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 .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 03/11/2026 and 12/23/2024 has been considered by the examiner.
Claim Rejections - 35 USC § 103
2. 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.
3. 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.
4. Claims 1-2, 6, 8, 12-14, 17-19, 23-24, 27, 32-33, 36-37 and 41-44 are rejected under 35 U.S.C. 103 as being unpatentable over Nord et al., U.S. Patent No. 11,010,605 B2, and further in view of Marciano et al., U.S. Patent Publication No. 2023/0186456 A1.
Regarding claim 1, Nord discloses A computing device-implemented method for object threat detection, the computing device including at least one processor, the method comprising:
scanning the body of an individual with a body scanner, the scanning producing a plurality of digital images (Fig. 1; col. 13, lines 24-28 – “the system configuration 100 includes at its core an object-detection system 110 that is coupled to one or more screening systems 120 by one or more links 130; col. 13, lines 33-39 – “Each screening system 120 includes at least one detection device 122 that is configured to scan or otherwise capture one or more images of a scene. Each detection device 122 can take a variety of forms, including but not limited to an X-ray machine, metal detector, MRI scanner, CT scanner, millimeter wave scanner, spectral band scanner, or other type of scanning device” – where detection device is a body scanner; col. 13, lines 40-60 – “In general, a scene comprises a two-dimensional (2D) area or three-dimensional (3D) space that can be scanned by at least one detection device 122 at a given point in time during which the area or space may contain one or more objects. There are various examples of scenes, such as a region within a screening system where baggage items, cargo containers, mail packages, etc. pass through to be scanned, a region within a screening system where a human stands to get scanned, or a region within a public location or the like where a crowd of individuals is scanned, among numerous other examples of scenes. Likewise, there are various examples of objects that may be within a scene, such as a baggage item (e.g., a purse, briefcase, piece of luggage, backpack, etc.), a human, a freight or cargo container, and mail or other packages, among numerous other examples. In practice, an object that is within a scanned scene typically also contains one or more objects itself (e.g., electronics, wallets, keys, books, clothing, food, lighters, contraband, etc.), and some of these objects may be considered an object of interest at a particular screening system 120”; col. 14, lines 28-36 – “a detection device 122 may capture multiple, different images of a scene. The multiple images may be from the same, single perspective or from multiple, different perspectives (e.g., a top-view image and a side-view image of a given piece of luggage). In some cases, the one or more images may comprise three-dimensional “slices” of a scene, where each slice represents a scan of the scene at a different level of scan depth.”;
processing the plurality of digital images using a first artificial intelligence model trained to detect known threats (col. 15, lines 8-10 – a detection device 122 may transmit data for the captured one or more images to the object-detection system 110; col. 15, lines 50-60 – the object detection system 110 is configured to perform object-detection functions based on images of scenes…identifying a perceived object of interest within an image of a scene”; col. 13, lines 61-67 – “an object of interest is an object that has one or more characteristics of an object within a type or class that has been deemed to be of interest (e.g., a security threat) such that additional attention should be given to the object”; col. 16, lines 22-44 – “The data storage 112 includes software that enables the object -detection system 110 to perform the functions disclosed herein…the data storage 112 may be provisioned with one or more object-detection models 113a, 113b, the object -detection system 110 may be configured as a multi-model object-detection system…each object-detection model 113a-b is configured to receive a given image of a scanned scene, evaluate the given image for one or more objects of interest, and the generate one or more sets of object-detection conclusions for the given image”; col. 16, line 67 – col. 17, line 2 – “some or each of the object-detection models 113a-b may take the form of a machine-learning object detection model”; col. 17, lines 7-9 – “each object-detection model 113a-b may take the form of one or more neural-network based object-detection models”; col. 17, lines 21-28 – “the two or more object-detection models 113a-b may have been trained to identify the same object of interest but trained differently”)
Claim 1 further recites “processing the plurality of digital images using a second artificial intelligence model that is trained to detect anomalies and utilizes an autoencoder”. As cited in Nord, Nord teaches object detection system that can use multi-model system where two or more neural artificial intelligence models can be used for object detection, but do not explicitly teach using a second artificial intelligence model that is trained to detect anomalies and utilizes an autoencoder. However, Marciano teaches using a second artificial intelligence model that is trained to detect anomalies and utilizes an autoencoder (Marciano - paras 0107-0109 – “As explained with reference to FIG. 2A, a first detection is performed on the image of the item to detect the area(s) of the image including the concealing elements. The method of FIG. 4A can use at least part of the output (see operation 400) of the method of FIG. 2A. In particular, the method of FIG. 4A includes performing a second detection using the second software module 113. The second software module 113 includes at least one second neural network (e.g. DNN) which is trained to detect prohibited objects in an area including a concealing element of an item”; para 0111 – “For each given area which has been