A method and device for correcting distortion of DR fluoroscopic images in security inspection equipment

Abstract

The present invention relates to a method and device for correcting distortion of DR (Dysplastic Resonance) (DR) fluoroscopic images in security inspection equipment. This method addresses the problem of existing security inspection distortion correction methods, where the virtual detector position is fixed and the influence of object height is not considered, resulting in distorted DR fluoroscopic images and low image judgment reliability and efficiency. The method comprises: obtaining a first CT tomographic image excluding a target object and second CT tomographic images containing the target object; determining the height of the target object based on the first and second CT tomographic images; setting the position of the virtual detector in the DR scanning imaging of the security inspection equipment based on the height of the target object; and mapping the original DR fluoroscopic image of the target object received by the DR detector onto the virtual detector to obtain a distortion-corrected DR fluoroscopic image of the target object.

Description

Technical Field

[0001] The present invention relates to the technical field of security inspection image correction, and in particular to a method and device for correcting distortion of DR (Droscope Driver) fluoroscopic images in security inspection equipment. Background Art

[0002] Currently, X-ray imaging-based security inspection systems are widely used in subways, airports, and other locations for luggage and parcel security checks. Traditional security inspection systems use two-dimensional perspective DR (Digital Radiography) images, produced by X-rays passing through an object, to determine whether it contains dangerous goods such as knives and guns. Because the DR detectors used are often L- or U-shaped, the image corresponding to the intersection of the vertical and horizontal parts of the DR detector exhibits significant abrupt changes, often requiring image distortion correction. Distortion correction in security inspection systems involves setting up a set of virtual detectors and mapping the corresponding rays from the original DR detectors onto these virtual detectors, thereby obtaining a distortion-corrected image.

[0003] Currently, virtual detectors on security inspection equipment are typically fixed in position. However, due to the wide variety of items inspected, their heights range, and the varying distances between objects and the DR radiation source, resulting in different magnification ratios, the images corresponding to the upper and lower planes of the calibrated object can differ significantly, which is inconsistent with human intuition. For example, if the virtual detector is set as a linear virtual detector located at the level of the conveyor belt, the upper and lower surfaces of the scanned luggage will be distorted due to the significant difference in magnification ratio, resulting in image distortion, which will affect the security inspector's image interpretation work.

[0004] Therefore, in the current distortion correction method of DR fluoroscopic images in security inspection equipment, the virtual detector position is set to be fixed, and the influence of different object heights is not considered, resulting in distortion of the obtained DR fluoroscopic images and low image recognition reliability and efficiency. Summary of the Invention

[0005] In view of the above analysis, the embodiments of the present invention aim to provide a method and device for distortion correction of DR fluoroscopic images in security inspection equipment, so as to solve the problems in existing distortion correction methods, in which the virtual detector position is set to be fixed, the influence of the object height is not considered, the obtained DR fluoroscopic images are distorted, and the image judgment reliability and efficiency are low.

[0006] In one aspect, an embodiment of the present invention provides a method for correcting distortion of a DR fluoroscopic image in a security inspection device, comprising the following steps:

[0007] Acquiring a first CT tomographic image that does not include the target object and each second CT tomographic image that includes the target object;

[0008] Obtaining a height of a target object according to the first CT tomographic image and each second CT tomographic image;

[0009] Setting the position of the virtual detector in the DR scanning imaging of the security inspection equipment based on the height of the target object;

[0010] The original DR perspective image of the target object received by the DR detector is mapped onto the virtual detector to obtain a DR perspective image of the target object after distortion correction.

[0011] Furthermore, each of the second CT tomographic images is a CT tomographic image containing the target object acquired sequentially within a set time, and the first CT tomographic image and each second CT tomographic image are two-dimensional images with width as rows and height as columns; wherein, the plane formed by the width direction and the height direction is perpendicular to the forward direction of the conveyor belt, and the height direction is perpendicular to the plane where the conveyor belt is located.

[0012] Furthermore, obtaining the height of the target object according to the first CT tomographic image and each second CT tomographic image includes:

[0013] sequentially processing each second CT tomographic image based on the first CT tomographic image to determine the height of the target object in each second CT tomographic image;

[0014] Calculating an average value of the heights of the target object in each second CT tomographic image, and using the average value as the height of the target object; or

[0015] The heights of the target objects in each second CT tomographic image are sorted from large to small, and an average value of the heights of a preset number of target objects is calculated, and the average value is used as the height of the target object.

