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 .
DETAILED ACTION
INFORMATION DISCLOSURE STATEMENT
The information disclosure statement (IDS) submitted on 2/15/24, 5/30/24, 9/12/24, 10/18/24, 3/18/25, 5/1/25, 10/9/25, 2/10/26 & 5/6/26 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
FOREIGN PRIORITY
A claim for foreign priority under 35 U.S.C § 119 (a) - (d), which was contained in the Declaration and Power of Attorney filed on 2/27/24 has been acknowledged. Acknowledgement of claimed foreign priority and receipt of priority documents is reflected in form PTO-326 Office Action Summary.
RESTRICTION RESPONSE
Restriction Requirement
Applicant's election with traverse of restriction requirement in the reply filed on 04/20/2026 is acknowledged. After carefully reviewing applicant arguments, restriction guidance and claim limitations, examiner respectfully disagrees.
Applicant Argument 1
Applicant submits claims all relate to identifying changes over time based on differences between images, and all claims use neural networks.
In response, in this case, Group I includes a skip connection between the input of the CNN encoder and the input of the ConvLSTM network. Group II discloses a phase data embodiment in which ConvLSTM input data further comprises phase data and the ConvLSTM processing includes convolving the phase data with the plurality of feature maps. Group III discloses a training computing network including pre-training the CNN encoder and training both CNN encoder and ConvLSTM network based on first/second feature classification data. Therefore, examiner submits even though claim groups maybe related they are distinct. MPEP 803 & 807 permit restriction where distinct claimed inventions would impose a serious search and/or examination burden. In view of above arguments, examiner submits restriction basis is sufficient and respectfully maintained.
Applicant Argument 2
Applicant submits Group III is linked to Group I because it trains the neural networks used in claim 23. Specifically, applicant submits Group III is clearly linked to Group I because claim 23 includes the CNN encoder and ConvLTSM network.
In response, examiner submits those claim groupings require different prior art searching: Group I requires searching operational CNN/ConvLSTM image change detection systems, while Group III requires searching neural network training strategies, imbalanced/scarace training data, pre-training, and frozen/unfrozen encoder weights. In view of above arguments, examiner submits restriction basis is sufficient and respectfully maintained.
Applicant Argument 3
Applicant submits Group II claim 12 should not be separated because phase data is spatial, not temporal.
In response, claim 12 requires that the ConvLSTM input data further comprises phase data indicative of a respective phase value of each pixel of each image and that propagating through the ConvLSTM includes convolving the phase data with the plurality of feature maps. Applicant claim 12 requires a different field of search from the broader CNN/ConvLTSM skip connection claims of Group I. In view of above arguments, examiner submits restriction basis is sufficient and respectfully maintained.
Applicant Argument 4
Applicant submits examiner classification rationale is weak because subclasses overlap, therefore no serious search burden exist. Applicant submits claim groups all relate to G06V 10/82 image analysis using neural networks, and all claims relate to a temporal dimension because they analyze changes between successive images. Applicant concludes submitting the same subclasses would need to be searched, so there is not substantial burden.
In response, classification overlap does not eliminate restriction when the claim limitations require different searches. MPEP 803 states serious search burden may be shown by separate classification, separate status in the art, or a different field of search, but classification is not the only basis for burden.
In this case, Group I includes a skip connection between the input of the CNN encoder and the input of the ConvLSTM network. Group II discloses a phase data embodiment in which ConvLSTM input data further comprises phase data and the ConvLSTM processing includes convolving the phase data with the plurality of feature maps. Group III discloses a training computing network including pre-training the CNN encoder and training both CNN encoder and ConvLSTM network based on first/second feature classification data. In view of above arguments, examiner submits restriction basis is sufficient and respectfully maintained.
Applicant Argument 5
Applicant submits claim 12 should be grouped with Group I because it overlaps with claim 11 as they both recite phase data.
In response, examiner submits areas of overlap is not the test for restriction. Claim 11 recites phase data as part of CNN input data and recites that the feature classification is based at least in part on the phase data. Claim 12 separately recites that ConvLSTM input data further comprises phase data and that ConvLSTM propagation includes convolving the phase data with the plurality of feature maps. The difference between claim scopes are sufficient to maintain a separate search burden for the Group II phase/ConvLSTM. In view of above arguments, examiner submits restriction basis is sufficient and respectfully maintained.
