# US 7,613,316 B2 · Methods and apparatus for detecting objects in baggage > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 7,613,316 B2 - Document ID: US7613316B2 - Status: Granted - Kind: B2 - Application number: 10/896,753 - Filed: 2004-07-22 - Published: 2005-02-17 - Granted: 2009-11-03 - Earliest claimed: 2003-07-22 - Assignees: L-3 Communications Security and Detection Systems, Inc. - Inventors: Yaghoub Mahdavieh - Web page: https://patentgrep.com/patent/US7613316B2 ## CPC - G01V 5/20: Detecting prohibited goods, e.g. weapons, explosives, hazardous substances, contraband or smuggled objects - G01V: GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS (means for indicating the location of accidentally buried, e.g. snow-buried, persons A63B29/02) ## Abstract A method of detecting a presence or absence of subject matter of interest, for example, sheet explosives and/or other potential threat objects is provided by various local and/or global gradient analysis methods including determining characteristics of gradient information of regions in an X-ray image to determine if the regions are associated with subject matter of interest for which detection may be desired, for example, contraband, explosives and/or other prohibited or unauthorized material. ## Claims ### Claim 1 A method of detecting a presence or an absence of at least one sheet explosive in an article of baggage containing a plurality of non-threat items from at least one x-ray image of the article of baggage, the method comprising acts of: determining a gradient field for at least a portion of the at least one image, the gradient field including a vector quantity indicating a gradient direction and gradient magnitude at each location for at least the portion of the at least one image; determining at least one characteristic of the gradient field including determining principal directions of variation of the gradient field; performing at least one comparison of the principal directions of variation to determine whether the at least one characteristic is consistent with characteristics of sheet explosives; and determining whether the at least one sheet explosive is present amongst the plurality of non-threat objects based, at least in part, on the at least one comparison. ### Claim 2 (depends on 1) The method of claim 1, wherein the act of determining at least one characteristic includes an act of determining a covariance matrix associated with the gradient field. ### Claim 3 (depends on 2) The method of claim 2, wherein the act of determining at least one characteristic includes an act of determining a plurality of eigenvalues associated with respective eigenvectors of the covariance matrix, and wherein the act of performing at least one comparison includes an act of performing at least one comparison of the plurality of eigenvalues. ### Claim 4 (depends on 3) The method of claim 3, wherein performing at least one comparison includes an act of determining at least one ratio between the plurality of eigenvalues. ### Claim 5 (depends on 4) The method of claim 4, wherein the act of determining the plurality of eigenvalues includes an act of determining a first eigenvalue, a second eigenvalue and a third eigenvalue respectively ordered according to decreasing magnitude. ### Claim 6 (depends on 5) The method of claim 5, wherein the act of determining at least one ratio includes an act of determining at least one of a first ratio of the first eigenvalue to the third eigenvalue, a second ratio of the first eigenvalue to the second eigenvalue, and a third ratio of the third eigenvalue to a product of the first and second eigenvalue, and wherein the act of determining whether the portion of the at least one image is associated with the subject matter of interest includes comparing at least one of the first ratio, the second ratio and the third ratio to a predetermined threshold. ### Claim 7 (depends on 1) The method of claim 1, wherein the portion of the at least one x-ray image includes a region of interest having a plurality of voxels. ### Claim 8 (depends on 7) The method of claim 7, further comprising an act of forming a first set of voxels from voxels in the region of interest based on the at least one characteristic. ### Claim 9 (depends on 8) The method of claim 8, wherein the act of determining the gradient field includes an act of determining for each target voxel in the region of interest a gradient field of a neighborhood of the target voxel. ### Claim 10 (depends on 9) The method of claim 9, further comprising an act of determining a covariance matrix associated with the gradient field. ### Claim 11 (depends on 10) The method of claim 10, further comprising an act of determining a plurality of eigenvalues associated with respective eigenvectors of the covariance matrix. ### Claim 12 (depends on 11) The method of claim 11, further comprising acts of: performing at least one comparison of the plurality of eigenvalues; and adding the target voxel to the first set of voxels depending on the at least one comparison. ### Claim 13 (depends on 12) The method of claim 12, wherein the act of performing at least one comparison includes an act of determining from the plurality of eigenvalues, a first eigenvalue having a greatest magnitude to a second eigenvalue having a least magnitude and a second ratio of a first eigenvalue to a third eigenvalue having a magnitude between the first eigenvalue and the second eigenvalue. ### Claim 14 (depends on 13) The method of claim 13, wherein the act of forming the first set of voxels