DE102004060608A1 - Compton scattering quantum attenuation correction method, involves producing common matrix at corrected scattering coefficients, where total correction of scattering signals results during matrix production - Google Patents

Abstract

The method involves dividing current examined high beam of an object into a two-dimensional matrix at scattering voxels (5) with an assigned dispersion coefficient. Attenuation factors are calculated for a primary beam (13) and a secondary beam of a Compton scattering quantum (6). A common matrix is produced at corrected scattering coefficients, where the total correction of scattering signals results during matrix production.

Description

The Invention concerned with a method for correcting the attenuation of Compton scattering quanta in an arrangement for measuring Compton scattering quanta.

It arrangements are known in which an X-ray source is used, which has a linearly extended focus and its fan-shaped radiation converges to a single focal point. There is also a Compton scatter collimator disposed between the object and the Compton detector, the one Streuvoxel in the object, which has a certain depth, a position at the two-dimensional Compton detector assigns.

Out The prior art also arrangements are known in two successive steps - usually in different Arrangements - a Baggage on dangerous Ingredients, especially explosives, examine. In the one Arrangement is doing the density information within the bag by means of Compton scattering determined and in the other arrangement, the diffraction information by means of coherent Scattering. When measuring the respective density or diffraction information However, false alarms occur because the known measuring methods and Detection methods make assumptions that lead to uncorrected results. A Correction is urgently needed since each passing through a piece of luggage X-ray a self-weakening in the material subject, without their consideration it to falsifications the result of the measurement comes.

task The invention is therefore to present a correction method, which the weakening both of the primary beam - before Impact on the Streuvoxel - as also the secondary weakening of the Scatter quantum determined, allowing a correction of the measured values obtained can be made, which leads to a reduction of false alarms leads.

The The object is achieved by a method having the features of the patent claim 1 solved. The weakening of the detected X-ray quantum within the object to be examined can best be determined by the dependent on the density Compton scattering can be determined. Therefore, according to the invention, that even with an arrangement in which both the Compton scattering quanta and also the elastically scattered scattering quanta can be detected, the Correction based on the data for the Compton scattering quantum occurs. According to the invention, a decomposition takes place the disk, which is interspersed by the inverse fan-shaped x-ray beam, into a two-dimensional matrix of scatter voxels. Compton scattering quanta those from the topmost layer, d. H. the detector-near layer, the Object, have not experienced secondary attenuation, but merely a primary weakening. After this for all Streuvoxel of the top row the respective scattering coefficients are determined, these are for taking into account the underlying second set of scatter voxels secondary attenuation by the above lying row of Streuvoxeln determined. After in the second Complete series the corrected scattering coefficients have been determined, becomes the next row advanced and repeated the process of the invention. This It continues until you reach the bottom row of Streuvoxeln, the the X-ray source the next lying, comes. After these have been corrected, is the entire matrix of corrected scattering coefficients for the straight examined slice before, so that a total correction for all measured Scattering quantum can be made, this not only applies to the Compton scattering quanta, but also by possibly in the same order at the same time measured coherent Scatter quanta. By the correction thus made, the false alarm rate can be to press, because the information obtained does not erroneously refer to any other content of the Close the object, when he actually exists.

Advantageously, the relationship between a scattering coefficient and the associated scatter signal is:

Here, F _s (i, j) are the attenuation coefficient of the secondary beam and F _P (i, j) are the attenuation coefficient of the primary beam 13 for each associated Streuvoxel 5 , K is a system constant and M indicates multiple scattering. As primary beam 13 the X-ray quanta become scattered at the respective scatter voxels 5 designated. As a secondary beam, the Compton scattering quanta 6 referred to by the respective Streuvoxel 5 out.

A further advantageous embodiment of the invention provides that for the Streuvoxel 5 in the top row, the attenuation coefficient F _{P of} the primary beam 13 is equal to the value T _{R1 obtained in} each case in the transmission detector and the attenuation coefficient of the secondary beam F _{s is set} equal to 1. This provides a very good approximation as the basis for the first set of scatter voxels 5 so that the subsequent determination of the further attenuation coefficients in the underlying rows of scatter voxels 5 has the smallest possible error. It is preferred that for the Streuvoxel 5 in the second row, the associated attenuation coefficient of the first row is multiplied by exp (σl _P ), where σ is the associated scattering coefficient of the respective scatter voxel 5 the first row and lp the path length of the primary beam 13 in the associated scatter voxel 5 is the first row and at the same time a multiplication of the attenuation coefficient of the secondary beam with exp (-σ'l _s ), where l _{s is} the path length of the secondary beam in the associated Streuvoxel 5 the first row and σ 'the Compton scattering coefficient at an average energy of the Compton scattering quanta 6 is.

