JP2009008467A - Method of evaluating x-ray baggage inspection device, device for evaluating x-ray baggage inspection device, and x-ray baggage inspection device evaluation program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To numerically evaluate a generation amount of image distortion of an X-ray baggage inspection device, and to compare the difference in distortion generation amount of the X-ray baggage inspection device which is different for every maker. <P>SOLUTION: A metal test chart frame 200/300 for evaluating distortion having a clear dimension shape is photographed by the X-ray baggage inspection device 400. A measuring device 100 measures numerically the degree of distortion with which the photographed image is photographed relative to an actual shape. Thus, the image distortion amount is clarified, and a distortion measurement result is evaluated. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an evaluation method for an X-ray luggage inspection apparatus, an evaluation apparatus for an X-ray luggage inspection apparatus, and an X-ray luggage inspection apparatus evaluation program, and evaluates the amount of distortion of an image taken by the X-ray luggage inspection apparatus. Regarding technology.

  The image taken with the X-ray baggage inspection apparatus is distorted. This can cause missed inspections and misjudgments for airport luggage inspectors. As one of the techniques for solving the image distortion of the X-ray baggage inspection apparatus, there is Patent Document 1. Since the technique of Patent Document 1 corrects the data of the L-shaped line sensor so that it is on a straight line perpendicular to the X-ray emission direction, image distortion can be greatly reduced.

  However, X-ray inspection apparatuses that are actually used for inspection of luggage at airports include apparatuses that are corrected on a straight line and apparatuses that are presumed not to be devised to reduce distortion. This will be described with reference to FIG. Here, FIG. 1 is a diagram summarizing the distortion correction method and correction characteristics of the X-ray inspection apparatus. In FIG. 1, the A model has an X-ray sensor arranged in an inverted L shape on the monitor screen side wall and the top, and X-rays are emitted from the back side of the tunnel (bright spot position: back side of the screen). The data of the line sensor is corrected so as to be on a straight line perpendicular to the X-ray emission direction. In the B model, X-ray sensors are arranged in an inverted L-shape on the side wall and the top of the monitor screen, and X-rays are emitted from the front side of the tunnel (bright spot position: in front of the screen). Is corrected to be on a straight line perpendicular to the X-ray emission direction.

  On the other hand, the C model has the same arrangement of the X-ray sensor and X-ray tube as the B model, but is corrected to the top projection. Furthermore, the D model is an apparatus having the same arrangement of the X-ray sensor and the X-ray tube as the C model and the B model, but is not subjected to X-ray sensor linear correction. In any model, the X-ray sensor is arranged in an inverted L shape so as not to increase the size of the X-ray inspection apparatus.

  Here, the correction features of the A model and the B model provide an image with a little discomfort even in a stereoscopic image. In addition, the correction feature of the C model is less uncomfortable with the planar image. On the other hand, the D model without correction is an image with a lot of uncomfortable feeling even in a flat object.

  At airports, X-ray inspection apparatuses of different A model or D model are used depending on the gate. Also, inspectors are not always located at the same gate. Therefore, the inspector needs to inspect while considering how the image of the X-ray inspection apparatus is seen. However, inadequate correction may cause the inspector to miss an inspection or make an erroneous determination. Specifically, when a knife or the like that is recognized as being straight is viewed as a bent image, there is a high risk of overlooking.

  With reference to FIG. 2, the mechanism of image distortion will be described. Here, FIG. 2A is a diagram for explaining the arrangement of the X-ray source, the inspection object, and the X-ray sensor. FIG. 2B is a diagram for explaining distortion in the horizontal X-ray sensor. FIG. 2C is a diagram for explaining distortion in the vertical X-ray sensor. FIG. 2D is a diagram for explaining distortion due to the inclination between the inspection object and the X-ray emission direction.

  The relationship between the X-ray source 423, the X-ray sensors 421 and 422 of the models B to D shown in FIG. Note that the baggage assuming the briefcase is placed horizontally, and the X-ray emission direction of the X-ray source 423 has an angle of 30 ° with the vertical direction. Also, the left in FIG. 2A is referred to as the front, and the right is referred to as the back.

