivable that the nominal duration is greater than the minimum possible duration for that apparatus. Thus, the nominal duration might be the duration to which the reset pulse would generally be set during operation of the system.

Decreasing the length of the reset pulse during online inspection may allow the system to better image objects passing through the detector which would generally absorb a high percentage of the radiation emitted from the source. Decreasing the length of the reset pulse may be considered synonymous with increasing the length of the exposure period.

The method should allow the system to be calibrated, without the power of the source being altered, such that the radiation can be set at a power which would cause the detector to saturate when no object is present which can improve the signal level output by the detector thereby improving the results of the scan.

In a second, additional, or alternative, use of the method, the reset pulse may remain substantially at the calibration duration during online use of the system. Such a method is advantageous for objects which do not absorb a high level of the radiation output from the system as the reduced exposure period can help ensure that the detector does not saturate during online inspection. In this manner, the radiation source may operate at a fixed high level, and the dose accrued at the detector adjusted by means of the reset pulse duration and/or conversely the exposure period duration.

The second use of the method may be thought of as using the system as close as possible to the saturation point of the detector, without allowing the detector to enter the saturation condition. This would be the case where an ideal contrast radiation dose for the object under inspection may not approach the higher levels of the possible outputs and varying the length of the exposure period can help to improve the use of the available output range of the detector.

The skilled person will appreciate that length of the reset pulse may be set to any duration between the minimum duration and the calibration duration thereby varying the exposure period duration with equal measure.

In other, alternative or additional, embodiments the reset pulse may be increased in length beyond the calibration duration thereby reducing the exposure further.

Conveniently, the output from the detector is digitised. In one particular embodiment, the output from the detector is digitised to 14 bits of resolution. In other embodiments, however, the output from the detector may be digitised to roughly any of the following: 8, 9, 10, 11 , 12, 13, 15, 16, 17 or more bits. A further embodiment may be a detector that outputs over an analog range, of perhaps 0-10 volts, without digitisation. The skilled person will appreciate that this voltage range is given by way of example only and that other voltages, and/or currents might be used.

During calibration, the duration of the reset pulse may be increased (i.e. the length of the exposure period is reduced) in a stepwise manner until an exposure period is set in which the output of the detector does not saturate.

In some embodiments the duration of the reset pulse may be reduced in steps of roughly 5%. The skilled person will appreciate that other steps may be used, such as 10%, 15%, 20%, 25%.

In other embodiments, the duration of the reset pulse may be reduced in steps by an absolute amount in each step.

In other embodiments, the system may be arranged to increase the duration of the reset pulse to a predetermined level and/or conversely to reduce the duration of the exposure period to a predetermined level. For example, the system may be arranged to reduce the exposure period to roughly 10% of its maximum period. The skilled person will appreciate that when the reset pulse has a minimum period then the exposure pulse has a maximum period. Other percentages are equally possible.

According to a second aspect of the invention there is provided an inspection system comprising: a source of radiation; a timer arranged to generate a control waveform having a measurement period and comprising a reset pulse, having a minimum period, and an exposure period; a detector arranged, in use, to allow the radiation to be detected during the exposure period of the control waveform; and processing circuitry arranged to allow the control waveform to be varied such that the length of the reset pulse is increased, during calibration of the system, beyond the minimum period.

In some embodiments, the detector comprises one or more, and generally a plurality of, photodiodes. Generally, these photodiodes will be arranged in a linear array.

In other embodiments, the detector may be provided by one or more photomultiplier tubes.

In some embodiments, a phosphorescent strip is mounted above the photodiodes. Such a strip is advantageous as it enhances the ability of the diodes to detect radiation, and in particular X-rays. However, in some embodiments, diodes may be used with no phosphorescent strip.

The system may be arranged to digitise the output of the detector. Conveniently, the system is arranged to keep the period of the control waveform substantially constant during calibration and use of the machine.

