DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments
The reply filed on 3 September 2025 has been entered. Applicant’s arguments with respect to claims 51-64 and 66-71 have been considered but are moot in view of new ground(s) of rejection caused by the amendments.
Claims 51-64 and 66-71 are pending in this application and have been considered below. Claims 1-50 and 65 are canceled by the applicant.
Priority
Receipt is acknowledged that application claims priority to foreign application with application number (Israeli) IL 294593 dated 7 July 2022. Copies of certified papers required by 37 CFR 1.55 have been received. Priority is acknowledged under 35 USC 119(e) and 37 CFR 1.78.
Information Disclosure Statement
The IDSs dated 15 April 2024 and 27 March 2023 that have been previously considered remain placed in the application file.
Specification - Abstract
The abstract has been amended. The objection to the abstract is withdrawn.
Specification - Drawings
Acknowledgement is made of the color drawings submitted 23 May 2023 in this application. Applicants are reminded that, absent a successful petition, the black and white drawings submitted on 23 May 2023 will be used. No petition is currently on file.
Claim Rejections - 35 USC § 112
claims 51, 69 and 70, have been amended. The rejection of claims 51, 69 and 70 under 35 USC 112 is withdrawn.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
PNG
media_image1.png
522
615
media_image1.png
Greyscale
Claims 51-53, 55-61, 64 and 66-71 are rejected under 35 U.S.C. 103 as obvious over US Patent Publication 2020 0333266 A1, (Li et al.) in view of US Patent Publication 2021 0125036 A1, (Tremblay et al.) and US Patent Publication 2020 0394442 A1 (Ahmed et al.).
Claim 51
[AltContent: textbox (Li et al. Fig. 14, showing prohibited objects in a security scanner image.)] Regarding Claim 51, Li et al. teach a system comprising one or more processing circuitries ("particular application with computed tomography (CT) security scanners configured to generate a three-dimensional image of an article under examination," paragraph [0002]) configured to:
obtain a three-dimensional (3D) representation of at least part of an item, wherein the three-dimensional representation includes a plurality of points each associated with an intensity obtained based on an acquisition of the at least part of the item by an acquisition device ("For example, high density objects, such as a metal plate, explosive, etc. may be represented by voxels having a higher CT value than lower density objects, such as empty spaces, clothing, etc., which may be represented by voxels having a lower CT value," paragraph [0054]),
determine at least one given cluster of points in the 3D representation ("first group of voxels representative of object regions corresponding to objects within the article 104 and a second group of voxels representative of void regions corresponding to voids within the article 104 are identified," paragraph [0053]), the given cluster of points being informative of at least one given object present in the at least part of the item, and
informative of a position and an orientation of the given object ("The 3D article image of the article 104 represents a volume of the article 104 and typically depicts one or more internal or interior aspects of the article 104," paragraph [0051] and "a gravitational bottom may be dependent upon the placement of the test item 1304 within the 3D article image 1300 and the orientation of the article during the examination, for example" paragraph [0082]).
[AltContent: textbox (Tremblay et al. Fig. 20, showing calculating orientation of several objects.)]
PNG
media_image2.png
489
454
media_image2.png
Greyscale
Li et al. do not explicitly teach all of determine data orientation.
However, Tremblay et al. teach use the given cluster of points of the 3D representation to determine data Dposition/orientation ("it is possible to train a state of art viewpoint (3D orientation) regressor for untextured symmetrical objects without any hand-labelling of object symmetries (or poses). In at least one embodiment, as a method only requires a 3D model and involves a non-differentiable black box, a sampling procedure for retrieving training gradients is introduced," paragraph [0126]).
It would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to modify “Combined Image Generation of Article under Examination and Image of Test Item” as taught by Li et al. to use “Determining Object Orientation from an Image with Machine Learning” as taught by Tremblay et al.
The suggestion/motivation for doing so would have been that, “In at least one embodiment, DLA(s) may quickly and efficiently execute neural networks, especially CNNs, on processed or unprocessed data for any of a variety of functions, including, for example and without limitation: a CNN for object identification and detection using data from camera sensors; a CNN for distance estimation using data from camera sensors; a CNN for emergency vehicle detection and identification and detection using data from microphones 3296; a CNN for facial recognition and vehicle owner identification using data from camera sensors; and/or a CNN for security and/or safety related events.” as noted by the Tremblay et al. disclosure in paragraph [0194].