A Further advantageous development of the invention provides that at the same time the coherent scattered radiation with a coherent Scattering detector is measured by passing between the examination area and the focal point another scatter collimator is arranged, the only coherent scattered radiation from an insertable into the examination area object, the is emitted at a fixed scattering angle. By the combination the measurement of the pulse transmission spectrum of elastically scattered X-ray quanta the measurement of the density profile (the image of the electron density) by means of Compton scattering will be for each scatter voxel has two material-specific parameters simultaneously be determined. Even if one of the two parameters is an uncertainty of the material in the streuvoxel may be due to the other Parameters in almost all cases the actual Nature of the material contained in the investigated Streuvoxel closed become. This significantly reduces the false alarm rate without the determination of the two parameters in different process steps or even spatially made separate equipment would have to which saves considerable time.

Further advantageous developments of the invention are the subject of the dependent claims. An advantageous Embodiment of the invention will be described with reference to the figures embodiment explained in more detail. in this connection demonstrate:

1 a schematic view of an arrangement without scattering collimators, with the method of the invention can be carried out and

2 a schematic representation of two beam paths, with reference to the system of the method according to the invention can be seen.

In 1 schematically an arrangement is shown, by means of which the inventive method can be performed. This has an anode extending in the Y direction 1 which has a series of juxtaposed discrete focus points when bombarded with an electron beam along the anode 1 hike. The X-ray quanta emanating from each individual focus point are passed through a primary collimator 2 , which has a fan shape, so limited that an inverse fan beam 3 to X-ray quanta as a primary beam 13 results. This inverse fan beam 3 is in the XY plane (the X-axis is not shown for clarity, but since it is a Cartesian coordinate system, it is perpendicular to the Y-axis and the Z-axis.) The inverse fan beam 3 converges to a single focal point 4 , which simultaneously forms the coordinate origin of the Cartesian coordinate system. The inverse fan beam 3 encounters an object in the object space, which for reasons of clarity only uses a single scatter voxel 5 is shown. This object is usually a suitcase in the most common application case of a luggage inspection system. This is then on a (not shown) conveyor belt, which is parallel to the Z-axis along the direction of movement 12 emotional. The inverse fan beam 3 penetrates the object along a thin slice in the XY plane. This disc is changed by a one-dimensional movement of the conveyor belt in the direction of movement 12 takes place, so that a complete scanning of the object can be done by the movement of the conveyor belt.

From the currently scanned thin disc is in 1 only the scatter voxel 5 shown. This is located at a predetermined depth along the X direction in the object.

The primary beam 13 gets at this scatter voxel 5 for a coherent scattered. Due to a first scattering collimator (not shown), only such coherent scattering quanta become 10 to a coherent scatter detector 11 let through, the under a fixed predetermined angle of scatter Θ scattered to the XY plane. The scatter collimator for coherent scattering quanta 10 is between the examination area, ie the Streuvoxel 5 , and the coherent scatter detector 11 arranged. These may, for example, be parallel metal plates that represent the X coordinate of the scatter voxel 5 to the Z coordinate of the coherent scatter detector 11 make. Such scattering collimators are known from the prior art and are not essential to the invention, so that their details are not discussed in more detail below.

The one from the Streuvoxel 6 not scattered trans mission brilliant 18 encounters a third, spatially resolving transmission detector 17 located on the X-axis between the examination area and the focal point 4 is arranged and extends in the Z direction. Hereby, the projection information can be evaluated in transmission and explosives are well detected. Likewise, a good shielding, for example by lead plates, can be detected. Finally, the attenuation correction can be calculated well.

The information from the transmission detector 17 is necessary to set the Compton scattering coefficients - as below 2 described - to be able to calculate.

In addition to the transmission quanta 14 that from the transmission detector 17 and the coherent scattering quanta 10 in the coherent scatter detector 11 be detected, go from the Streuvoxel 5 also Compton scattering quanta 6 out. These are based on a Compton scattering collimator, not shown on a Compton detector 7 displayed. As the Compton scattering collimator, for example, a plate of a high X-ray absorbing material such as lead, which has a slit parallel to the Y axis, may be used. This turns the X coordinate of each scatter voxel 5 to an associated unique Z coordinate in the Compton detector 7 displayed. In a slot-shaped Compton scatter collimator, the highest scattering voxel becomes 5 , ie closest to the detectors, with the highest absolute Z coordinate. It also follows that the lowest-lying scatter voxel 5 - that is, that which is closest to the X-ray source - is mapped to the smallest absolute Z coordinate, that is the Y axis closest to the Compton detector 7 incident.

The Compton detector 7 has strip-shaped detector elements parallel to the Y-axis 8th on. This causes the solid angle, below which the Compton scattering quanta 6 from the strevox 5 are detected, increased and there are more Compton scattering quanta 6 which leads to a better signal-to-noise ratio.