By way of background, it is known that some systems, may be arranged to vary the period of the control waveform, perhaps by varying the length of a reset portion of the control waveform, and such arrangements may be used if it is desirable to adjust the scan rate of the objects passing through the system.

The system may be arranged such that the amount by which the exposure period can be reduced can be varied. Such a system may be beneficial as it would allow the calibration of the machine to be made according to the type of object that was to be scanned. Dense objects may require a higher dose in operation that would otherwise undesirably saturate the detector during calibration stage, whereas a degree of saturation may be preferable during operation.

According to a third aspect of the invention there is provided a machine readable medium containing instructions which when read by a machine cause that machine to perform the method of the first aspect of the invention.

According to a fourth aspect of the invention there is provided a machine readable medium containing instructions which when read by a machine cause that machine to function as the machine of the second aspect of the invention.

The machine readable medium in any of the above aspects of the invention may be any of the following: a floppy disk; a CDROM; a DVD

(including + R/ + RW, -R/-RW, RAM) ; a hard disk; a memory (including memory sticks and the like) ; a tape; a transmitted signal (including an Internet download, an ftp transfer and the like) ; a wire; or the like.

According to a fifth aspect of the invention there is provided a method of controlling an inspection system which generates radiation from a source and allows radiation to be detected by a detector during an exposure period of a control waveform used to drive the detector, the control waveform having a substantially constant period and comprising an exposure period in addition to a reset pulse, the reset pulse having a nominal duration and the method comprising calibrating the inspection system by decreasing the relative length of the exposure period, to a calibration duration, when compared to the reset pulse duration whilst maintaining the substantially constant period of the control waveform.

The skilled person will appreciate that it is possible to use a feature described in relation to any one of the above aspects of the invention with any other aspects of the invention.

Brief description of the drawings

There now follows by way of example only a detailed description of the present invention with reference to the accompanying drawings in which

Figure 1 shows a typical arrangement of the components of an x- ray inspection system; and

Figure 2 shows a timing diagram for circuitry used to drive the photo-detector array shown in the earlier Figures. Detailed description of the drawings

Figure 1 shows a general arrangement of an x-ray inspection system 100. It is convenient to describe this invention in relation to X-ray inspection systems although it has wider applicability. The skilled person will readily appreciate how to make the necessary changes to allow the teachings to be modified for other forms of radiation.

Figure 1 is intended to put embodiments of the invention into context but may also be applicable to prior art systems. The system is intended to inspect materials and/or objects (hereinafter referred to as an object), during an on-line inspection process, to ensure that the inspected object is suitable and/or safe for its intended purpose. If the object were a foodstuff, or a pharmaceutical then the inspection may be to determine whether there are foreign bodies or voids therein, or an absence of product within the packaging. If the object is an item of baggage then the inspection may be to determine whether there banned goods in the baggage; for example to inspect baggage before an airline flight.

The system 100 comprises an x-ray source 102, providing a source of radiation, which is supplied from a high voltage power supply 104. The x-ray source is cooled by a cooler 106 to ensure that its temperature is maintained within an operating range. The power supply 104 and the cooler 106 are controlled by the processing circuitry within a controller 108 which is discussed hereinafter.

The x-rays produced by the x-ray source 102 are collimated, in a known manner, to provide a thin beam of x-rays of generally a fan shape 110 (where the fan shape exists in a plane perpendicular to the plane of the page) and typically having a width of roughly lmm. In Figure 1 the fan shape is viewed from one side and is represented by a row of dots.

A conveyor 112, having an upstream end 114 from which objects flow and a downstream end 116 to which objects flow, is provided and arranged to move an object 118 to be inspected through an irradiation zone 120 situated in a region below the x-ray source 102 and above an x- ray detector 122, which comprises a plurality of detector elements which are typically an array of photo-diodes. The conveyor 112 is shown as a belt conveyor but could be any other suitable form of mechanism arranged to transfer objects 118 through the irradiation zone 120, such a Bandolier or web conveyor mechanisms, gravity feed, free flight, or the like. It will be appreciated that if the direction of travel of the conveyor 112 is reversed then the upstream end 114 will become the downstream end 116 and vice versa.