Li et al. and Tremblay et al. do not explicitly teach all of a generate a two-dimensional representation informative of the given object.
[AltContent: textbox (Ahmed et al. Fig. 3, showing using a CNN to analyze objects for labeling.)]
PNG
media_image3.png
309
623
media_image3.png
Greyscale
However, Ahmed et al. teach use the data Dposition/orientation informative of the position and the orientation of the given object to generate a two-dimensional representation informative of the given object ("The computing system 120 can display a result on or more monitors visible to the security office and/or traveler. This quick classification regardless of shape or material of the prohibited item allows for faster movement through checkpoints," paragraph [0024]), the two-dimensional representation being generated with a given orientation selected based on detectability by a detection algorithm, of the given object in the two-dimensional representation as a prohibited object ("The classification module 133 labels the data based on the training data comprised by the neural network 200 so that the computing system 120 can determine that the compartment does or does not contain a prohibited object. The label may be threat or no threat," paragraph [0030] where labeling the data teaches two-dimensional representations, or the label of "threat").
It would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to modify “Combined Image Generation of Article under Examination and Image of Test Item” as taught by Li et al. and “Determining Object Orientation from an Image with Machine Learning” as taught by Tremblay et al. to use “Screening Technique for Prohibited Objects at Security Checkpoints” as taught by Ahmed et al.
The suggestion/motivation for doing so would have been that, “Items of interest, such as prohibited items, can be difficult to detect within this environment due to a range of orientation, clutter, and density confusion in a traditional two-dimensional (2D) X-ray projection. Specifically, the problem of objects occluding each other is a limitation of 2D X-ray scanners, which makes detection ( automatically or by human operators) particularly challenging.” as noted by the Ahmed et al. disclosure in paragraph [0014].
The rejection of system claim 51 above applies mutatis mutandis to the corresponding limitations of method claim 69 and Computer readable medium claim 70 while noting that the rejection above cites to both device and method disclosures. Claims 69 and 70 are mapped below for clarity of the record and to specify any new limitations not included in claim 51.
Claim 52
Regarding claim 52, Li et al. teach the system of claim 51, as noted above.
Li et al. do not explicitly teach all of the region is determined using Dposition/orientation.
However, Tremblay et al. teach configured to: determine a region of the 3D representation, wherein the region is determined using Dposition/orientation ("In at least one embodiment, neural network may take as its input at least some subset of parameters, such as bounding box dimensions, ground plane estimate obtained (e.g. from another subsystem), output from IMU sensor(s) 3266 that correlates with vehicle 3200 orientation, distance, 3D location estimates of object obtained from neural network and/or other sensors (e.g., LIDAR sensor(s) 3264 or RADAR sensor(s) 3260), among others," paragraph [0207])
Li et al. and Tremblay et al. do not explicitly teach all of generate the two-dimensional representation informative of the given object.
However, Ahmed et al. teach use at least said region to generate the two-dimensional representation informative of the given object ("The classification module 133 labels the data based on the training data comprised by the neural network 200 so that the computing system 120 can determine that the compartment does or does not contain a prohibited object. The label may be threat or no threat," paragraph [0030] where labeling the data teaches two-dimensional representations, or the label of "threat").
Li et al., Tremblay et al. and Ahmed et al. are combined as per claim 51.
Claim 53
Regarding claim 53, Li et al. teach the system of claim 52, wherein the region is a three-dimensional region, wherein the system is configured to use Dposition/orientation to convert intensity of points of the three-dimensional region into the two-dimensional representation informative of the given object with the given orientation wherein the given orientation matches a desired orientation ("In this way, the data is converted from projection space to image space, a domain that may be more understandable by a user 134 viewing the image(s), for example," paragraph [0043]).
Claim 55
Regarding claim 55, Li et al. teach the system of claim 52, wherein Dposition/orientation includes one or more weights informative of a geometric distribution of the given object along one or more directions, wherein the system is configured to use the one or more weights to determine the region ("ensuring that a relatively even distribution of abutment locations 1400 about the center of mass of the test item 1304," paragraph [0089]).
Claim 56
Regarding claim 56, Li et al. teach the system of claim 51, wherein the region is determined based on Dposition/orientation for a first type of prohibited objects differently than for a second type of prohibited objects, wherein the second type is different from the first type ("the test item 302 is a gun, although other threat items and/or non-threat items are also contemplated. In some examples, such as the example illustrated in FIG. 3, the 3D test image 300 can represent more than just the test item 302," paragraph [0061]).