[0016] Furthermore, the sequentially processing each second CT tomographic image based on the first CT tomographic image to determine the height of the object in each second CT tomographic image includes:

[0017] Accumulating pixel values ​​of each row of the first CT tomographic image and the current second CT tomographic image and calculating an average value to obtain first one-dimensional tomographic data and second one-dimensional tomographic data;

[0018] Subtracting the first one-dimensional fault data from the current second one-dimensional fault data and taking the absolute value to obtain difference fault data;

[0019] Based on a set threshold, binarization processing is performed on the difference fault data to obtain binary difference fault data;

[0020] Counting the number of data whose pixel values ​​are continuously 1 in the binary difference fault data, and selecting the maximum value among the numbers as the number of height data;

[0021] Multiply the number of height data by the pixel length to obtain the height of the object in the current second slice data.

[0022] Furthermore, the setting of the position of the virtual detector in the DR scanning imaging based on the height of the target object includes:

[0023] The point where the CT rotation center is vertically projected onto the plane where the DR detector is located is taken as the center point. The point on the plane where the DR detector is located that passes through the center point and is vertically projected onto the conveyor belt plane is taken as the origin. The direction from the origin to the center point is taken as the longitudinal axis direction, and the direction perpendicular to the longitudinal axis is taken as the transverse axis direction. A DR scanning coordinate system is established on the plane where the DR detector is located.

[0024] Determine the necessary points on the vertical axis based on the height of the object and the set height multiple, wherein the set height multiple is a fractional multiple;

[0025] Based on the geometric shape of the DR detector and the position of the DR ray source, the rotation angle of the virtual detector relative to the horizontal axis is determined;

[0026] Set the position of the virtual probe based on the required points and rotation angle.

[0027] Furthermore, the height multiple is set to times.

[0028] Furthermore, mapping the original DR perspective image of the target object received by the DR detector onto the virtual detector to obtain the distortion-corrected DR perspective image of the target object includes:

[0029] By mapping the original DR perspective image line data received by the DR detector at each acquisition moment of the target object onto the virtual detector, the distortion-corrected DR perspective image line data of the target object at each acquisition moment is obtained;

[0030] The distortion-corrected DR perspective image line data of the target object obtained on the virtual detector at all acquisition moments are arranged in the order of the acquisition moments to obtain the distortion-corrected DR perspective image of the target object.

[0031] Furthermore, the method maps the original DR fluoroscopic image line data received by the DR detector at each acquisition moment of the target object onto the virtual detector to obtain the distortion-corrected DR fluoroscopic image line data of the target object at each acquisition moment, including:

[0032] determining the number of virtual detector units in the virtual detector according to a set unit size of the virtual detector unit and a length of the virtual detector;

[0033] The original DR fluoroscopic image line data of the DR detector at the current acquisition moment is obtained. Based on the original DR fluoroscopic image line data, each virtual detector unit in the virtual detector is processed in sequence to obtain a pixel value of each virtual detector unit. The pixel value of each virtual detector unit constitutes the distortion-corrected DR fluoroscopic image line data at the current acquisition moment. The obtained pixel value of each virtual detector unit is expressed as:

[0034] Assume that the DR ray source passes through the center point of the current virtual detector unit and projects to point K on the DR detector. Point K is located between the center points M and N of the two DR detector units m and n. Then the pixel value Y of the current virtual detector unit is expressed as:

[0035]

[0036] Among them, l represents the distance from point K to point M, k represents the distance between the center points M and N; R m 、R n denote the pixel values ​​of DR detector units m and n respectively.

[0037] On the other hand, an embodiment of the present invention provides a device for correcting distortion of DR fluoroscopic images in a security inspection device, comprising:

[0038] a CT data acquisition module, configured to acquire a first CT tomographic image excluding the target object and each second CT tomographic image containing the target object;

[0039] an object height determination module, configured to obtain a height of a target object based on the first CT tomographic image and each second CT tomographic image;

[0040] A virtual detector setting module is used to set the position of the virtual detector in DR scanning imaging based on the height of the target object;

[0041] The distortion correction module is used to map the original DR perspective image of the target object received by the DR detector to the virtual detector to obtain a DR perspective image of the target object after distortion correction.