It will be appreciated that other arrangements of the system 100 are equally possible. For example, the X-rays may travel in an upwards direction rather than the downwards shown in the Figures. In other embodiments, the X-rays may travel in a horizontal direction with the objects to be inspected moving in a vertical direction. The beam may also be oriented across the conveyor belt so as to image the object in a vertical direction - for example, looking through the sides of cans or jars on a conveyor belt.

Some conveyor mechanisms may use packaging of the object as the conveyor (such as in packaging of pharmaceuticals) . Other conveyor mechanisms may provide conduits for fluids or powders such as soups, or the like. In such an embodiment the fluid or powder is the object to be inspected. However, it is likely to still be desirable to inspect the content of objects carried by such transport mechanisms to ensure that the product is suitable and/or safe to be released.

The detector 122 is arranged to output data indicative of the amount of x- rays incident thereupon. The x-rays emitted from the source 102 generally pass through the object 118 when it is in the irradiation zone 120, but are attenuated by the object 118 according to its composition, and are then detected by the x-ray detector 122. The amount of x-rays received at a point along the detector (i.e. into or out of the page as viewed in Figure 1) give an indication of the composition of the object 118 at that point along the detector 122 at that point in time.

As the object 118 (which may be a fluid, powder, or packaging that should contain an object) is moved through the irradiation zone 120 by the conveyor 112 a two dimensional image of the object can be constructed from the data output from the detector 122. That is, the data output from the detector can be taken at predetermined intervals (typically roughly lms) and stitched together to form an image after suitable processing. In this embodiment, an output 124 from the detector 122 is processed by the processing circuitry of the controller 108 which generates a video display which is output to a display 126.

In some embodiments, the controller 108 may also perform other processing on the data output from the detector 122, for example to determine whether the product being scanned should be rejected by making an output on an Output reject mechanism 128. In such embodiments if the controller 108 determines that the object being scanned is below a predetermined standard (may be because it contains a foreign body above a predetermined size, it contains a void, a portion of the packaging is unfilled or the like) then it can cause a rejection mechanism to remove the object from the conveyor 112. Such rejection mechanisms are well known and will not be described further.

In some embodiments, the display 126 may be optionally omitted and the machine may perform automatic inspection of an object passing through the irradiation zone 120 without a visible image. During automatic inspection, if the controller 108 determines that a product falls outside acceptable criteria then the output to the reject mechanism 128 can be utilised to remove the product from the conveyor 112.

The processing circuitry of the controller 108 typically comprises a processor such as an Intel™ Pentium™, AMD™ Athlon™, IBM™ PowerPC™, or other such processor. However, in other embodiments the processing circuitry may also comprise dedicated electronics as provided by one or more Application Specific Integrated Circuits (or the like) . It is also possible that the detector output could be processed electronically by entirely analog means.

The processor is arranged to run code held in a memory accessible by the processor. The memory may or may not be provided within the system 100 and may be accessible over a network connection to the system 100. Further, it is likely that the memory comprises both a volatile portion (e.g. RAM) and a non-volatile portion (e.g. ROM, EPROM, a hard drive, or the like) .

The display 126 is typically a Liquid Crystal Display (LCD) but could be any other type of display such as a Cathode Ray Tube (CRT) display, a Light Emitting Polymer (LEP) display or the like. ^""-

Before the system 100 can be used in an on-line inspection process it should be calibrated using a calibration process. Such a calibration process (i.e. an offline operation) will usually make the output of the detector 122 uniform.

When a photodiode array is used, each individual diode within the system will have a different response to dark conditions, and exposure to X-rays. This is due to differences in internal 'dark current' noise, and varying gain in manufacture of individual diodes. This phenomenon is well known in the design of linear detectors.