Claim 57
Regarding claim 57, Li et al. teach the system of claim 51, configured to: obtain one or more weights informative of a geometric distribution of the given object along one or more directions ("In some embodiments, a weight applied to the CT value of the voxel of the 3D article image 400 relative to a weight applied to the corresponding CT value of the voxel," paragraph [0065]), and
use one or more of the weights to determine whether the given cluster of points is informative of a given object which has a shape matching a shape of a prohibited object ("In some embodiments, a weight applied to the CT value of the voxel of the 3D article image 400 relative to a weight applied to the corresponding CT value of the voxel of the 3D test image may be a function of the location of respective voxels relative to the selection region. For example, a higher relative weight may be applied to voxels of the 3D article image 400 than to voxels of the 3D test image 300 near an outer boundary of the first selection region 504, whereas a higher relative weight may be applied to voxels of the 3D test image 300 than voxels of the 3D article image 400 near an inner core of the first selection region 504," paragraph [0065]).
Claim 58
Regarding claim 58, Li et al. teach the system of claim 51, wherein Dposition/orientation is informative of three axes Xobj, Yobj, Zobj informative of axes of the given object, wherein the system is configured to: determine a region of the 3D representation comprising at least part of a plane which comprises two axes Xobj, Yobj among the axes Xobj, Yobj, Zobj ("2D graphics operations are performed using one or more components of graphics processing engine (GPE) 5010. In at least one embodiment, GPE 5010 is a compute engine for performing graphics operations, including three-dimensional (3D) graphics operations and media operations," paragraph [0064]).
Li et al. do not explicitly teach all of determine a region of the 3D representation comprising at least part of a plane which is orthogonal to the axis Z.
However, Tremblay et al. teach determine a region of the 3D representation comprising at least part of a plane which is orthogonal to the axis Zobj ("techniques described herein may be used to identify both symmetry and possible orientations of objects pictured in FIG. 7," paragraph [0096]).
Li et al. and Tremblay et al. are combined as per claim 51.
Claim 59
Regarding claim 59, Li et al. teach the system of claim 51, as noted above.
Li et al. do not explicitly teach all of teach wherein Dposition/orientation is informative of three axes Xobj, Yobj, Zobj.
However, Tremblay et al. teach wherein Dposition/orientation is informative of three axes Xobj, Yobj, Zobj informative of axes of the given object ("plane equations are transmitted to coarse raster engine to generate coverage information (e.g., an x, y coverage mask for a tile) for primitive; output of coarse raster engine is transmitted to culling engine where fragments associated with primitive that fail a z-test are culled, and transmitted to a clipping engine where fragments lying outside a viewing frustum are clipped. In at least one embodiment, fragments that survive clipping and culling are passed to fine raster engine," paragraph [0486]), wherein the system is configured to:
obtain, for the axis Xobj, a weight Wx informative of geometric distribution along said axis X ("confidence may be represented or interpreted as a probability, or as providing a relative "weight" of each detection compared to other detections. In at least one embodiment, confidence enables a system to make further decisions regarding which detections should be considered as true positive detections rather than false positive detections," paragraph [0207]),
obtain, for the axis Yobj, a weight Wy informative of geometric distribution along said axis Y ("confidence may be represented or interpreted as a probability, or as providing a relative "weight" of each detection compared to other detections. In at least one embodiment, confidence enables a system to make further decisions regarding which detections should be considered as true positive detections rather than false positive detections," paragraph [0207]),
obtain, for the axis Zobj, a weight Wz informative of geometric distribution along said axis Z ("confidence may be represented or interpreted as a probability, or as providing a relative "weight" of each detection compared to other detections. In at least one embodiment, confidence enables a system to make further decisions regarding which detections should be considered as true positive detections rather than false positive detections," paragraph [0207]), perform (i) or (ii):
(i) determining a region of the 3D representation comprising at least part of a plane which includes two axes Xobj, Yobj, wherein the weights Wx, Wy and Wz indicate that geometric distribution is greater along the two axes Xobj, Yobj than along the axis Zobj ("plane equations are transmitted to coarse raster engine to generate coverage information (e.g., an x, y coverage mask for a tile) for primitive; output of coarse raster engine is transmitted to culling engine where fragments associated with primitive that fail a z-test are culled, and transmitted to a clipping engine where fragments lying outside a viewing frustum are clipped. In at least one embodiment, fragments that survive clipping and culling are passed to fine raster engine," paragraph [0486]),
(ii) determining a region of the 3D representation along a plane which is orthogonal to axis Zobj, wherein the weights Wx, Wy and Wz indicate that geometric distribution along the axis Zobj is smaller than along the two axes Xobj, Yobj ("plane equations are transmitted to coarse raster engine to generate coverage information (e.g., an x, y coverage mask for a tile) for primitive; output of coarse raster engine is transmitted to culling engine where fragments associated with primitive that fail a z-test are culled, and transmitted to a clipping engine where fragments lying outside a viewing frustum are clipped. In at least one embodiment, fragments that survive clipping and culling are passed to fine raster engine," paragraph [0486]).