[0042] Furthermore, each of the second CT tomographic images is a CT tomographic image containing the target object acquired sequentially within a set time, and the first CT tomographic image and each second CT tomographic image are two-dimensional images with width as rows and height as columns; wherein, the plane formed by the width direction and the height direction is perpendicular to the forward direction of the conveyor belt, and the height direction is perpendicular to the plane where the conveyor belt is located.

[0043] Compared with the prior art, the present invention can achieve the following beneficial effects:

[0044] The present invention provides a method and device for correcting distortion of DR fluoroscopic images in security inspection equipment. The method obtains the height of the target object through the tomographic data in the CT scanning imaging in the security inspection equipment, sets the position of the virtual detector in the DR scanning imaging in the security inspection equipment according to the height, and then maps the image data received by the DR detector to the virtual detector, thereby obtaining a DR fluoroscopic image after distortion correction. The method can set the position of the virtual detector according to the actual height of the target object and can be adjusted according to the height of different target objects. The method has stronger adaptability, so that the DR fluoroscopic image corrected by the virtual detector is more realistic, which helps to improve the reliability and efficiency of image judgment.

[0045] In the present invention, the above-mentioned technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of the present invention will be described in the following description, and some advantages will become apparent from the description or be learned through practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the contents particularly pointed out in the description and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings are only for the purpose of illustrating particular embodiments and are not to be considered limiting of the present invention. Like reference symbols denote like parts throughout the drawings.

[0047] Figure 1 A schematic flow chart of a method for correcting distortion of DR fluoroscopic images in security inspection equipment provided by an embodiment of the present invention;

[0048] Figure 2 This is a schematic structural diagram of a security inspection device according to an embodiment of the present invention;

[0049] Figure 3 Schematic diagram of the position setting of the virtual detector in DR scanning imaging according to an embodiment of the present invention;

[0050] Figure 4 Schematic diagram of virtual detector distortion correction in DR scanning imaging according to an embodiment of the present invention;

[0051] Reference numerals:

[0052] 1-CT ray source; 2-CT slip ring; 3-CT detector; 4-target object; 5-conveyor belt; 6-DR ray source; 7-DR detector; 8-virtual detector. DETAILED DESCRIPTION

[0053] The preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, and are not used to limit the scope of the present invention.

[0054] Current security inspection equipment usually sets the virtual detector position to be fixed when performing DR perspective image distortion correction, and does not consider the impact of different heights of each target object. Since the height range of the target object is relatively large, there will be a large difference in the images corresponding to the upper and lower planes of the object after correction, resulting in image distortion, which makes the reliability and efficiency of image judgment low. X-ray computed tomography imaging technology (CT) obtains a three-dimensional image of the scanned object by reconstructing the scanning data of the object, thereby realizing the identification of dangerous goods in the scanned object. The CT tomographic image obtained in CT imaging can obtain the three-dimensional geometric information of the scanned object, that is, the height information. Therefore, this application proposes a method and device for distortion correction of DR perspective images in security inspection equipment with both CT imaging and DR imaging. The object of the security inspection equipment with both CT imaging and DR imaging moves with the conveyor belt, passes through the CT imaging scanning area and the DR imaging scanning area in turn, and passes through the CT rays in the CT imaging scanning area. The source and CT detector rotate around the target object as the CT slip ring rotates, acquiring attenuation data at different positions of the target object. The acquired attenuation data is then reconstructed to obtain tomographic images of the target object at different positions. In the DR imaging scanning area, the DR detector receives the X-ray attenuation line data transmitted by the DR ray source through the target object, and then performs distortion correction on the line data through a fixed virtual detector. Finally, all the collected line data of the target object are arranged to obtain a distortion-corrected DR perspective image. The virtual detector is fixed in DR imaging and does not take into account relevant information about the target object, nor does it change position due to differences in the target object. In the present invention, the height of the target object is obtained based on the CT tomographic image, and the position of the virtual detector in the DR scanning imaging is then set based on the height of the target object, so that the obtained distortion-corrected DR perspective image is more consistent with the actual situation, thereby improving the reliability and efficiency of image judgment.