includes acts of: comparing the first ratio to a first threshold value; and adding the target voxel to the first set of voxels when the first ratio exceeds the first threshold. ### Claim 15 (depends on 13) The method of claim 13, wherein the act of forming the first set of voxels includes acts of: comparing the first ratio to a first threshold value; comparing the second ratio to a second threshold value; and adding the target voxel to the first set of voxels when the first ratio exceeds the first threshold value and the second ratio exceeds the second threshold value. ### Claim 16 (depends on 12) The method of claim 12, further comprising an act of forming a first set of objects from the first set of voxels, each object in the first set of objects being comprised of voxels from the first set of voxels that are respectively connected. ### Claim 17 (depends on 16) The method of claim 16, further comprising an act of determining at least a mass of each object in the first set of objects. ### Claim 18 (depends on 17) The method of claim 17, further comprising an act of comparing each mass to a mass threshold value, and wherein the act of determining whether the portion is associated with the subject matter of interest includes an act of determining each object having a mass exceeding the mass threshold as including the subject matter of interest ### Claim 19 (depends on 1) The method of claim 1, wherein the portion of the at least one x-ray image includes a first segmented object having a plurality of voxels, and wherein the act of determining the gradient field includes an act of determining a gradient field for the first segmented object. ### Claim 20 (depends on 19) The method of claim 19, wherein the act of determining gradient field includes determining a gradient field vector for each of the plurality of voxels. ### Claim 21 (depends on 20) The method of claim 20, wherein the act of determining at least one characteristic includes an act of comparing a magnitude of each of the plurality of gradient field vectors with a first and second threshold value. ### Claim 22 (depends on 21) The method of claim 21, further comprising an act of grouping together each voxel having a magnitude greater than the first threshold value and less than the second threshold value to form a second segmented object. ### Claim 23 (depends on 22) The method of claim 22, further comprising acts of: comparing a number of voxels in the second segmented object to a number of voxels in the first segmented object; and eliminating the second segmented object when the number of voxels in the second segmented object is less than a predetermined percentage of the number of voxels in the first segmented object. ### Claim 24 (depends on 23) The method of claim 23, further comprising an act of determining a mean vector from the plurality of gradient field vectors of voxels in the second segmented object. ### Claim 25 (depends on 24) The method of claim 24, wherein the act of determining gradient field includes an act of determining a covariance matrix of the gradient field of the second segmented object based on the mean vector. ### Claim 26 (depends on 25) The method of claim 25, wherein the act of determining at least one characteristic includes an act of determining a plurality of eigenvalues associated with eigenvectors of the convariance matrix, and wherein the act of determining whether the portion is associated with the subject matter of interest includes an act of performing at least one comparison of the plurality of eigenvalues. ### Claim 27 (depends on 26) The method of claim 26, wherein the act of determining at least one characteristic includes an act of determining at least one of a first ratio of a greatest magnitude eigenvalue of the covariance matrix to a least magnitude eigenvalue of the convariance matrix and a second ratio of the least magnitude eigenvalue of the covariance matrix to a product of the greatest magnitude eigenvalue and a median magnitude eigenvalue of the covariance matrix. ### Claim 28 (depends on 27) The method of claim 27, wherein the act of determining whether the portion is associated with the subject matter of interest includes an act of identifying the second segmented object as including the subject matter of interest when the first ratio is less than a predetermined threshold value. ### Claim 29 (depends on 27) The method of claim 27, wherein the act of determining whether the portion is associated with an object of interest includes an act of identifying the second segmented object as an object of interest when the first ratio is less than a first predetermined threshold value and the second ratio is less than a second predetermined threshold value. ### Claim 30 A method of detecting a presence or an absence of one or more sheet explosives in an article of baggage from at least one image of the article of baggage, the method comprising acts of: obtaining a gradient field of at least a portion of the at least one image, the gradient field including a vector quantity indicating a gradient direction and a gradient magnitude at each location for at least the portion of the at least one image; determining at least one characteristic of the gradient information indicative of a sheet explosive including determining principal directions of variation of the gradient information; and determining whether the portion of the at least one image includes at least part of a sheet explosive based on the at least one characteristic including performing at least one comparison of the principal directions