In a first step of this calibration process, the processing circuitry of the controller 108 is caused to acquire a sample of the signal output from each diode within the detector 122 under dark (no x-rays) conditions. During subsequent on-line operation of the system this level will be subtracted from the output of each diode in order to obtain a uniform zero point for the diodes within the detector 122.

In a further calibration step the processing circuitry of the controller 108 is caused to sample the signal from diodes within the detector 122 under the operating X-ray dose level (i.e. with the X-ray source emitting the level of X-rays that will be emitted during use) . A correction factor is obtained that is used during online operation to create a signal level of uniform level for the diodes within the array. Typically, (but not limited to the range of figures mentioned) the data from the detector arrives at the processing circuitry as a 14 bit data stream comprising 0-16,383 levels. After correction, the processing circuitry normalises the 16,383 levels to give a corrected 0-255 linear 'greyscale' as an output which is used to drive the display 126 should one be used. It should be noted that the correction process need not be limited to correction over a 0-255 range, and may be output over other digital ranges, such as 0-1023, 0- 8192 or the like or as an analogue voltage and/or current. The 16,383 points relate to the full range of detector 122 sensitivity, before saturation (inability for the diodes to produce any further output) occurs. That is a level of 16,383 indicates that the diode is at maximum output. A system will not typically operate to provide a level of X-rays to provide as many as 16,383 points of signal.

Generally, the X-ray dose (i.e. the level of X-rays emitted by the source 102) is set to provide the correct level of penetration through the object 118 during the online inspection process. For example, for a low density object, such as a pouch of soup powder 5mm thick, a low level of X-rays (for example enough to generate 2,000 points of signal from the diodes) may be used. For objects of greater density, such as a frozen block of meat, much higher levels of X-rays are required, which will produce high levels of output during calibration time, or may even cause the array to enter saturation point, or be driven beyond this extent. (The detector 122 will receive the excess dose, but will cease to translate it to an increasing level of signal past the saturation point and in this example a level of 16,383 will be continuously output) .

It is desirable to run the detector 122 during calibration at the same energy level of the X-ray source 102 that will be used during the online inspection process. However, if the detector has reached saturation, with diodes levels at 16,383, these will not be representative of actual diode gain, and merely represent a saturated condition. In this state, the resultant 'gain' calculation will be meaningless, and the detector will run in operation in an uncorrected state.

For example,, if the object 118 that will be inspected are dense, such as blocks of frozen meat, it will be desirable to set the level of X-rays emitted from the system at a level which would cause the detector 122 to saturate during the calibration process. Dense objects passing through the inspection system tend to absorb a high percentage of the x-rays generated by the source and as such only a small percentage of the x-rays reach the detector. As such, the detector only operates over a small percentage of its useable range and detail may be lost if the detector is forced to operate during inspection at a lesser dose level in order to accommodate 'saturation free' calibration.

As a further example, a detector running at 'clear beam' (i.e. with no objects 118 present) with an x-ray dose from the X-ray source 102 may cause an output of 16,000 points from the detector 122. The same set up may only produce as little as 500 points of data from the detector 122 when a dense object 118 under online inspection is present (which using the above example, may be a frozen block of meat) .

An advantage can be gained by applying an excess of X-ray dose to the detector during online inspection, effectively driving it past its saturation point. If the dose is increased by a factor of four, the effective output of the array (disregarding the saturation effect) at 'clear beam' would be 4x 16,000 = 64,000 points (clipped at the detector output at 16,383) . After penetration of the dense object, however, the detector comes out of saturation, and provides a useable signal of 4x the original at 2,000 points in this example (was 500 points) .

There now follows a discussion in relation to Figure 2 in which embodiments of the present invention may employ in order to allow the detector 122 to be calibrated correctly and which helps to allow the outputs of the detector to make more use of the full range of outputs.