  In FIG. 2B, when the baggage indicated by the arrow is placed perpendicular to the X-ray emission direction, the projected image is projected onto the horizontal X-ray sensor 421 disposed above the X-ray inspection apparatus. Is an image with the front side contracted and the back side extended.

  In FIG. 2C, when the baggage indicated by the arrow is placed perpendicular to the X-ray emission direction, it is projected onto the vertical X-ray sensor 422 disposed at the back of the X-ray inspection apparatus. The image is an image in which the front side is extended and the back side is contracted.

  In FIGS. 2B and 2C, it is assumed that the baggage indicated by the arrow is placed perpendicular to the X-ray emission direction, but the baggage and the X-ray emission direction are actually inclined by 30 °. Therefore, as shown in FIG. 2 (d), even if the screen is converted to a screen perpendicular to the X-ray emission direction, the front side is extended and the back side is reduced.

  In FIG. 1, the D model is an image in which the distortions in FIGS. 2B to 2D are superimposed. The C model is an image in which the distortion of FIG. 2B and the distortion of FIG. Furthermore, the distortion slightly recognized in the A model and the C model is due to the explanation of FIG. From these explanations, it can be seen that the performance is qualitatively higher in the order of A model / B model> C model> D model. However, since the manufacturer of the X-ray inspection apparatus does not clarify what kind of correction is applied, it is also taken as a subjective judgment from the image.

JP 63-079043 A

  The present invention provides an evaluation method for an X-ray baggage inspection apparatus, an evaluation apparatus for an X-ray baggage inspection apparatus, and an X-ray baggage inspection apparatus evaluation program for quantitatively evaluating the amount of image distortion generated by different X-ray baggage inspection apparatuses. is there.

  The above-described problems are a step of flowing a distortion evaluation test chart consisting of a plurality of sides to the X-ray baggage inspection apparatus, a step of measuring the length of a plurality of sides from a transmission image of the distortion evaluation test chart output by the X-ray baggage inspection apparatus, and Calculating the average reduction rate of the X-ray luggage inspection apparatus from the measured lengths and the actual dimensions of the sides, multiplying the actual dimensions and the average reduction rate, This can be achieved by an evaluation method for an X-ray baggage inspection apparatus comprising a step of calculating a difference and a step of calculating an evaluation value from a plurality of differences.

  An image processing unit that measures the length of a plurality of sides from a transmission image of a strain evaluation test chart obtained by flowing a strain evaluation test chart consisting of a plurality of sides through an X-ray luggage inspection apparatus, and a plurality of measured lengths This can be achieved by an evaluation apparatus for an X-ray baggage inspection apparatus, which includes a distortion calculation unit that calculates an evaluation value of image distortion by the X-ray baggage inspection apparatus based on the actual dimensions of a plurality of sides.

  Furthermore, an image processing unit that measures the length of a plurality of sides from a transmission image of a distortion evaluation test chart obtained by flowing a distortion evaluation test chart consisting of a plurality of sides through an X-ray luggage inspection apparatus, a plurality of measured This can be achieved by an X-ray baggage inspection apparatus evaluation program for functioning as a distortion calculation unit that calculates a distortion evaluation value of an image of the X-ray baggage inspection apparatus based on the length and the actual dimensions of a plurality of sides.

  By numerically evaluating the amount of distortion generated in the image, the inspector can select an X-ray baggage inspection apparatus with a smaller amount of distortion generation, thereby reducing the risk of missed inspections and erroneous determination.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings using examples. Note that substantially the same parts are denoted by the same reference numerals, and description thereof will not be repeated.

  The embodiment will be described with reference to FIGS. Here, FIG. 3 is a perspective view for explaining the X-ray luggage inspection apparatus and its evaluation apparatus. FIG. 4 is a hardware block diagram for explaining an evaluation apparatus for the package X inspection apparatus. FIG. 5 is a functional block diagram for explaining the evaluation device of the package X inspection device. FIG. 6 is a diagram for explaining the test chart frame. FIG. 7 is a diagram for explaining a flat test chart frame. FIG. 8 is a diagram for explaining dimension measurement from an X-ray image. FIG. 9 is a diagram for explaining the extraction of the distortion error from the actual size and scale ratio of the test frame chart. FIG. 10 is a flowchart for explaining a procedure for extracting a distortion error.