[0055] A specific embodiment of the present invention discloses a method for correcting distortion of DR fluoroscopic images in security inspection equipment, such as Figure 1 As shown, the following steps are included:

[0056] Step S1: Acquire a first CT slice image excluding a target object and second CT slice images including the target object. It is understood that the first CT slice image is image data under an empty scan condition, that is, image data when the CT scanning area does not include the target object.

[0057] Specifically, the security inspection equipment in this embodiment has both CT imaging and DR imaging, as shown in the schematic diagram. Figure 2As shown, the security inspection equipment includes a CT ray source 1, a CT slip ring 2, a CT detector 3, an object 4, a conveyor belt 5, a DR ray source, and a DR detector 7. The working process is that the target object 4 moves along the conveyor belt 5, passes through the CT imaging scanning area and the DR imaging scanning area in sequence, and obtains a CT tomographic image and a DR fluoroscopic image, respectively.

[0058] During implementation, each of the second CT images is a CT image containing the target object, acquired sequentially within a set time period. The first CT image and each second CT image are two-dimensional images with width as rows and height as columns. The plane formed by the width and height directions is perpendicular to the conveyor belt's travel direction, and the height direction is perpendicular to the plane of the conveyor belt. In other words, the plane of the CT images is perpendicular to the conveyor belt's travel direction. The plane of the CT images is decomposed into two directions perpendicular to the conveyor belt's travel direction: one direction perpendicular to the conveyor belt's plane is the height direction, and the other direction is the width direction. It is understood that the height direction sequentially represents the conveyor belt and the target object.

[0059] Specifically, the set time for obtaining each second CT tomographic image can be set according to actual needs. It can be set to the time required to complete the scanning of the target object, or it can be set to a fixed interval time period. The height of the entire target object or the height of a part of the target object can be obtained, making the subsequent virtual detector position setting more flexible.

[0060] Step S2: Obtain the height of the target object based on the first CT slice image and each second CT slice image. It is understandable that the height of the target object in its entirety or in part can be obtained by obtaining the height of the target object in different second CT slice images.

[0061] During implementation, in step S2, obtaining the height of the target object based on the first CT tomographic image and each second CT tomographic image includes:

[0062] sequentially processing each second CT tomographic image based on the first CT tomographic image to determine the height of the target object in each second CT tomographic image;

[0063] Calculating an average value of the heights of the target object in each second CT tomographic image, and using the average value as the height of the target object; or

[0064] The heights of the target objects in each second CT tomographic image are sorted from large to small, and an average value of the heights of a preset number of target objects is calculated, and the average value is used as the height of the target object.

[0065] It should be noted that when the height variation of the target object in each obtained CT tomographic image is small, the average value of the object height in all the second CT tomographic images is selected to be calculated, and the average value is used as the height of the target object; when the height variation of the target object in each obtained CT tomographic image is large, the average value of the object height of a preset proportional number is selected to be calculated as the height of the target object, so that the obtained target object height can better reflect the actual height of the target object and be more adaptable to different height variations of the target object, so that the distortion correction effect of the virtual detector set according to the height of the target object is better.

[0066] Furthermore, the height change of the target object is judged by setting a height change threshold. If the height change of the target object is greater than the height change threshold, it is judged that the height change of the target object is small; otherwise, it is judged that the height change of the target object is large. The height change of the target object can be represented by calculating the variance of the target object height obtained in all CT tomographic images. The height change threshold can be set according to actual needs.

[0067] Preferably, the preset ratio can be set to 50%.

[0068] In a specific implementation, the second CT tomographic images are processed separately based on the first CT tomographic image to determine the height of the object in each second CT tomographic image. It can be understood that the same processing process is performed on each acquired second CT tomographic image, including:

[0069] Accumulating pixel values ​​of each row of the first CT tomographic image and the current second CT tomographic image and calculating an average value to obtain first one-dimensional tomographic data and second one-dimensional tomographic data;

[0070] Subtracting the first one-dimensional tomographic data from the current second one-dimensional tomographic data and taking the absolute value to obtain difference tomographic data; it can be understood that the first one-dimensional tomographic data is obtained from the first CT tomographic image, which does not contain the object but only contains the conveyor belt. Based on this, subtracting the first one-dimensional tomographic data from the current second one-dimensional tomographic data and taking the absolute value can eliminate the height of the conveyor belt, thereby obtaining difference tomographic data containing only the height of the object;

[0071] Based on a set threshold, binarization processing is performed on the difference fault data to obtain binary difference fault data;

[0072] Counting the number of data whose pixel values ​​are continuously 1 in the binary difference fault data, and selecting the maximum value among the numbers as the number of height data;

[0073] The height of the object in the current second CT slice image is obtained by multiplying the number of height data by the pixel length.