Each of the detector elements is generally a photo-diode with which there may be an associated scintillating layer of material (generally a strip of phosphorous) . This is well known in the art. Further, the photo-diodes are generally reversed biased so that they function as a charged coupled device: as x-rays hit the silicon device or scintillating layer light (or a direct electrical charge within the diode) is generated; the generated light causes charge to be stored in the photo- diode; the magnitude of the charge on any one diode is read at a predetermined interval; and after the level of charge is read the diode is reset so that the accumulated charge is removed therefrom. The level of charge, on any one photo-diode, read in this manner gives an indication of the amount of x-rays that were incident upon the scintillating material in a region above or within that photo-diode since the last reset of the diode (which may be referred to as during the current line scan time, or exposure period) . Thus, photo-diodes in the detector 122 are reset at regular intervals which are generally kept constant in order that the charge measured from the photo-diode is measured over a constant time period.

Figure 2a shows a suitable control waveform 200 for clocking the photo- diodes in the detector during use of the detector in a scanning process so that they are reset at appropriate times and also so that they cause a measurement to be taken at appropriate times. The waveform has a reset period T which comprises a low, reset, pulse 202 of period R which is used to reset the photo-diode and a high pulse of period C which allows charge to be accumulated on the photo diodes; the period C may be thought of as an exposure period 204. The output from the detector is generally read at an end region of this exposure period before the detector is reset. During the reset pulse 202 accumulated charge is removed from the photo-diodes in order that they can re-accumulate charge during the following exposure period 204.

It will be seen that the period T is substantially constant for the waveform 200 such that the edges of the reset pulse occur at a predetermined time. Further it will be seen that the relationship between the control waveform period T, the exposure period C and the reset pulse duration R is: T = C + R. C and R vary simultaneously in order to maintain T as substantially constant. Thus for a known substantially constant period T a reset pulse duration R may be calculated from a chosen or known value of C by evaluation of the expression: T-C, and likewise an exposure period C may be calculated from a chosen or known value of R by evaluation of the expression C = T-R.

In the embodiment being described calibration is achieved as follows. During calibration of the detector 122, the waveform 250 used to clock the detector 122 is modified to that shown in Figure 2b. This waveform 250 also has a period T, substantially the same as the waveform 200 shown in Figure 2a. Reset pulse 252 is longer when compared to the waveform 200 used during an online scanning process. Consequently, the exposure period C 254 during which x-rays can cause light to be emitted from the phosphor to be accumulated on the detector 122 is shorter when compared to the waveform 200. Thus, there is less time for the charge to build up on the photodiodes.

The new exposure period C may be chosen such that, during calibration, the photo-diodes in the detector no longer saturate when the system is running at clear beam at predetermined level of X-ray dose emitted from the X-ray source 102. It is perhaps desirable that the photodiodes more or less reach saturation as maximum x-rays are output in order that the greatest useable output is made from the detector 122.

Reducing the length of the exposure period C is initiated if saturation of the detector 122 outputs is detected during calibration. This might be determined if the output remains at a maximum level for more than a predetermined time. In this embodiment reduction is made in a stepwise manner, reducing the length of the exposure period 254 until saturation is no longer detected.

In a first mode of operation, once the calibration has been set the detector 122 is, during online use, clocked with the waveform 200 shown in Figure 2a, i.e. the exposure period C is returned to its pre-calibration level. As such, the detectors may now be thought of as running in an over exposed mode since the exposure period C in which x-rays are allowed to accumulate on the photodiodes is greater than during the calibration process. However, because the level of X-rays emitted from the source is constant between calibration and online use the calibration should still be valid. In the first mode of operation objects 118 passing through the irradiation zone 120 absorb a proportion of the x-rays emitted from the source 102 and the x-rays reaching the detector 122 are likely to fall within the calibration of the detector 122. This technique is perhaps most effective with substantially homogenous objects since if the density of the object reduces the output of the detector 122 will again be taken into saturation. Parts of the image which are under saturation will appear as 'peak white' on any display (i.e. maximum output) and will therefore not be resolved. Those parts of a dense object (where contaminants are more likely than not to be dense items) will be resolved as contaminants tend not to be the least dense parts of the image.