Applicant Argument 6
Applicant submits Group III does not mention satellite imaging, so its classification is wrong. Applicant submits Group III should not be classified in G06V 20/13 for satellite image scene analysis because Group III claims do not expressly mention satellite imaging.
In response, distinctness in claim sets requiring sufficient burden for restriction is grounded in group III reciting a method of training the computing network, including providing first/second feature classification training data, pre-training the CNN encoder, training both the CNN encoder and ConvLSTM network, and limitations concerning scarce training data and frozen/unfrozen weights. Examiner submits the issue of classification correction would not eliminate the need for a separate search based on claim scope differences. In view of above arguments, examiner submits restriction basis is sufficient and respectfully maintained.
Applicant Argument 7
Applicant submits all claims use neural networks, so they should be examined together. In response, neural networks is too high level to establish a single invention. Inventions in many different classifications include neural networks. The relevant question is whether the claimed inventions require distinct searches.
In response, the critical inquiry here is whether the claimed invention require different searches. Group I requires a CNN encoder and ConvLSTM network connected through a skip connection for image change map generation. Claim 1 expressly recites propagating a copy of the CNN input data through the skip connection and convolving that copied image data with respective feature maps. Group II claim 12 requires phase data in the ConvLSTM input and convolving the phase data with feature maps. Group III requires training the CNN encoder and ConvLSTM network, including pre-training the CNN encoder and training both networks using first/second feature classification data. Examiner submits these groups contain different technical mechanisms and require different searches. In view of above arguments, examiner submits restriction basis is sufficient and respectfully maintained.
Applicant Argument 8
Applicant submits no substantial burden because the limitations overlap.
In response, the critical inquiry here is whether the claimed invention require different searches. Group I requires a CNN encoder and ConvLSTM network connected through a skip connection for image change map generation. Claim 1 expressly recites propagating a copy of the CNN input data through the skip connection and convolving that copied image data with respective feature maps. Group II claim 12 requires phase data in the ConvLSTM input and convolving the phase data with feature maps. Group III requires training the CNN encoder and ConvLSTM network, including pre-training the CNN encoder and training both networks using first/second feature classification data. Examiner submits these groups contain different technical mechanisms and require different searches. In view of above arguments, examiner submits restriction basis is sufficient and respectfully maintained.
After reviewing applicant restriction arguments are not found persuasive as MPEP 802.01 discloses that only invention should be claimed per patent application. The requirement is still deemed proper and is therefore made FINAL.
Claims 12, 25-28 & 31 are withdrawn from further consideration pursuant to 37 CFR 1.142(b), as being drawn to a nonelected group/species, there being no allowable generic or linking claim. Applicant timely traversed the restriction (election) requirement in the reply filed on 04/10/2026. Applicant is reminded claims 12, 25-28 & 31 are required to be cancelled.
CLAIM REJECTIONS - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-4, 6-8, 11, 13-22 & 30 are rejected under 35 U.S.C. 101 because the claimed invention is directed to as ineligible under subject eligibility test. In the Subject Matter Eligibility Test for Products and Processes (Federal Register, Vol. 79, No. 241, dated Tuesday, December 16, 2014, page 74621), The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the additional device elements, which are recited at a high level of generality, provide conventional computer functions that do not add meaningful limits to practicing the abstract idea.
Claim 1
Step 1
This step inquires “is the claim to a process, article of machine, manufacture or composition of matter?” Yes,
Claim 1 – “Method” is a process.
Step 2A - Prong 1
This step inquires “does the claim recite an abstract idea, law or natural phenomenon”. This claim appears to directed to an abstract idea.