[0074] Specifically, the threshold value may be set as a multiple of the minimum value in the first one-dimensional slice data, or may be set as a multiple of the first one-dimensional slice data, wherein the multiple is set according to actual conditions.

[0075] Preferably, after the differential fault data is binarized, filtering may be performed to eliminate isolated points and sudden interruptions in the data, thereby increasing the accuracy of the data. Preferably, the convolution kernel of the filtering may be [1 1 1] / 3.

[0076] Preferably, a height threshold can be set. If the height of the object in the current second CT slice image is less than the height threshold, the second CT slice data is deleted to avoid the influence of data errors in the second CT slice image on the subsequent determination of the height of the target object. The setting of the height threshold can be determined according to actual conditions. For example, if it is set to 0, it means that the second CT slice images that do not contain the target object in each second CT slice image are deleted.

[0077] Step S3: Setting the position of the virtual detector in the DR scanning imaging based on the height of the target object. It is understood that setting the position of the virtual detector based on the height of the target object can avoid distortion of the DR fluoroscopic image caused by the different heights of different objects, making the DR fluoroscopic image closer to reality and helping to improve the reliability and efficiency of subsequent image interpretation.

[0078] Specifically, the virtual detector may be a linear detector or an arc detector.

[0079] During implementation, in step S3, setting the position of the virtual detector in DR scanning imaging based on the height of the target object includes:

[0080] The point where the CT rotation center is vertically projected onto the plane where the DR detector is located is taken as the center point. The point on the plane where the DR detector is located that passes through the center point and is vertically projected onto the conveyor belt plane is taken as the origin. The direction from the origin to the center point is taken as the longitudinal axis direction, and the direction perpendicular to the longitudinal axis is taken as the transverse axis direction. A DR scanning coordinate system is established on the plane where the DR detector is located.

[0081] Determine the necessary points on the vertical axis based on the height of the object and the set height multiple, wherein the set height multiple is a fractional multiple;

[0082] Based on the geometric shape of the DR detector and the position of the DR ray source, the rotation angle of the virtual detector relative to the horizontal axis is determined;

[0083] Set the position of the virtual probe based on the required points and rotation angle.

[0084] Specifically, the rotation angle of the virtual detector is determined according to the actual geometric shape of the DR detector and the position of the DR ray source, so that the virtual detector can cover the target object and ensure the integrity of the target object data. After the rotation angle is selected, it does not change with different target objects.

[0085] Preferably, the set height multiple is It is understandable that setting the necessary point of the virtual detector at half the height of the target object can make the DR perspective data after distortion correction the best and most realistic.

[0086] For example, an L-shaped linear DR detector is selected, the height of the target object is H, and Set the position of the virtual detector, such as Figure 3 FIG. 1 is a schematic diagram showing the position setting of the virtual detector in DR scanning imaging in this embodiment; specifically:

[0087] The CT rotation center is vertically projected onto point a on the plane where the DR detector is located as the center point. The point o on the plane where the DR detector is located, which passes through the center point a and is vertically projected onto the conveyor belt plane, is taken as the origin. The direction from the origin to the center point is taken as the longitudinal axis y, and the direction perpendicular to the longitudinal axis y is taken as the transverse axis x. The DR scanning coordinate system xoy is established on the plane where the DR detector is located.

[0088] Select the distance from the origin on the vertical axis y Point b is used as the necessary point for the virtual detector 8, and the two ends of the virtual detector 8 are set to intersect with the rays p and q to ensure the integrity of the projection data of the target object, thereby setting the position of the virtual detector 8; wherein, the DR ray source 6 emits X-rays to project the target object 4, and the DR detector 7 receives the attenuation signal projected through the object 4, and the rays p and q emitted by the DR ray source 6 correspond to the edge DR detector units on the DR detector 7.