  3, the X-ray baggage inspection apparatus 400 includes a front slider 410, a rear slider 430, an X-ray inspection apparatus body 420 sandwiched between the front slider 410 and the rear slider 430, and an X-ray baggage inspection imaging result display 440. It consists of. In the X-ray inspection apparatus main body 420, an X-ray source (not shown), an X-ray sensor arranged in an inverted L shape, and a conveyor for moving a load are arranged. The X-ray inspection apparatus main body 420 generates a two-dimensional image by moving the baggage by the conveyor and displays it on the X-ray baggage inspection imaging result display 440. Usually, for the baggage (not shown) inserted from the front slider 410, the inspector inspects the baggage while looking at the image of the X-ray baggage inspection imaging result display 440.

  The evaluation apparatus 100 of the X-ray luggage inspection apparatus includes a control device 110, a communication interface 120, a display display 130, and a LAN cable 140. The control device 110 is connected to the X-ray baggage inspection apparatus main body 420 via a LAN cable 140 that transmits an imaging result image. The control device 110 is an evaluation device body that automatically measures the numerical value of how much the imaging result of the X-ray baggage inspection device 400 is imaged with respect to the actual object.

  The test frame chart 200 placed on the front slider 410 of the X-ray luggage inspection apparatus 400 is used for grasping the image distortion of the X-ray luggage inspection apparatus 400. Specifically, the test frame chart 200 is inserted into the X-ray inspection apparatus main body 420 and the obtained image is evaluated by the evaluation apparatus 100.

  With reference to FIG. 4, a hardware block for explaining the evaluation apparatus for the package X inspection apparatus will be described. 4, the evaluation apparatus 100 includes a control device 110, an IEEE 802 IFU (Interface Unit) 120 that is a communication interface, a display display 130, and a LAN cable connected to the IFU 120, as described with reference to FIG. 140. The control device 110 also includes an MPU (Micro Processor Unit) 112, an MMU (Main Memory Unit) 113, a GDU (Graphic Display Unit) 114, a DCU (Disk Controller Unit) 115, and a DCU 115 connected to the bus 111. And a hard disk 116 connected to the. The IFU 120 is illustrated as a part of the control device 110 in FIG. A display display 130 is connected to the GDU 114.

  Here, the control device 110 is a computer, and operates as a part of the evaluation device 100 of the package inspection device when the MPU 112 executes a program on the MMU 113 loaded from the hard disk 116. In FIG. 5, the control device 110 includes an image processing unit 117 and a distortion calculation unit 118. Both the image processing unit 117 and the distortion calculation unit 118 are realized by executing a program.

  A distortion evaluation test chart frame will be described with reference to FIGS. FIG. 6A is a perspective view of a rectangular parallelepiped distortion evaluation test chart frame. FIG. 6B is a plan view of parts for assembly. FIG. 7 is a plan view of a planar shape distortion evaluation test chart frame.

  In FIG. 6, a rectangular parallelepiped distortion evaluation test chart frame 200 is assembled by screwing three types of metal square bars 210, 220, and 230 each having four lengths, using a total of 12 bars. . When the three types have the same length, a cube distortion evaluation test chart frame is obtained. Further, it is desirable that the type of metal is a 16 mm square aluminum material because it is lightweight. Here, the aluminum material is used because aluminum has a high X-ray transmittance, and the coloration in the X-ray inspection apparatus is intermediate between the organic substance and iron.

  In FIG. 7, the flat-shaped distortion measurement test chart frame 300 is assembled by screwing a thin acrylic plate using four aluminum bars of three different lengths, four in total.