The limitation of “receiving, at a convolutional neural network (CNN) encoder, CNN input data comprising data associated with each pixel of each of the plurality of images; propagating the CNN input data through the CNN encoder to generate a plurality of feature maps, wherein each feature map comprises a feature classification of each pixel of a respective image of the plurality of images according to a feature classification scheme, wherein the feature classification scheme is generated by the CNN encoder based on training data; providing a skip connection between an input of the CNN encoder and an input of a convolutional Long Short-Term Memory (ConvLSTM) network; propagating a copy of the CNN input data to the input of the ConvLSTM network through the skip connection; convolving the data associated with each of the plurality of images in the copy of the CNN input data with its respective feature map generated by the CNN encoder, to generate the ConvLSTM input data, the ConvLSTM input data comprising the plurality of feature maps generated by the CNN encoder; receiving, at ConvLSTM network the ConvLSTM input data; and propagating the ConvLSTM input data through the ConvLSTM network to generate a change map, wherein the change map comprises change data indicative of one or more changes across the plurality of images.”, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind (e.g. mathematical concepts, mental processes or certain methods of organizing human activity) but for the recitation of generic computer components.
STEP 2A – PRONG 1 - CONCLUSION
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claim recites an abstract idea.
Step 2A - Prong 2
This step inquires “does the claim recite additional elements that integrate the judicial exception into a practical application”. This judicial exception is not integrated into a practical application
STEP 2A – PRONG 2 - CONCLUSION
Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea.
Step 2B
The critical inquiry here is does the claim recite additional elements that amount to “significantly more” than the judicial exception? The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claim is not patent eligible.
Dependent Claims
As to claim 2, this claim is directed to mental process (“observing/measuring image differences and quantifying the degree of change”) and insignificant extra-solution activity (“outputting of a quantitative result”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claim 3, this claim is directed to mental process (“comparing two-pixel classification and determining same/different maps easily to observation/comparison/judgment”) and insignificant extra-solution activity (“outputting a label after analysis”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claim 4, this claim is directed to mental process (“binary changed/unchanged classification”) and insignificant extra-solution activity (“output labeling”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claim 6, this claim is directed to insignificant extra-solution activity (“specifies type of input data.”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claim 7, this claim is directed to mental process (“object classification into two categories”) and insignificant extra-solution activity (“abstract classification framework”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claim 8, this claim is directed to insignificant extra-solution activity (“Property of the training data”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claim 11, this claim is directed to insignificant extra-solution activity (“Input Data Type”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claim 13, this claim is directed to mental process (“Visual Observation and Temporal Comparison”) and insignificant extra-solution activity (“Use Context Data Selection”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claim 14, this claim is directed to insignificant extra-solution activity (“Temporal Processing”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claim 15, this claim is directed to insignificant extra-solution activity (“Property of input images or technical environment”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claim 16, this claim is directed to insignificant extra-solution activity (“Scale Field of Use Limitation”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claim 17, this claim is directed to insignificant extra-solution activity (“Output/Capacity Result”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claim 18, this claim is directed to mental process (“Observation, Classification & Comparison”) and insignificant extra-solution activity (“Field of Use”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claim 19, this claim is directed to mental process (“Observational Classification Task”) and insignificant extra-solution activity (“Application/Field of Use”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claim 20, this claim is directed to insignificant extra-solution activity (“field data source limitation”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claim 21, this claim is directed to insignificant extra-solution activity (“Data Source Limitation”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claim 22, this claim is directed to insignificant extra-solution activity (“data source”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claim 30, this claim is directed to mental process (“CRM format does not change claim 1 abstract idea”) and insignificant extra-solution activity (“CRM implementation of the method”). Thus, this claim does not integrate the abstract idea into a practical application or constitute significantly more than the abstract.
As to claims 9-10 & 23, these claims are rejected due to their dependence on claim 1 and are rejected for the same reasons.
CLAIM REJECTIONS - 35 USC § 103
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 of this title, 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.
Claims 1-4, 7, 10, 13, 23 & 30 are rejected under 35 U.S.C. 103 as being unpatentable over
Gubbi et al. (U.S. Publication 2021/0065354) in view of Braman et al. (U.S. Publication 2021/0110532) & Rhodes et al. (U.S. Publication 2020/0160528)
As to claims 1 & 23, Gubbi discloses a computer-implemented method for identifying one or more changes across a plurality of images ([0003] discloses change detection is performed to identify changes that may have happened in a particular place and/or on an object being considered, over a period of time.), ([0021] discloses a test image and a reference image of the scene/object are fed as input to the system 100.)([0034] discloses a reference image and a test image as input, wherein the reference image and the test image are of the same scene, captured at different time points.)