Li et al. and Tremblay et al. are combined as per claim 51.
Claim 60
Regarding claim 60, Li et al. teach the system of claim 51, as noted above.
Li et al. do not explicitly teach all of at least part of a plane of the 3D representation determined using Dposition/orientation.
However, Tremblay et al. teach configured to determine a region of the 3D representation which includes: at least part of a plane of the 3D representation determined using Dposition/orientation ("real-time ray-tracing hardware accelerator may be used to quickly and efficiently determine positions and extents of objects (e.g., within a world model), to generate real-time visualization simulations," paragraph [0203]);
a volume of the 3D representation, wherein a position of the volume with respect to the plane is determined using Dposition/orientation ("real-time ray-tracing hardware accelerator may be used to quickly and efficiently determine positions and extents of objects (e.g., within a world model), to generate real-time visualization simulations," paragraph [0203] where extents is volume).
Li et al. and Tremblay et al. are combined as per claim 51.
Claim 61
Regarding claim 61, Li et al. teach the system of claim 60, wherein at least one of (i) or (ii) is met: (i) the volume has a height H+ above the plane along a direction Zobj, or has a height H- below the plane along the direction Zobj, wherein Zobj has been determined using Dposition/orientation ("the example embodiment 1200 comprises adjusting a z-value of the test item 1304 to adjust a distance between the test item 1304 and the nearest object," paragraph [0090]); or
(ii) one or more dimensions of the volume is selected based on an estimate of one or more dimensions of a prohibited object ("The 3D article image of the article 104 represents a volume of the article 104 and typically depicts one or more internal or interior aspects of the article 104," paragraph [0051]).
Claim 64
Regarding claim 64, Li et al. teach the system of claim 51, configured to perform at least one of (i), (ii) or (iii), (i) dividing the three-dimensional representation of the at least part of the item into a plurality of clusters, wherein points which belong to a same cluster meet a proximity criterion ("In some embodiments, a weight applied to the CT value of the voxel of the 3D article image 400 relative to a weight applied to the corresponding CT value of the voxel of the 3D test image may be a function of the location of respective voxels relative to the selection region. For example, a higher relative weight may be applied to voxels of the 3D article image 400 than to voxels of the 3D test image 300 near an outer boundary of the first selection region 504, whereas a higher relative weight may be applied to voxels of the 3D test image 300 than voxels of the 3D article image 400 near an inner core of the first selection region 504," paragraph [0065]); as noted above.
Li et al. do not explicitly teach all of thresholds.
However, Tremblay et al. teach (ii) dividing the three-dimensional representation of the at least part of the item into a plurality of clusters, wherein points which belong to a same cluster meet a proximity criterion, and selecting the given cluster as a cluster of the plurality of clusters which includes a number of points which is above a threshold("confidence may be represented or interpreted as a probability, or as providing a relative "weight" of each detection compared to other detections. In at least one embodiment, confidence enables a system to make further decisions regarding which detections should be considered as true positive detections rather than false positive detections," paragraph [0207]);
(iii) dividing the three-dimensional representation of the at least part of the item into a plurality of clusters wherein points which belong to a same cluster meet a proximity criterion and determining that one or more given clusters of the plurality of clusters do not represent a prohibited object, based on a detection that one or more geometrical parameters of each of the one or more given clusters do not comply with an expected range ("set a threshold value for confidence and consider only detections exceeding threshold value as true positive detections," paragraph [0207]).
Li et al. and Tremblay et al. are combined as per claim 51.