  As described above, not only the flat-shaped distortion measurement test chart frame 300 but also the rectangular parallelepiped distortion-evaluation test chart frame 200 is used for the X-ray inspection apparatus having a good result in the flat-shaped distortion measurement test chart frame 300. However, it is because the test chart frame 200 for distortion evaluation having a rectangular parallelepiped shape is not always good.

  With reference to FIG. 8, an image processing procedure for extracting a dimension measurement part from an X-ray image in the image processing unit and measuring the dimension will be described. FIG. 8A (a) shows a template image 501 for searching for a photographed image. The image processing unit 117 searches for a photographed image by pattern matching image processing. In FIG. 8A (b), the operator designates a portion to be measured from the template image 502 using the mask pattern 511.

  FIG. 8A (c) is a result image 503 obtained by photographing a distortion evaluation test chart frame with the X-ray luggage inspection apparatus. The image processing unit 117 performs an image matching calculation process in the pattern matching image process between the template image 501 and the test result frame shooting result image 503. The calculation result is shown in FIG. 8A (d). The image processing unit 117 displays it as a matched image 504. In the photographed result image 504, reference numeral 512 denotes a mask pattern corresponding portion of an image that is matched by pattern matching image processing. The image processing unit 117 cuts out a mask pattern corresponding portion that matches the image in the photographing result image 504 by an AND operation, and sets it as a length measurement target partial image 505 shown in FIG. 8A (e). The image processing unit 117 performs thinning image processing as the next stage in order to measure the length from the cut length measurement target partial image 505. The thinned image processing result image is a thinned image 506 shown in FIG. 8B (f).

  In the thinned image 506, a shadow corresponding to the width of the mask pattern 512 remains. This remaining beard needs to be removed. The image processing unit 117 removes unnecessary whiskers and leaves only the necessary length measurement line segments in the direction code extraction calculation process. The result of removing unnecessary whiskers is a length measurement line segment image 507 in FIG. 8B (g). Since the length measurement line segment image 507 has an inclination, it is necessary to calculate the angle of the inclination and rotate the image. The image processing unit 117 rotates the length measurement line segment image 507 to obtain a horizontal image 508 in FIG. 8B (h). In order to measure the length of the line segment from the horizontal image 508, coordinate extraction processing by histogram image processing is necessary. A histogram result of the histogram processed image 509 by the image processing unit 117 is indicated by reference numeral 510.

  With the series of image processing described above, the measurement of the portion of the result image obtained by photographing the distortion evaluation test chart frame to be measured and the portion to be measured designated by the mask pattern is completed. Similarly, by preparing a plurality of template images in which the positions of the mask patterns are changed, a plurality of parts are measured. In general, four different points in a plane substantially perpendicular to the X-ray emission direction are measured.

  With reference to FIG. 9, extraction of distortion errors in an X-ray image will be described based on the actual dimensions and scale ratio of a distortion evaluation test chart frame. The actual dimensions L1, L2, and L3 (= L1) of three frames made of metal bars of the distortion evaluation test chart frame 200 are known. However, since the manufacturer of the X-ray inspection apparatus does not disclose the scale ratio from the actual size to the display display, the strain calculation unit 118 calculates the average scale ratio of the three frames. An image obtained by photographing the distortion evaluation test chart frame 200 with the X-ray baggage inspection apparatus should be equal to a photographed image size determined by an ideal scale ratio if there is no distortion. However, since there is actually distortion, the images in the L1, L2, and L3 portions each have a distortion error. This distortion error is obtained by multiplying the actual dimension by the average scale ratio and making a difference from the image measurement value. In FIG. 9, the length measurement parts are described as three places. However, similarly to the explanation of FIG. 8, four different places in a plane substantially perpendicular to the X-ray emission direction may be measured.

  With reference to FIG. 10, the processing procedure will be described based on the measured value of the part. In FIG. 10, the image processing unit 117 of the control device 100 measures the lengths (three points) of the line segments of the captured image (S701). The strain calculation unit 118 calculates a ratio between the known actual dimensions (three points) and the measured measurement dimensions (three points) (S702). The distortion calculator 118 calculates the average value of the three scales (ratio) (S703). The distortion calculation unit 118 sets the average value to the scale ratio of the captured image (S704). The distortion calculation unit 118 multiplies the set scale ratio by the actual dimension (three points) and calculates a difference (distortion error) from the measurement dimension (three points) (S705). The distortion calculation unit 118 finally averages the absolute value of the distortion error to obtain the average distortion rate of the imaging result (S706). As described above, the distortion rate evaluation result of the photographing result was obtained.