the method comprising: receiving, at a convolutional neural network (CNN) encoder, CNN input data comprising data associated with each pixel of each of the plurality of images;
([0024] discloses I1 and I2 are fed as inputs to the adaptive correlation layer. Each set of the Siamese CNN architecture has 6 convolution layers each, the convolution layers extract features containing semantic information about the test image and the reference image. )
([0028] discloses in the images being processed, the information may not be uniformly distributed across all the pixels in the images. Some of the pixels may contain more information in comparison with other pixels in the image.)
propagating the CNN input data through the CNN encoder to generate a plurality of feature maps,
([0024] discloses the convolution layers extract features containing semantic information about the test image and the reference image.)([0035] discloses the system 100 generates at least one feature map separately for the test image and the reference image. The feature map of any image contains information on features specific to semantic information in the images.)
wherein each feature map comprises a feature classification of each pixel of a respective image of the plurality of images according to a feature classification scheme. ([0025] discloses this results in an output of T2 correlation values for every pixel in f1. )([0036] discloses step 404 the system 100 extracts semantic features from the feature map of the test image and the reference image. Assign every pixel with probability of having a change.)
wherein the feature classification scheme is generated by the CNN encoder based on training data;
([0027] discloses introduction of new parameterized layers in the network enables training it end-to end, the network learns to identify changes in a pair of images from any scene;)
propagating the ConvLSTM input data through the ConvLSTM network to generate a change map,
([0036] discloses step 406, the system 100 computes at least one correlation map between the feature maps of the test image and the reference image, by computing correlation using (1). step 408, the system 100 determines one or more semantic changes, by processing the correlation map. step 408, the system 100 passes the computed at least one correlation map through a plurality of 2D convolution layers and a softmax layer )
wherein the change map comprises change data indicative of one or more changes across the plurality of images.
([0036] discloses assign every pixel with probability of having a change.)([0037] discloses
the system 100 highlights differences between the test image and the reference image at output. The system 100 outputs the region containing the person as the identified semantic change.)
Gubbi is silent to a convolutional Long Short-Term Memory (ConvLSTM) network; receiving, at ConvLSTM network the ConvLSTM input data; the ConvLSTM input data comprising the plurality of feature maps generated by the CNN encoder; providing a skip connection between an input of the CNN encoder and an input of a convolutional Long Short-Term Memory (ConvLSTM) network; propagating a copy of the CNN input data to the input of the ConvLSTM network through the skip connection; convolving the data associated with each of the plurality of images in the copy of the CNN input data with its respective feature map generated by the CNN encoder, to generate the ConvLSTM input data, the ConvLSTM input data comprising the plurality of feature maps generated by the CNN encoder.
However, Braman discloses a convolutional Long Short-Term Memory (ConvLSTM) network;
([0001] discloses a detection model that uses convolutional long short-term memory (Conv-LSTM) units to scan through an image volume and remember signatures of a disease on and between slices)
([0030] discloses Conv-LSTM is a variant of LSTM that uses convolutional operations to distinguish variations in spatial patterns. Conv-LSTM excels in detecting spatiotemporal patterns.)
receiving, at ConvLSTM network the ConvLSTM input data ([0031] discloses
Features extracted from each slice individually by previous convolutional layers are fed one at a time into a convolutional LSTM)([0032] discloses each unit 102 features two 2D convolutional layers 104, which extract features from each 2D slice of a 3D volume separately. The outputs from the convolutional layers for each slice are then processed sequentially by a Conv-LSTM layer 108)
propagating a copy of the CNN input data to the input of the ConvLSTM network;
([0031] discloses the Conv-LSTM implementation uses convolutional Conv-LSTM to “scan” through an imaging volume The convolutional LSTM has the capability to “remember” image patterns associated with a disease from previous slices, as well as identify changes between slices that are indicative of a disease.)([0033] discloses The last Conv-LSTM layer 108 in the network 100 outputs a single set of features, which represents the network's findings after processing through the imaging volume multiple times.)
It would have been obvious to one of ordinary skill in the art at the time of effective filing to modify Gubbi’s disclosure to include the above limitations in order to account for spatial temporal relationships across the plurality of images before outputting the pixel level change map.