The Compton scattering quanta 6 be used for determining the density in the scatter voxel 5 used as a close link with the electron density in Streuvoxel 5 given is. In contrast, the coherent scattering quanta 10 measured the pulse transmission spectrum with high accuracy.

Both the Compton detector 7 as well as the coherent scatter detector 11 are symmetrical to an axis of symmetry 9 , which corresponds to the Z-axis formed. In addition, it is also possible to perform a symmetrical training with respect to the Y-axis, so that in each case two coherent scattering detectors 11 with associated scatter collimators and two Compton detectors 7 with associated scattering collimators are present. This is possible because the momentum transfer function at a scattering angle Θ of about 5 ° for the explosives in question, which are to be detected, have a strong peak at the usual x-ray energies resulting in larger scattering angles Θ decreases sharply. In contrast, it is for the detection of Compton scattering quanta 6 advantageous if at the given X-ray energies Compton scattering angle Γ of about 10 ° are detected. The symmetrical arrangement to the Y-axis gives an even better signal-to-noise ratio, since twice the number of Compton scattering quanta 6 or coherent scattering quanta 10 be detected.

Based on 2 can very well the basic inventive method for correcting the self-attenuation within the object in the by the inverse fan beam 3 illuminated disc can be explained.

The examined slice of the object becomes a two-dimensional matrix of scatter voxels 5 decomposed with associated Compton scattering coefficients σ _ij . Where i stands for the row and j for the column. The X coordinate is correlated with the line number i. This is - as already stated above - based on the Compton secondary collimator on the respective strip-shaped detector element 8th of the Compton detector 7 displayed. The information about the column j of the Streuvoxel 5 is determined by the primary beam angle α, the primary beam 13 to the X-axis, uniquely determined in connection with the line information i. Thus, for each measurement of Compton scattering quanta 6 the associated Streuvoxel 5 and thus the corresponding Compton scattering coefficient σ _ij in the two-dimensional matrix can be determined.

Depending on the location of Streuvoxels 5 learns the primary beam 13 to the Streuvoxel 5 a weakening within the object and beyond is experienced by the Compton scattering quantum 6 after the strevox 5 in the remaining part of the object a further weakening. This implies that the measured Compton scattering coefficient of a voxel 5 not the actual one. If there is no correction of the Compton scattering coefficient σ _ij , then a false alarm may possibly be obtained, since a false density of the material in the scatter voxel 5 which may indicate the presence of explosive, although there is another density which does not belong to an explosive. To reduce the false alarm rate, a correction according to the invention of the Compton scattering coefficients σ _ij is therefore necessary.

According to the invention, the procedure is such that the uppermost row of scattering coefficients σ _{1j is} first determined. For this purpose, the weakening in the object can be determined very easily. Since the Compton scattering quanta 6 when leaving the Streuvoxels 5 no longer pass through the object, there is no secondary attenuation, but only a primary attenuation of the primary beam 13 along its trajectory through the object. Because the transmission beam 15 from the transmission detector in the focal point 4 is measured and the energy of the X-ray tube is known, the primary attenuation can be determined very easily. Thus, for the entire upper row of scattering coefficients σ _1j, the correct value is obtained after performing the aforementioned correction.

erhält man die korrigierten Streukoeffizienten σ_1j, indem man für F_P das Transmissionssignal einsetzt und für F_s den Wert 1, da keine Sekundärschwächung vorliegt. Die beiden Konstanten M für die Mehrfachstreuung sowie K als Systemkonstante sind aus Referenzmessungen bestimmt worden.From the above equation

One obtains the corrected scattering coefficients σ _1j by using the transmission _signal for F _P and the value 1 for F _s , since there is no secondary attenuation. The two constants M for the multiple scattering and K as the system constant have been determined from reference measurements.

Assuming these corrected Compton scattering coefficients σ _{1j of} the first series, one can calculate the Compton scattering coefficients σ _{2j of} the second series. Because in the second row of the primary beam 13 passes through less material, the attenuation coefficient F _{P of} the primary beam 13 multiplied by the factor exp (σl _P ), where σ is the Compton scattering coefficient of the scatter voxel 5 in the overlying row and l _{p is} the path length of the primary beam 13 represents in this cell. This is necessary as the primary beam 13 no longer the full length to the top row passes through the object, but only up to the second row. For the second series, there is an additional weakening of the Compton scattering quanta 6 because they still have to pass through the top row. The attenuation coefficient F _{s of} the secondary beam is therefore multiplied by the factor exp (-σ'l _s ). Here, l _{s means} the path length of the Compton scattering quantum 6 in the overlying scatter voxel 5 and σ 'the Compton scattering coefficient at the average energy of the Compton scattering quanta 6 ,

However, it should be noted that this is an approximation, for which it is assumed that the scattering coefficient in the direction of the Z-axis along the trajectory of the Compton scattering quantum 6 does not change. This must be done because the Compton scatter quantum 6 is scattered out of the illustrated XY plane and thus not by the overlying Streuvoxel 5 - whose scattering coefficient is just used for the calculation - occurs, but by a lying behind or before - so moved in the Z direction - Streuvoxel 5 , Because the Streuvoxel 5 however, if they are very small and the Compton scattering angles Γ are also not large, this assumption leads to no noteworthy errors in the determination of the corrected Compton scattering coefficients σ _2j .