Claim 66
Regarding claim 66, Li et al. teach the system of claim 51, and informative of the position and the orientation of the given object ("The 3D article image of the article 104 represents a volume of the article 104 and typically depicts one or more internal or interior aspects of the article 104," paragraph [0051] and "a gravitational bottom may be dependent upon the placement of the test item 1304 within the 3D article image 1300 and the orientation of the article during the examination, for example" paragraph [0082]), and
use data Dposition/orientation to generate the two-dimensional representation informative of the given object ("a CT imaging system presents an operator with 3D volumetric images and/or two-dimensional (2D) projections (e.g., projected from the 3D volumetric images) of articles in the imaged volume," paragraph [0030]) as noted above.
Li et al. do not explicitly teach all of determine data Dflatness informative of a flatness of areas.
However, Tremblay et al. teach configured to: determine data Dflatness informative of a flatness of areas of the three-dimensional representation ("plane equations are transmitted to coarse raster engine to generate coverage information (e.g., an x, y coverage mask for a tile) for primitive; output of coarse raster engine is transmitted to culling engine where fragments associated with primitive that fail a z-test are culled," paragraph [0486] where a plane is flatness),
use the data Dflatness to determine the given cluster of points in the three-dimensional representation, ("plane equations are transmitted to coarse raster engine to generate coverage information (e.g., an x, y coverage mask for a tile) for primitive; output of coarse raster engine is transmitted to culling engine where fragments associated with primitive that fail a z-test are culled," paragraph [0486] where culling is within the interpretation of informative of at least one given object present), and
use the given cluster of points to determine the data Dposition/orientation ("it is possible to train a state of art viewpoint (3D orientation) regressor for untextured symmetrical objects without any hand-labelling of object symmetries (or poses). In at least one embodiment, as a method only requires a 3D model and involves a non-differentiable black box, a sampling procedure for retrieving training gradients is introduced," paragraph [0126]).
Li et al. and Tremblay et al. are combined as per claim 51.
Claim 67
Regarding claim 67, Li et al. teach the system of claim 51, and determine a given three-dimensional area of the 3D representation, said given three-dimensional area including said given point ("first group of voxels representative of object regions corresponding to objects within the article 104 and a second group of voxels representative of void regions corresponding to voids within the article 104 are identified," paragraph [0053]),
determine the given cluster of points within said given three-dimensional area, use the given cluster of points to determine the data Dposition/orientation informative of the position and the orientation of the given object ("The 3D article image of the article 104 represents a volume of the article 104 and typically depicts one or more internal or interior aspects of the article 104," paragraph [0051] and "a gravitational bottom may be dependent upon the placement of the test item 1304 within the 3D article image 1300 and the orientation of the article during the examination, for example" paragraph [0082]),
determine a region of the 3D representation ("first group of voxels representative of object regions corresponding to objects within the article 104 and a second group of voxels representative of void regions corresponding to voids within the article 104 are identified," paragraph [0053]), wherein the region is determined using at least part of the given cluster of points and Dposition/orientation("The 3D article image of the article 104 represents a volume of the article 104 and typically depicts one or more internal or interior aspects of the article 104," paragraph [0051] and "a gravitational bottom may be dependent upon the placement of the test item 1304 within the 3D article image 1300 and the orientation of the article during the examination, for example" paragraph [0082]), and
use at least said region to generate the two-dimensional representation informative of the given object("a CT imaging system presents an operator with 3D volumetric images and/or two-dimensional (2D) projections (e.g., projected from the 3D volumetric images) of articles in the imaged volume," paragraph [0030]).
Li et al. do not explicitly teach all of data Dflatness informative of a flatness of said given point.
However, Tremblay et al. teach configured to: determine, for each given point of a plurality of points of the three-dimensional representation, data Dflatness informative of a flatness of said given point ("plane equations are transmitted to coarse raster engine to generate coverage information (e.g., an x, y coverage mask for a tile) for primitive; output of coarse raster engine is transmitted to culling engine where fragments associated with primitive that fail a z-test are culled," paragraph [0486] where a plane is flatness),
Li et al. and Tremblay et al. are combined as per claim 51.