  In the above-described embodiment, the dimension of the image is automatically measured, but the length may be measured on the screen or from a hard copy of the screen. Further, the average of the absolute values of distortion errors may be a mean square.

  According to the above-described embodiment, an inspector who performs X-ray baggage inspection can select an X-ray baggage inspection apparatus with less image distortion by comparing the evaluation results for each manufacturer model.

It is the figure which put together the correction method of the X-ray inspection apparatus, and the characteristic of correction. It is a figure explaining the mechanism of generation | occurrence | production of an image distortion. It is a perspective view explaining an X-ray baggage inspection apparatus and its evaluation apparatus. It is a hardware block diagram explaining the evaluation apparatus of a package X inspection apparatus. It is a functional block diagram explaining the evaluation apparatus of the package X inspection apparatus. It is a figure explaining a test chart frame. It is a figure explaining the test chart frame for planes. It is a figure explaining the dimension measurement from a radiography image (the 1). It is a figure explaining the dimension measurement from a radiography image (the 2). It is a figure explaining extraction of a distortion error from the actual size and scale rate of a test frame chart. It is a flowchart explaining the extraction procedure of a distortion error.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 100 ... Measuring apparatus, 110 ... Control apparatus, 111 ... Bus, 112 ... MPU, 113 ... MMU, 114 ... GDU, 115 ... DCU, 116 ... Hard disk, 117 ... Image processing part, 118 ... Distortion calculation part, 120 ... Interface card , 130 ... Display display, 140 ... LAN cable, 200 ... Test chart frame for distortion evaluation, 300 ... Test chart frame for distortion evaluation, 400 ... X-ray baggage inspection device, 410 ... Front slider, 420 ... X-ray inspection device body, 430 ... rear slider, 440 ... X-ray baggage inspection and photographing result display.

Claims (5)

Flowing a distortion evaluation test chart consisting of a plurality of sides to the X-ray baggage inspection apparatus;
Measuring the lengths of the plurality of sides from a transmission image of the distortion evaluation test chart output by the X-ray luggage inspection device;
Calculating an average reduction rate of the X-ray baggage inspection apparatus from a plurality of measured lengths and actual dimensions of the plurality of sides;
Multiplying a plurality of actual dimensions and the average reduction ratio, and calculating a difference between the corresponding plurality of lengths;
An evaluation method for an X-ray luggage inspection apparatus, comprising: calculating an evaluation value from a plurality of the differences. An evaluation method for an X-ray luggage inspection apparatus according to claim 1,
The method for evaluating an X-ray baggage inspection apparatus, wherein the plurality of sides are made of metal. An evaluation method for an X-ray baggage inspection apparatus according to claim 1 or 2,
The step of calculating the evaluation value is a step of calculating an average value of the absolute values of the differences. An image processing unit for measuring the lengths of the plurality of sides from a transmission image of the distortion evaluation test chart obtained by flowing a distortion evaluation test chart consisting of a plurality of sides in an X-ray luggage inspection apparatus;
An evaluation apparatus for an X-ray baggage inspection apparatus, comprising: a distortion calculation unit that calculates an evaluation value of image distortion by the X-ray baggage inspection apparatus based on a plurality of measured lengths and actual dimensions of the plurality of sides. Computer
An image processing unit that measures the lengths of the plurality of sides from a transmission image of the distortion evaluation test chart obtained by flowing a distortion evaluation test chart consisting of a plurality of sides in an X-ray luggage inspection apparatus;
A strain calculator that calculates a distortion evaluation value of an image of the X-ray baggage inspection apparatus based on a plurality of measured lengths and actual dimensions of the plurality of sides;
X-ray baggage inspection device evaluation program to function as

Images