[0089] Step S4: Mapping the original DR fluoroscopic image of the target object received by the DR detector onto the virtual detector to obtain a distortion-corrected DR fluoroscopic image of the target object. It will be appreciated that the distortion correction process involves mapping the image data on the DR detector onto the virtual detector based on the geometric relationship between the DR ray source, the DR detector, and the virtual detector. The DR fluoroscopic image obtained on the virtual detector is the distortion-corrected DR fluoroscopic image.

[0090] During implementation, in step S4, mapping the original DR perspective image of the target object received by the DR detector onto the virtual detector to obtain the distortion-corrected DR perspective image of the target object includes:

[0091] By mapping the original DR perspective image line data received by the DR detector at each acquisition moment of the target object onto the virtual detector, the distortion-corrected DR perspective image line data of the target object at each acquisition moment is obtained;

[0092] The distortion-corrected DR perspective image line data of the target object obtained on the virtual detector at all acquisition moments are arranged in the order of the acquisition moments to obtain the distortion-corrected DR perspective image of the target object.

[0093] In a specific implementation, the method maps the original DR perspective image line data received by the DR detector at each acquisition moment of the target object onto the virtual detector to obtain the distortion-corrected DR perspective image line data of the target object at each acquisition moment, including:

[0094] The number of virtual detector units in the virtual detector is determined according to the set unit size of the virtual detector unit and the length of the virtual detector; wherein the unit size is set according to actual conditions, such as the geometric structure of the DR detector and actual DR detector parameters.

[0095] Obtaining original DR fluoroscopic image line data of the DR detector at the current acquisition moment, and sequentially processing each virtual detector unit in the virtual detector based on the original DR fluoroscopic image line data to obtain a pixel value of each virtual detector unit, wherein the pixel values ​​of each virtual detector unit constitute the distortion-corrected DR fluoroscopic image line data at the current acquisition moment; it is understandable that the original DR fluoroscopic image line data of the DR detector is composed of the pixel values ​​of all DR detector units in the DR detector;

[0096] Among them, such as Figure 4 As shown, the pixel value of each virtual detector unit is obtained as follows:

[0097] Assume that the ray emitted by the DR ray source 6 passes through the center point T of the current virtual detector unit and is projected onto point K on the DR detector. Point K is located between the center points M and N of the two DR detector units m and n. The pixel value Y of the current virtual detector unit is expressed as:

[0098]

[0099] Among them, l represents the distance from point K to point M, k represents the distance between the center points M and N; R m 、R n denote the pixel values ​​of DR detector units m and n respectively.

[0100] Combine Figure 2Schematic diagram of the security inspection equipment structure in FIG. The specific process of the method of this embodiment is as follows: first, the target object (baggage) 4 is placed on the conveyor belt 5, and then moves at a constant speed along the conveyor belt 5 into the CT scanning area for scanning; the CT slip ring 2 rotates at a constant speed, the CT ray source 1 emits an X-ray beam to transmit the object 4, and the CT detector 3 receives the attenuation signal transmitted through the target object 4 and continuously transmits the received signal to the data processing computer for data reconstruction to obtain CT tomographic data at different positions of the target object; the height of the target object is obtained based on the CT tomographic data, and the position of the virtual detector in DR imaging is set according to the height; then, the target object (baggage) 4 leaves the CT scanning area and enters the DR scanning area; the DR ray source 6 emits an X-ray beam to transmit the object 4, and the DR detector 7 receives the attenuation signal transmitted through the target object 4, i.e., line data; distortion correction is then performed based on the set virtual detector to obtain distortion-corrected line data; finally, the distortion-corrected line data collected at each moment are arranged in chronological order to obtain a distortion-corrected DR fluoroscopic image.

[0101] Compared with the prior art, this embodiment provides a method for correcting distortion of DR fluoroscopic images in security inspection equipment. The height of the target object is obtained through the tomographic data in the CT scanning imaging in the security inspection equipment, and the position of the virtual detector in the DR scanning imaging in the security inspection equipment is set according to the height. The image data received by the DR detector is then mapped to the virtual detector to obtain the DR fluoroscopic image after distortion correction. The position of the virtual detector can be set according to the actual height of the target object and can be adjusted according to the height of different target objects. It has stronger adaptability, so that the DR fluoroscopic image corrected by the virtual detector is more realistic, which helps to improve the reliability and efficiency of image judgment.