Claim 68
Regarding claim 68, Li et al. teach the system of claim 51, wherein the one or more processing circuitries are configured to execute said detection algorithm, wherein said detection algorithm comprises a trained machine learning model, wherein the one or more processing circuitries are configured to: feed the two-dimensional representation informative of the given object to the trained machine learning model, and use the trained machine learning model to determine whether the given object is a prohibited object ("the test item 302 is a gun, although other threat items and/or non-threat items are also contemplated. In some examples, such as the example illustrated in FIG. 3, the 3D test image 300 can represent more than just the test item 302," paragraph [0061]),
wherein at least one of (i) or (ii) is met: (i) most of the images of prohibited objects used to train the machine learning model include prohibited objects observed with a viewing angle which matches an optimal viewing angle according to a criterion ("defining the first selection region comprises defining the first selection region as a function of the size and/or shape of the test image of the test item (e.g., to select a cluster of voxels that best approximates the size and/or shape of the test item)," paragraph [0056]where best approximates is within the interpretation of optimal); or
(ii) the trained machine learning model has been trained with images of prohibited objects, wherein most of the images of prohibited objects used to train the machine learning model include prohibited objects observed with a viewing angle which matches a viewing angle of the given object in the two-dimensional image according to a criterion ("defining the first selection region comprises defining the first selection region as a function of the size and/or shape of the test image of the test item (e.g., to select a cluster of voxels that best approximates the size and/or shape of the test item)," paragraph [0056]where best approximates is a criterion).
Claim 69
Regarding claim 69, Li et al. teach a computer-implemented method ("method 200 for generating a 3D combined image representative of an article undergoing a radiation examination and representative of a test item not comprised within the article during the radiation examination," paragraph [0051]) comprising:
obtaining a three-dimensional (3D) representation of at least part of an item, wherein the three-dimensional representation includes a plurality of points each associated with an intensity obtained based on an acquisition of the at least part of the item by an acquisition device("For example, high density objects, such as a metal plate, explosive, etc. may be represented by voxels having a higher CT value than lower density objects, such as empty spaces, clothing, etc., which may be represented by voxels having a lower CT value," paragraph [0054]),
determining at least one given cluster of points in the 3D representation ("first group of voxels representative of object regions corresponding to objects within the article 104 and a second group of voxels representative of void regions corresponding to voids within the article 104 are identified," paragraph [0053]), the given cluster of points being informative of at least one given object present in the at least part of the item ("The 3D article image of the article 104 represents a volume of the article 104 and typically depicts one or more internal or interior aspects of the article 104," paragraph [0051] and "a gravitational bottom may be dependent upon the placement of the test item 1304 within the 3D article image 1300 and the orientation of the article during the examination, for example" paragraph [0082]),
Li et al. do not explicitly teach all of using the given cluster of points of the 3D representation to determine data Dposition/orientation informative of a position and an orientation.
However, Tremblay et al. teach using the given cluster of points of the 3D representation to determine data Dposition/orientation informative of a position and an orientation of the given object("it is possible to train a state of art viewpoint (3D orientation) regressor for untextured symmetrical objects without any hand-labelling of object symmetries (or poses). In at least one embodiment, as a method only requires a 3D model and involves a non-differentiable black box, a sampling procedure for retrieving training gradients is introduced," paragraph [0126])
Li et al. and Tremblay et al. do not explicitly teach all of generate a two-dimensional representation informative of the given object.
However, Ahmed et al. teach using the data Dposition/orientation informative of the position and the orientation of the given object to generate a two-dimensional representation informative of the given object ("The computing system 120 can display a result on or more monitors visible to the security office and/or traveler. This quick classification regardless of shape or material of the prohibited item allows for faster movement through checkpoints," paragraph [0024]) the two-dimensional representation being generated with a given orientation selected based on detectability, by a detection algorithm, of the given object in the two dimensional representation as a prohibited object ("The classification module 133 labels the data based on the training data comprised by the neural network 200 so that the computing system 120 can determine that the compartment does or does not contain a prohibited object. The label may be threat or no threat," paragraph [0030] where labeling the data teaches two-dimensional representations, or the label of "threat").
Li et al., Tremblay et al. and Ahmed et al. are combined as per claim 51.