identified as including a concealing element by the method of FIG. 2A, the method of FIG. 4A includes detecting (operation 410) whether this given area includes a prohibited object.”; paras 0181-0182 – “The method of FIG. 6 is particularly beneficial when the second neural network is an Auto-Encoder, or a Generative Adversarial Network (the GAN can be combined with a discriminator). This is however not limitative, and other deep neural networks can be used. The given second neural network is therefore trained specifically for detecting, in an image, prohibited objects concealed using a concealing element of the given type. The given second neural network is trained to detect anomalies. Indeed, since it has been trained on images which are free of prohibited objects, if, during the prediction stage, it encounters an image with a prohibited object, it is able to provide a score informative of the probability of a presence of a prohibited object (presence of the prohibited object is viewed as an anomaly by the second neural network). Since a large training set of data is used (which covers various scenarios including unprohibited objects), this generally prevents from raising an alarm when an unprohibited object is concealed using the concealing element”. Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to replace Nord’s second artificial intelligence model with that of Marciano that is trained to detect anomalies and utilizes an autoencoder. A person having ordinary skill in the art would have been motivated before the effective filing date of the claimed invention to replace Nord’s second artificial intelligence model with that of Marciano that is trained to detect anomalies and utilizes an autoencoder, in order to flag strange or unusual deviation from normal data, which is an anomaly, as it sees an item/object of interest it was not trained to see; and autoencoder, as well known, when used provides unsupervised efficiency, as it learns normal image features without needing explicit examples of every possible concealed threat during training.
The combined invention of Nord and Marciano teaches “determining based on the processing performed using both the first artificial intelligence model and the second artificial intelligence model, that the plurality of digital images includes at least one suspect image” (see Marciano - paras 0107-0109 – “As explained with reference to FIG. 2A, a first detection is performed on the image of the item to detect the area(s) of the image including the concealing elements. The method of FIG. 4A can use at least part of the output (see operation 400) of the method of FIG. 2A. In particular, the method of FIG. 4A includes performing a second detection using the second software module 113. The second software module 113 includes at least one second neural network (e.g. DNN) which is trained to detect prohibited objects in an area including a concealing element of an item”; para 0111 – “For each given area which has been identified as including a concealing element by the method of FIG. 2A, the method of FIG. 4A includes detecting (operation 410) whether this given area includes a prohibited object.” – where second artificial intelligence model relies on output of first intelligence model, and the input images – thus the output of 2nd model is based on both 1st and 2nd model); and
The combined invention of Nord and Marciano further teaches “generating an alert regarding the suspect image” (See Nord – col.7, lines 49-57 – “the multi-model object-detection system may present one or more indications of one or more of the representative conclusions from the reconciled set of object-detection conclusions at a computer display or the like. For example, the multi-model object-detection system may present a graphical visualization of a bounding box and/or label for a given object of interest at a security screener's workstation at a screening checkpoint at an airport or the like”).
Regarding claim 2, the combined invention of Nord and Marciano teaches “The method of claim 1, wherein the alert includes an image of the individual with a graphical indicator identifying an area of concern” (See Nord – col.7, lines 49-57 – “the multi-model object-detection system may present one or more indications of one or more of the representative conclusions from the reconciled set of object-detection conclusions at a computer display or the like. For example, the multi-model object-detection system may present a graphical visualization of a bounding box and/or label for a given object of interest at a security screener's workstation at a screening checkpoint at an airport or the like”).
Regarding claim 6, the combined invention of Nord and Marciano teaches “The method of claim 1, further comprising: adjusting one or more of a confidence threshold and a size threshold of objects to be detected for either or both of the first artificial intelligence model and the second artificial intelligence model via a user interface.” (See Nord – col. 36, lines 34-45 – adjusting confidence thresholds; see Marciano’s – para 0060 – the weighting and/or threshold values of the DNN can be initially selected prior to training, and can be further iteratively adjusted or modified during training to achieve an optimal set of weighting and/or threshold values in a trained ML network).
Regarding claim 8, the combined invention of Nord and Marciano teaches “The method of claim 1, wherein the body scanner is a millimeter wave scanner” (see Nord - col. 13, lines 33-39 – “Each screening system 120 includes at least one detection device 122 that is configured to scan or otherwise capture one or more images of a scene. Each detection device 122 can take a variety of forms, including but not limited to an X-ray machine, metal detector, MRI scanner, CT scanner, millimeter wave scanner, spectral band scanner, or other type of scanning device”).
Regarding claim 12, claim 12 has been similarly analyzed and rejected as per citations made in the rejection of claim 1.
Regarding claim 13, claim 13 has been similarly analyzed and rejected as per citations made in the rejection of claim 2.