[0102] Another specific embodiment of the present invention discloses a device for correcting distortion of DR fluoroscopic images in security inspection equipment, comprising:

[0103] a CT data acquisition module, configured to acquire a first CT slice image of an empty scan that does not include the target object and each second CT slice image that includes the target object;

[0104] an object height determination module, configured to obtain a height of a target object based on the first CT tomographic image and each second CT tomographic image;

[0105] A virtual detector setting module is used to set the position of the virtual detector in DR scanning imaging based on the height of the target object;

[0106] The distortion correction module is used to map the original DR perspective image of the target object received by the DR detector to the virtual detector to obtain a DR perspective image of the target object after distortion correction.

[0107] During implementation, each of the second CT tomographic images is a CT tomographic image containing the target object that is acquired sequentially within a set time, and the first CT tomographic image and each second CT tomographic image are two-dimensional images with width as rows and height as columns; wherein, the plane formed by the width direction and the height direction is perpendicular to the forward direction of the conveyor belt, and the height direction is perpendicular to the plane where the conveyor belt is located.

[0108] The specific implementation process of the embodiment of the present invention can be found in the above method embodiment, and this embodiment will not be repeated here.

[0109] Since the principles of this embodiment are the same as those of the above method embodiment, the present device also has the corresponding technical effects of the above method embodiment.

[0110] Those skilled in the art will appreciate that all or part of the process steps of the above-described embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, such as a magnetic disk, an optical disk, a read-only memory, or a random access memory.

[0111] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with this technical field within the technical scope disclosed in the present invention should be covered by the scope of protection of the present invention.

Claims

1. A method for correcting distortion of DR fluoroscopic images in security inspection equipment, characterized in that: The following steps are involved: Acquiring a first CT tomographic image that does not include the target object and each second CT tomographic image that includes the target object; Obtaining a height of a target object according to the first CT tomographic image and each second CT tomographic image; comprising: Based on the first CT tomographic image, each second CT tomographic image is processed separately to determine the height of the target object in each second CT tomographic image; specifically, the pixel values ​​of each row of the first CT tomographic image and the current second CT tomographic image are accumulated and the average value is calculated to obtain first one-dimensional tomographic data and second one-dimensional tomographic data; the first one-dimensional tomographic data is subtracted from the current second one-dimensional tomographic data and the absolute value is taken to obtain difference tomographic data; based on a set threshold, the difference tomographic data is binarized to obtain binary difference tomographic data; the number of data with pixel values ​​continuously equal to 1 in the binary difference tomographic data is counted, and the maximum value among the number is selected as the number of height data; the number of height data is multiplied by the pixel length to obtain the height of the object in the current second tomographic data; Calculating an average value of the heights of the target objects in each of the second CT tomographic images, and using the average value as the height of the target object; or sorting the heights of the target objects in each of the second CT tomographic images from largest to smallest, calculating an average value of the heights of a preset proportion of the target objects, and using the average value as the height of the target object; Setting the position of the virtual detector in the DR scanning imaging of the security inspection equipment based on the height of the target object; The original DR perspective image of the target object received by the DR detector is mapped onto the virtual detector to obtain a DR perspective image of the target object after distortion correction.

2. The method for correcting distortion of DR fluoroscopic images in security inspection equipment according to claim 1, characterized in that: Each of the second CT tomographic images is a CT tomographic image containing the target object that is acquired sequentially within a set time. The first CT tomographic image and each second CT tomographic image are two-dimensional images with width as rows and height as columns; wherein the plane formed by the width direction and the height direction is perpendicular to the forward direction of the conveyor belt, and the height direction is perpendicular to the plane where the conveyor belt is located.

3. The method for correcting distortion of DR fluoroscopic images in security inspection equipment according to claim 1, characterized in that: The step of setting the position of the virtual detector in the DR scanning imaging based on the height of the target object includes: The point where the CT rotation center is vertically projected onto the plane where the DR detector is located is taken as the center point. The point on the plane where the DR detector is located that passes through the center point and is vertically projected onto the conveyor belt plane is taken as the origin. The direction from the origin to the center point is taken as the longitudinal axis direction, and the direction perpendicular to the longitudinal axis is taken as the transverse axis direction. A DR scanning coordinate system is established on the plane where the DR detector is located. Determine the necessary points on the vertical axis based on the height of the object and the set height multiple, wherein the set height multiple is a fractional multiple; Based on the geometric shape of the DR detector and the position of the DR ray source, the rotation angle of the virtual detector relative to the horizontal axis is determined; Set the position of the virtual probe based on the required points and rotation angle.