After the whole Compton scattering coefficients σ _{2j of} the second row have been corrected, they are again used as a starting point for the correction of the Compton scattering coefficients σ _{3j of} the underlying row. This iterative process continues until the Compton scattering coefficients of the lowest row of scatter voxels 5 have been corrected.

The inventive method can even be extended to the system constant K variable is made and of an inconsistent distribution of multiple scattering M is assumed. This preserves even better corrections and even lower false alarm rates.

1

anode

2

primary collimator

3

inverse fan beam

4

focal point

5

voxel

6

Compton scattering Quant

7

Compton detector

8th

detector element

9

axis of symmetry

10

coherent scatter quantum

11

coherent scatter detector

12

movement direction

13

primary beam

14

transmission Quant

15

transmission beam

16

transmission detector

17

conveyor belt

Θ

coherent scattering angle

Γ

Compton scattering angle

α

Primary beam angle

Claims (6)

Method for correcting the attenuation of Compton scattering quanta ( 6 ) in an arrangement for measuring Compton scattering quanta ( 6 ) and transmission quanta ( 14 ) passing through an object using an X-ray source having a linearly extended focus and its inverse fan beam (FIG. 3 ) on a single focal point ( 4 converges; Also, a Compton scatter collimator is between the object and one Compton detector ( 7 ), which is a Streuvoxel ( 5 ) in the object, which has a certain depth, an assigned position on the two-dimensional Compton detector ( 7 ) assigns; Furthermore, a transmission detector ( 16 ) between the focal point ( 4 ) and the object on the X-axis and extends over the inverse fan beam ( 3 ) in the Y direction, with the following steps: - decomposing the currently examined slice of the object into a two-dimensional matrix of scatter voxels ( 5 ) each having an associated scattering coefficient; - calculation of the attenuation factor for the primary beam ( 13 ), which reaches the Streuvoxel ( 5 ) occurs; Calculation of the attenuation factor for the secondary beam of Compton scattering quanta ( 6 ), which after Streuvoxel ( 5 ) occurs; - starting with the series of scatter voxels ( 5 ) farthest from the X-ray source and correction of the respective associated scattering coefficient; Further correction of each row of scatter voxels following each preceding row 5 ) taking into account the corrected scattering coefficients associated with the preceding row; - Preservation of the entire matrix of corrected scattering coefficients, from which the overall correction of the scattering signal results. A method according to claim 1, characterized in that the relationship between a scattering coefficient and the associated scattering signal is:

where F _s (i, j) is the attenuation coefficient of the secondary beam and F _P (i, j) is the attenuation coefficient of the primary beam ( 13 ) for the respectively assigned Streuvoxel ( 5 ), K is a system constant and M indicates the multiple scatter. Method according to claim 2, characterized in that for the Streuvoxel ( 5 ) in the uppermost row the attenuation coefficient F _{P of} the primary beam ( 13 ) is equal to the value T _{R1 obtained in} each case in the transmission detector and the attenuation coefficient F _{s of} the secondary beam is set equal to 1. Method according to claim 3, characterized in that for the Streuvoxel ( 5 ) in the second row the associated attenuation coefficient F _{P of} the primary beam ( 13 ) of the first row is multiplied by exp (σl _P ), where σ is the assigned scattering coefficient of the first row and lp is the path length of the primary beam ( 13 ) in the associated scatter voxel ( 5 ) of the first row and at the same time a multiplication of the attenuation coefficient F _{s of} the secondary beam with exp (-σ'l _s ) takes place, where l _{s is} the path length of the secondary beam in the associated scatter voxel ( 5 ) of the first series and σ 'the Compton scattering coefficient at an average energy of the Compton scattering quanta ( 6 ). Method according to claim 4, characterized in that for each further series of scatter voxels ( 5 ) an iteration according to the preceding claim takes place. Method according to one of the preceding claims, characterized in that at the same time the coherent scattered radiation ( 10 ) with a coherent scatter detector ( 11 ) is measured by between the examination area and the focal point ( 7 ) a further scatter collimator is arranged, the only coherent scattered radiation ( 10 ) passes through an object that can be introduced into the examination area and that is emitted at a fixed scattering angle (Θ).