of variation. ### Claim 31 (depends on 30) The method of claim 30, wherein the act of obtaining gradient information includes an act of determining at least one Hessian matrix associated with the portion of the at least one image. ### Claim 32 (depends on 31) The method of claim 31, wherein the act of determining at least one characteristic of the gradient information includes an act of determining a set of eigenvalues associated with respective eigenvectors of the at least one Hessian matrix. ### Claim 33 A computer readable medium encoded wit a program for execution on at least one processor, the program, when executed on the at least one processor, performing a method of detecting a presence or an absence of at least one sheet explosive in an article of baggage containing a plurality of non-threat objects from at least one x-ray image of the article of baggage, the method comprising acts of: obtaining a gradient field of at least a portion of the at least one image, the gradient field including a vector quantity indicating a gradient direction and gradient magnitude at each location for at least the portion of the at least one image; determining at least one characteristic of the gradient field including determining principal directions of variation of the gradient field; performing at least one comparison of the principal directions of variation to determine whether the at least one characteristic is consistent with characteristics of sheet explosives; and determining whether the at least one sheet explosive is present amongst the plurality of non-threat objects based, at least in part, on the at least one comparison. ### Claim 34 (depends on 33) The computer readable medium of claim 33, wherein the portion of the at least one image includes a region of interest having a plurality of voxels. ### Claim 35 (depends on 34) The computer readable medium of claim 34, wherein the act of determining whether the portion is associated with the subject matter of interest includes an act of forming a first set of voxels from voxels in the region of interest based on the at least one characteristic. ### Claim 36 (depends on 35) The computer readable medium of claim 35, wherein the act of determining gradient information includes an act of determining for each target voxel in the region of interest a gradient field of a neighborhood of the voxel. ### Claim 37 (depends on 36) The computer readable medium of claim 36, further comprising an act of determining principal directions of variation associated with the gradient field. ### Claim 38 (depends on 37) The computer readable medium of claim 37, further comprising acts of: performing at least one comparison of the principal directions of variation; and adding the target voxel to the first set of voxels depending on the at least one comparison. ### Claim 39 (depends on 38) The computer readable medium of claim 38, further comprising an act of forming a first set of objects from the first set of voxels, each object in the first set of objects being comprised of voxels from the first set of voxels that are respectively connected. ### Claim 40 (depends on 39) The computer readable medium of claim 39, further comprising an act of determining at least a mass of each object in the first set of objects. ### Claim 41 (depends on 40) The computer readable medium of claim 40, further comprising an act of comparing each mass to a mass threshold value. ### Claim 42 (depends on 41) The computer readable medium of claim 41, wherein the act of determining whether the portion is associated with the subject matter of interest includes an act of determining each object having a mass exceeding the mass threshold as including the subject matter of interest. ### Claim 43 (depends on 33) The computer readable medium of claim 33, wherein the portion of the at least one image includes a first segmented object having a plurality of voxels. ### Claim 44 (depends on 43) The computer readable medium of claim 43, wherein the act of obtaining gradient information includes an act of determining a gradient field for the first segmented object. ### Claim 45 (depends on 44) The computer readable medium of claim 44, wherein the act of determining at least one characteristic of the gradient information includes an act of determining principal directions of variance for the gradient field. ### Claim 46 (depends on 43) The computer readable medium of claim 43, wherein the act of determining gradient information includes determining a gradient field vector for each of the plurality of voxels. ### Claim 47 (depends on 46) The computer readable medium of claim 46, wherein the act of determining at least one characteristic includes an act of comparing a magnitude of each of the plurality of gradient field vectors with a first and second threshold value. ### Claim 48 (depends on 47) The computer readable medium of claim 47, further comprising an act of grouping together each voxel having a magnitude greater than the first threshold value and less than the second threshold value to form a second segmented object. ### Claim 49 (depends on 48) The computer readable medium of claim 48, further comprising acts of: comparing a number of voxels in the second segmented object to a number of voxels in the first segmented object; and eliminating the second segmented object when the number of voxels in the second segmented object is less than a predetermined percentage of the number of voxels in the first segmented object. ### Claim 50 (depends on 33) The computer readable medium of claim 33, in