4. The method for correcting distortion of DR fluoroscopic images in security inspection equipment according to claim 3, characterized in that: The set height multiple is times.

5. The method for correcting distortion of DR fluoroscopic images in security inspection equipment according to claim 3, characterized in that: The method of mapping the original DR perspective image of the target object received by the DR detector onto the virtual detector to obtain the distortion-corrected DR perspective image of the target object includes: By mapping the original DR perspective image line data received by the DR detector at each acquisition moment of the target object onto the virtual detector, the distortion-corrected DR perspective image line data of the target object at each acquisition moment is obtained; The distortion-corrected DR perspective image line data of the target object obtained on the virtual detector at all acquisition moments are arranged in the order of the acquisition moments to obtain the distortion-corrected DR perspective image of the target object.

6. The method for correcting distortion of DR fluoroscopic images in security inspection equipment according to claim 5, characterized in that: The method maps the original DR perspective image line data received by the DR detector at each acquisition moment of the target object onto the virtual detector to obtain the distortion-corrected DR perspective image line data of the target object at each acquisition moment, including: determining the number of virtual detector units in the virtual detector according to a set unit size of the virtual detector unit and a length of the virtual detector; The original DR fluoroscopic image line data of the DR detector at the current acquisition moment is obtained. Based on the original DR fluoroscopic image line data, each virtual detector unit in the virtual detector is processed in sequence to obtain a pixel value of each virtual detector unit. The pixel value of each virtual detector unit constitutes the distortion-corrected DR fluoroscopic image line data at the current acquisition moment. The obtained pixel value of each virtual detector unit is expressed as: Set the DR ray to pass through the center point of the current virtual detector unit and project to point K on the DR detector. Point K is located between the two DR detector units. 、 The pixel value Y of the current virtual detector unit is expressed as: ； in, Indicates the distance from point K to point M, Represents the distance between the center points M and N; 、 Represents DR detector units 、 The pixel value of .

7. A device for correcting distortion of DR fluoroscopic images in security inspection equipment, characterized in that: include: a CT data acquisition module, configured to acquire a first CT tomographic image excluding the target object and each second CT tomographic image containing the target object; An object height determination module is configured to obtain the height of a target object based on the first CT tomographic image and each second CT tomographic image; and comprises: Based on the first CT tomographic image, each second CT tomographic image is processed separately to determine the height of the target object in each second CT tomographic image; specifically, the pixel values ​​of each row of the first CT tomographic image and the current second CT tomographic image are accumulated and the average value is calculated to obtain first one-dimensional tomographic data and second one-dimensional tomographic data; the first one-dimensional tomographic data is subtracted from the current second one-dimensional tomographic data and the absolute value is taken to obtain difference tomographic data; based on a set threshold, the difference tomographic data is binarized to obtain binary difference tomographic data; the number of data with pixel values ​​continuously equal to 1 in the binary difference tomographic data is counted, and the maximum value among the number is selected as the number of height data; the number of height data is multiplied by the pixel length to obtain the height of the object in the current second tomographic data; Calculating an average value of the heights of the target objects in each of the second CT tomographic images, and using the average value as the height of the target object; or sorting the heights of the target objects in each of the second CT tomographic images from largest to smallest, calculating an average value of the heights of a preset proportion of the target objects, and using the average value as the height of the target object; A virtual detector setting module is used to set the position of the virtual detector in DR scanning imaging based on the height of the target object; The distortion correction module is used to map the original DR perspective image of the target object received by the DR detector to the virtual detector to obtain a DR perspective image of the target object after distortion correction.

8. The device for correcting distortion of DR fluoroscopic images in security inspection equipment according to claim 7, characterized in that: Each of the second CT tomographic images is a CT tomographic image containing the target object that is acquired sequentially within a set time. The first CT tomographic image and each second CT tomographic image are two-dimensional images with width as rows and height as columns; wherein the plane formed by the width direction and the height direction is perpendicular to the forward direction of the conveyor belt, and the height direction is perpendicular to the plane where the conveyor belt is located.

Images