Claim 70
Regarding claim 70, Li et al. teach a non-transitory computer readable medium storing instructions that when executed by one or more processing circuitries, cause the one or more processing circuitries to:
obtain a three-dimensional (3D) representation of at least part of an item, wherein the three-dimensional representation includes a plurality of points each associated with an intensity obtained based on an acquisition of the at least part of the item by an acquisition device("For example, high density objects, such as a metal plate, explosive, etc. may be represented by voxels having a higher CT value than lower density objects, such as empty spaces, clothing, etc., which may be represented by voxels having a lower CT value," paragraph [0054]),
determine at least one given cluster of points in the 3D representation ("first group of voxels representative of object regions corresponding to objects within the article 104 and a second group of voxels representative of void regions corresponding to voids within the article 104 are identified," paragraph [0053]), the given cluster of points being informative of at least one given object present in the at least part of the item ("The 3D article image of the article 104 represents a volume of the article 104 and typically depicts one or more internal or interior aspects of the article 104," paragraph [0051] and "a gravitational bottom may be dependent upon the placement of the test item 1304 within the 3D article image 1300 and the orientation of the article during the examination, for example" paragraph [0082]).
Li et al. do not explicitly teach all of use the given cluster of points of the 3D representation to determine data Dposition/orientation informative of a position and an orientation.
However, Tremblay et al. teach use the given cluster of points of the 3D representation to determine data Dposition/orientation informative of a position and an orientation of the given object ("it is possible to train a state of art viewpoint (3D orientation) regressor for untextured symmetrical objects without any hand-labelling of object symmetries (or poses). In at least one embodiment, as a method only requires a 3D model and involves a non-differentiable black box, a sampling procedure for retrieving training gradients is introduced," paragraph [0126]).
Li et al. and Tremblay et al. do not explicitly teach all of generate a two-dimensional representation informative of the given object.
However, Ahmed et al. teach use the data Dposition/orientation informative of the position and the orientation of the given object to generate a two-dimensional representation informative of the given object ("The computing system 120 can display a result on or more monitors visible to the security office and/or traveler. This quick classification regardless of shape or material of the prohibited item allows for faster movement through checkpoints," paragraph [0024]), the two-dimensional representation being generated with a given orientation selected based on detectability by a detection algorithm of the given object in the two-dimensional representation as a prohibited object ("The classification module 133 labels the data based on the training data comprised by the neural network 200 so that the computing system 120 can determine that the compartment does or does not contain a prohibited object. The label may be threat or no threat," paragraph [0030] where labeling the data teaches two-dimensional representations, or the label of "threat").
Li et al., Tremblay et al. and Ahmed et al. are combined as per claim 51.
Claim 71
Regarding claim 71, Li et al. teach the system of claim 68, as noted above.
Li et al. do not explicitly teach all of informative of prohibited objects with an orientation which matches an optimal orientation.
However, Tremblay et al. teach wherein at least one of (i) or (ii) is met:
(i) most of images of prohibited objects used to train the machine learning model are informative of prohibited objects with an orientation which matches an optimal orientation according to a criterion ("In at least one embodiment, a viewpoint regressor coupled with a 3D renderer, serving as generator, is trained simultaneously with a discriminator. In at least one embodiment, a discriminator's goal is to challenge a generator to improve its predictions," paragraph [0065] where a 3D renderer teaches optimal orientation); or
(ii) the machine learning model has been trained with images of prohibited objects, wherein most of images of prohibited objects used to train the machine learning model are informative of prohibited objects with an orientation which matches the given orientation of the given object in the two-dimensional representation according to a criterion ("In at least one embodiment, a key idea is to train a viewpoint regressor using an adversarial training paradigm where adversarial component learns to provide a loss function that is consistent with an object's symmetries. In at least one embodiment, techniques described herein utilize a viewpoint regressor coupled with a 3D renderer, serving as generator, simultaneously with a discriminator," paragraph [0073]).
Li et al. and Tremblay et al. are combined as per claim 51.
Allowable Subject Matter
Claims 54, 62 and 63 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
Reference Cited
The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure.
International Patent Publication 2023 009557 A1 to Khinvasara et al. discloses use data obtained from an inferred object to determine whether to re-infer the same inferred object. In at least one embodiment, data obtained from inferencing a tracked object among a set of images is used to adjust a set of conditions.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HEATH E WELLS whose telephone number is (703)756-4696. The examiner can normally be reached Monday-Friday 8:00-4:00.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Ms. Jennifer Mehmood can be reached on 571-272-2976. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/H.E.W./Examiner, Art Unit 2664
Date: 7 November 2025
/JENNIFER MEHMOOD/Supervisory Patent Examiner, Art Unit 2664