combination with the at least one processor. ### Claim 51 (depends on 50) The combination of claim 50, in further combination with at least one X-ray scanning device, the at least one X-ray scanning device capable of providing image data to the at least one processor. ### Claim 52 (depends on 51) The combination of claim 51, wherein the at least one X-ray scanning device includes at least one X-ray computed tomography (CT) device, and wherein the at least one image includes at least one CT image. ### Claim 53 (depends on 51) The combination of claim 51, wherein the at least one X-ray scanning device includes at least one X-ray detection device adapted to scan baggage. ### Claim 54 An apparatus adapted to detect a presence or an absence of at least one sheet explosive in an article of baggage containing a plurality of non-threat objects from at least one x-ray image of the article of baggage, the apparatus comprising: at least one input adapted to receive the at least one image; and at least one controller, coupled to the at least one input, the at least one controller adapted to obtain a gradient field of at least a portion of the at least one image, the gradient field including a vector quantity indicating a gradient direction and gradient magnitude at each location for at least the portion of the at least one image, the at least one controller farther adapted to determine at least one characteristic of the gradient field including principal directions of variation of the gradient field, and to perform at least one comparison of the principal directions of variation to determine whether the at least one characteristic is consistent with sheet explosives to determine whether the at least one sheet explosive is present amongst the plurality of non-threat objects based, at least in part, on the at least one comparison. ### Claim 55 (depends on 54) The apparatus of claim 54, wherein the at least one controller comprises means for obtaining gradient information of at least the portion of the at least one image, means for determining the at least one characteristic of the gradient information, and means for determining whether the portion of the at least one image is associated with the at least one sheet explosive based on the at least one characteristic. ### Claim 56 (depends on 30) The method of claim 30, wherein the act of determining whether the portion of the at least one image is associated with the one or more sheet explosives includes an act of determining whether the gradient information of the portion of the at least one image varies predominantly in one general direction relative to other directions. ### Claim 57 (depends on 30) The method of claim 30, wherein the act of determining the at least one characteristic of the gradient information includes an act of determining principal directions of variation of the gradient information of the portion of the at least one image, and wherein the act of determining whether the portion of the at least one image is associated with one or more sheet explosives includes an act of identifying a predominant principal direction of variation. ### Claim 58 A method of detecting a presence or an absence of one or more sheet explosives in an article of baggage from at least one image of the article of baggage, the method comprising acts of: obtaining gradient information of at least a portion of the at least one image; determining at least one characteristic of the gradient information indicative of a sheet explosive including determining principal directions of variation of the gradient information of the portion of the at least one image, wherein determining the principal directions includes determining eigenvectors representing the principal directions of variation of the gradient information, the predominant principal direction of variation corresponding to an eigenvector having a greatest magnitude; and determining whether the portion of the at least one image includes at least part of a sheet explosive based on the at least one characteristic including comparing the eigenvectors to determine whether the relative magnitudes associated with the eigenvectors is characteristic of sheet explosives. ### Claim 59 (depends on 30) The method of claim 30, further comprising an act of segmenting the image into a plurality of regions, each of the plurality of regions including voxels that meet a first predetermined criteria, and wherein the act of determining gradient information includes an act of determining gradient information for each of the plurality of regions. ### Claim 60 (depends on 59) The method of claim 59, wherein the at least one image is segmented such that each of the plurality of regions is formed from voxels that meet a selected intensity threshold requirement and wherein each voxel in each respective region is adjacent to at least one other voxel in the region, and wherein a region in the plurality of regions is determined to contain a sheet explosive when principal directions of variation of the region meet a second predetermined criteria. ### Claim 61 (depends on 59) The method of claim 59, wherein the at least one image is segmented such that each of the plurality of regions is formed from voxels that meet a first gradient variation requirement local to each voxel and wherein each voxel in each respective region is adjacent to at least one other voxel in the region, and wherein a region in the plurality of regions is determined to contain a sheet explosive when the region meets a second gradient variation requirement global to the region.