In a second mode of operation, once the calibration has been set the detector 122 is, during online use, clocked with the waveform 250 shown in Figure 2b which was used to calibrate the system; i.e. the exposure period C remains at that used during calibration. This second mode of operation may be particularly suitable for objects which do not absorb a significant amount of the X-rays emitted from the source 102 such as, using the example given above, thin packets of soup. Thus, the second mode of operation helps to prevent the detector 122 from becoming saturated by keeping the exposure period C at the shortened period used for calibration.

Providing a specific example of the lengthening of the reset pulse for calibration, it is assumed that the example scan rate of lms (i.e. period T is set to lms (with a reset pulse of 12μS and an exposure period C of 988μS) results in a saturated detector 122. In this embodiment, during calibration, the exposure period 254 is reduced in a stepwise manner reducing in steps of 10% (or other incremental amounts) until saturation no longer occurs. In the example being given this occurs when the exposure period C is reduced to 0.1m/s; i.e. 1/lOth of its previous value. Thus, charge accumulated on the photo diodes is 1/lOth of the previous value and the output is no longer saturated. If a level of 1/lOth were not appropriate (i.e. there was either not enough charge accumulating on the photodiodes or still too much charge was accumulating and the outputs were still saturated) then it would be possible to choose other ratios.

Claims

1. A method of controlling an inspection system which generates radiation from a source and allows the radiation to be detected by a detector during an exposure period of a control waveform used to drive the detector, the control waveform comprising a reset pulse in addition to the exposure period, the reset pulse having a nominal duration and the method comprising calibrating the inspection system by increasing the relative length of the reset pulse, to a calibration duration, when compared to the exposure period whilst maintaining a substantially constant period of the control waveform.

2. A method according to claim 1 in which the radiation used is X- rays.

3. A method according to claim 1 or 2 in which the reset pulse is reduced in length, when compared to the calibration duration used to calibrate the system, during online inspection of articles.

4. A method according to claim 3 in which the reset pulse is reduced in length to have, or to be close to having the nominal duration.

5. A method according to claim 4 in which the nominal duration is substantially equal to its minimum duration.

6. A method according to claim 1 in which the reset pulse remains substantially at the calibration duration during online use of the system to inspect articles.

7. A method according to any preceding claim in which the duration of the reset pulse is increased once it has been detected that saturation of the detector is occurring.

8. A method according to claim 7 in which the duration of the reset pulse is increased until saturation of the detector no longer occurs.

9. A method according to claim 7 or 8 in which the duration of the reset pulse is increased in a stepwise manner.

10. A method according to claim 9 in which the duration of the reset pulse is increased in steps of roughly 5%.

11. A method according to claim 7 or 8 in which the duration of the reset pulse is reduced to a predetermined level once saturation of the detector has been detected.

12. An inspection system comprising: a source of radiation; a timer arranged to generate a control waveform comprising a reset pulse, having a nominal period, and an exposure period; a detector arranged, in use, to allow radiation to be detected during the exposure period of the control waveform; and processing circuitry arranged to allow the control waveform to be varied such that the length of the reset pulse is increased, during calibration of the system, beyond the nominal period whilst maintaining a substantially constant period of the control waveform.

13. A system according to claim 12 in which the source of radiation is a source of X-rays.

14. A system according to claim 12 or 13 in which the detector comprises one or more, and generally a plurality of, photodiodes.

15. A system according to claim 14 in which the photodiodes are arranged in a linear array.

16. A system according to any of claims 12 to 15 which is arranged such that the amount by which the duration of the reset pulse can be increased can be varied.

17. A machine readable medium containing instructions which when read by a machine cause that machine to perform the method of any of claims 1 to 11.

18. A machine readable medium containing instructions which when read by a machine cause that machine to function as the machine of any of claims 12 to 16.