Regarding claim 14, the combined invention of Nord and Marciano teaches “The method of claim 12, further comprising: training the second artificial intelligence model in an unsupervised manner on a plurality of normal condition images of objects” (see Marciano as cited in the rejection of claim 1 – paras 0181-082 – Autoencoder trained on images which are free of prohibited objects (which are normal condition images); and autoencoders are inherently classified as unsupervised learning because they train on raw, unlabeled data without requiring human-provided target labels (clean/normal images as cited in Marciano).
Regarding claim 17, claim 17 has been similarly analyzed and rejected as per citations made in the rejection of claim 6.
Regarding claim 18, the combined invention of Nord and Marciano teaches “The method of claim 12, further comprising: adjusting a size threshold of objects to be detected via a user interface” (see Marciano – para 0125 – “data output can include e.g. the size and/or shape and/or type…of the prohibited object; para 0135 – different size categories would require different size thresholds to be selected, when performing detection).
Regarding claim 19, the combined invention of Nord and Marciano teaches “The method of claim 12, wherein the object scanner is a CT scanner or x-ray scanner” (see Nord - col. 13, lines 33-39 – “Each screening system 120 includes at least one detection device 122 that is configured to scan or otherwise capture one or more images of a scene. Each detection device 122 can take a variety of forms, including but not limited to an X-ray machine, metal detector, MRI scanner, CT scanner, millimeter wave scanner, spectral band scanner, or other type of scanning device”).
Regarding claim 23, claim 23 has been similarly analyzed and rejected as per citations made in the rejection of claim 1.
Regarding claim 24, claim 24 has been similarly analyzed and rejected as per citations made in the rejection of claim 2.
Regarding claim 27, claim 27 has been similarly analyzed and rejected as per citations made in the rejection of claim 6.
Regarding claim 32, claim 32 has been similarly analyzed and rejected as per citations made in the rejection of claim 1.
Regarding claim 33, claim 33 has been similarly analyzed and rejected as per citations made in the rejection of claim 2.
Regarding claim 36, claim 36 has been similarly analyzed and rejected as per citations made in the rejection of claim 17.
Regarding claim 37, claim 37 has been similarly analyzed and rejected as per citations made in the rejection of claim 18.
Regarding claim 41, claim 41 has been similarly analyzed and rejected as per citations made in the rejection of claim 1.
Regarding claim 42, claim 42 has been similarly analyzed and rejected as per citations made in the rejection of claim 8.
Regarding claim 43, claim 43 has been similarly analyzed and rejected as per citations made in the rejection of claim 1.
Regarding claim 44, claim 44 has been similarly analyzed and rejected as per citations made in the rejection of claim 19.
5. Claims 4-5, 15-16 and 35 are rejected under 35 U.S.C. 103 as being unpatentable over Nord et al., U.S. Patent No. 11,010,605 B2, further in view of Marciano et al., U.S. Patent Publication No. 2023/0186456 A1; and further in view of Bastian et al., U.S. Patent No. 12,656,277 B2.
Regarding claim 4, claim 4 recites “The method of claim 1, further comprising: using augmentations added to the normal condition images to train the anomaly detector”. The combined invention of Nord and Marciano as cited in the rejection of claim 1 teaches anomaly detector trained using normal condition images (see Marciano – para 0182), but does not teach “adding augmentations to the training data”. However, Bastian teaches “For example, in the experiment in which an anomaly detector trained using 2D radiograph data, further experimentation was performed to find the optimal set of hyperparameters in conjunction with data augmentation” (col. 16, lines 45-48). Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to use Bastian’s teachings of adding augmentations to the training data in the combined invention of Nord and Marciano. A person having ordinary skill in the art would have been motivated before the effective filing date of the claimed invention to use Bastian’s teachings of adding augmentations to the training data in the combined invention of Nord and Marciano, as adding augmentation helps to train an anomaly detector by teaching the model what normal variations look like, and stops false alarms by making the model tough.
Regarding claim 5, the combined invention of claim Nord, Marciano and Bastian teaches “The method of claim 4 wherein the augmentations include one or more of Gaussian noise, precision reduction, random shape cutouts, horizontal flips, Gaussian blur and face blur” (see Bastian - col. 16, lines 36-48 – “Data augmentation methodologies included rotations, flips, crops, and other classical image manipulation techniques. Data augmentation also included various forms of noise generation, including Gaussian noise”).
Regarding claim 15, claim 15 has been similarly analyzed and rejected as per citations made in the rejection of claim 4.
Regarding claim 16, claim 16 has been similarly analyzed and rejected as per citations made in the rejection of claim 5.
Regarding claim 35, claim 35 has been similarly analyzed and rejected as per citations made in the rejection of claim 4.
6. Claims 9-11, 20-22, 29-31 and 38-40 are rejected under 35 U.S.C. 103 as being unpatentable over Nord et al., U.S. Patent No. 11,010,605 B2, further in view of Marciano et al., U.S. Patent Publication No. 2023/0186456 A1; and further in view of Schneider et al., 2022, “Autoencoders – A comparative analysis in the realm of anomaly detection” (pp. 1986-1992).
Regarding claim 9, claim 9 recites “The method of claim 1 wherein the autoencoder is a convolutional autoencoder, a variational autoencoder or an adversarial autoencoder”. As cited in the rejection of claim1, Marciano teaches using an autoencoder, but does not explicitly teach “the autoencoder is a convolutional autoencoder, a variational autoencoder or an adversarial autoencoder”. However, as well-known autoencoder can be a convolutional autoencoder, a variational autoencoder or an adversarial autoencoder; and Schneider provides the evidentiary teachings. Schneider teaches in Section 2 – “Autoencoders for Anomaly Detection” that Autoencoders can be selected from convolutional autoencoder, a variational autoencoder or an adversarial autoencoder. Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to use Schneider’s teachings on autoencoder in the combined invention of Nord and Marciano. A person having ordinary skill in the art would have been motivated before the effective filing date of the claimed invention to use Schneider’s teachings on autoencoder in the combined invention of Nord and Marciano, in order to effectively select an autoencoder as per use case, for example, one would choose convolutional autoencoder as it demands the simplest implementation and training procedure, which leads to the fastest measured training times (see Schneider – page 1991 - right column – 3rd paragraph).
Regarding claim 10, claim 10 recites “The method of claim 1, wherein the autoencoder includes an encoder and decoder that is trained in an unsupervised manner and optimized to reduce reconstruction loss” (see Marciano as cited in the rejection of claim 1 teaches use of Autoencoder – Autoencoder as known inherently includes an encoder and decoder that is trained in an unsupervised manner and optimized to reduce reconstruction loss. Claim 10 presents the exact core definition and inherent mechanism of a standard autoencoder). Further citing Schneider – Schneider here is not required to be cited for inherent mechanism as cited in claim 10, but for reference purpose see Schneider sections 1 and 2 providing details on inherent structure of an autoencoder. A person having ordinary skill in the art would have been motivated before the effective filing date of the claimed invention to use Schneider’s teachings on autoencoder in the combined invention of Nord and Marciano, in order to effectively select an autoencoder as per use case.
Regarding claim 11, claim 11 recites “The method of claim 10, wherein teaches the encoder compresses the original image into a latent space representation, the decoder reconstructs the image from the latent space representation, and the loss between the original and reconstructed images is backpropagated to adjust the model weights” (see Marciano as cited in the rejection of claim 1 teaches use of Autoencoder – claim 11 presents the exact core definition and inherent mechanism of a standard autoencoder). Further citing Schneider – Schneider here is not required to be cited for inherent mechanism as cited in claim 11, but for reference purpose see Schneider sections 1 and 2 providing details on inherent structure of an autoencoder. A person having ordinary skill in the art would have been motivated before the effective filing date of the claimed invention to use Schneider’s teachings on autoencoder in the combined invention of Nord and Marciano, in order to effectively select an autoencoder as per use case.
Regarding claim 20, claim 20 has been similarly analyzed and rejected as per citations made in the rejection of claim 9.
Regarding claim 21, claim 21 has been similarly analyzed and rejected as per citations made in the rejection of claim 10.
Regarding claim 22, claim 22 has been similarly analyzed and rejected as per citations made in the rejection of claim 11.
Regarding claim 29, claim 29 has been similarly analyzed and rejected as per citations made in the rejection of claim 9.
Regarding claim 30, claim 30 has been similarly analyzed and rejected as per citations made in the rejection of claim 10.
Regarding claim 31, claim 31 has been similarly analyzed and rejected as per citations made in the rejection of claim 11.
Regarding claim 38, claim 38 has been similarly analyzed and rejected as per citations made in the rejection of claim 9.
Regarding claim 39, claim 39 has been similarly analyzed and rejected as per citations made in the rejection of claim 10.
Regarding claim 40, claim 40 has been similarly analyzed and rejected as per citations made in the rejection of claim 11.
7. Claims 3, 7, 25-26 and 28 are objected to as being dependent upon a rejected base claim 1, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Manav Seth whose telephone number is (571) 272-7456. The examiner can normally be reached on Monday to Friday from 8:30 am to 5:00 pm.
If attempts to reach the examiner by telephone are unsuccessful, the examiner's supervisor, Sumati Lefkowitz, can be reached on (571) 272-3638. 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:/Awww.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
/Manav Seth/
Primary Examiner, Art Unit 2672
August 17, 2026
Prosecution Insights