Gubbi in view of Braman is silent to providing a skip connection between an input of the CNN encoder and an input of a convolutional Long Short-Term Memory (ConvLSTM) network;
propagating a copy of the CNN input data to the input of the ConvLSTM network through the skip connection; convolving the data associated with each of the plurality of images in the copy of the CNN input data with its respective feature map generated by the CNN encoder, to generate the ConvLSTM input data, the ConvLSTM input data.
However, Rhodes discloses CNN input data comprising data associated with each pixel.
([0026] discloses the term frame in the context of a CNN input indicates a 2D data structure having a feature value for each pixel of the frame.)
A copy of the CNN input data. ([0027] discloses context feature volume indicates features that are from and provide context to the current video frame.)([0036] discloses context feature volume 130 may include current video frame 111 (Xt) of input video 110, a previous video frame 112 (Xt−1) of input video 110,)
Feature map generated by the CNN encoder.
([0028] discloses the object classification convolutional neural network is applied to the current video frame and, for some or all of the convolutional layers of the object classification convolutional neural network, feature values are attained.)([0047] discloses after application, multiple feature maps may be retrieved from the object classification CNN such that each feature map corresponds to a layer of the object classification CNN with each feature map having a feature value corresponding to a pixel of current video frame 111. )([0049] discloses features volume 130 are feature maps extracted from convolutional layers of the object detection CNN. feature maps from convolutional layers may be copied and stacked to form features volume 130, which includes a volume of pixel wise features.) Skip connection.
([0031] discloses the segmentation network includes context aware skip connections. the term context aware skip connection indicates a skip connection that combines (e.g., concatenates) an output from a previous convolutional layer with the previously discussed context feature volume to generate a convolutional layer input volume for an immediately next convolutional layer of the segmentation network. the context aware skip connections discussed herein provide the context feature volume (e.g., current video frame, previous video frame, etc.) as input to some or all of the convolutional layers of the segmentation network. )
Propagating a copy of the CNN input data through the skip connection.
([0074] discloses the context feature volume is also provided, along with feature frames 119, to the first layer of the segmentation network. Such context aware skip connections do not provide skip connections for the convolutional layer output but instead provide a skip connection for the context feature volume to each (or one or more) convolutional layer subsequent to the first layer of the segmentation network.)
([0075] discloses Such context aware connections allow features from the context feature volume to bypass layers and remain undiluted to deep convolutional layers)
Convolving the data associated with each of the plurality of images in the copy of the CNN input data with its respective feature map generated by CNN encoder.
([0078] discloses convolutional layer output volume 812 is then concatenated with context feature volume 130 at concatenation operation 822 to generate a convolutional layer input volume 813 including a combination of context feature volume 130 and convolutional layer output volume 812. Convolutional layer input volume 813 is provided to second convolutional layer 802, which processes the convolutional layer input volume 813 by applying any number of convolutional filters using pretrained filter weights.)([0082] discloses these features (as provided by context feature volume 130) are concatenated with the previous convolutional layer output and passed to the current convolutional layer of segmentation network 800.)
It would have been obvious to one of ordinary skill in the art at the time of effective filing to modify Gubbi in view of Braman’s disclosure to include the above limitations in order to preserve low level pixel image context together with CNN generated feature maps at the recurrent input stage, thereby reducing loss of task relevant image context while the ConvLSTM models spatial temporal changes across the plurality of images.
As to claim 2, Gubbi in view of Braman & Rhodes discloses everything as disclosed in claim 1. In addition, Gubbi discloses wherein the change data includes quantitative data indicative of the degree of the one or more changes across the plurality of images. ([0025] discloses
This results in an output of T2 correlation values for every pixel in f1.)([0036] discloses the system 100 processes the at least one feature map of the test image and the reference image. the system 100 passes the computed at least one correlation map through a plurality of 2D convolution layers and a softmax layer to assign every pixel with probability of having a change.)
As to claim 3, Gubbi in view of Braman & Rhodes discloses everything as disclosed in claim 1. In addition, Gubbi discloses the change data includes a change classification of each pixel of a selected image of the plurality of images, and for a given pixel of the selected image, the change classification of said pixel is indicative of whether the feature classification for said pixel is the same as or different from the feature classification for a corresponding pixel of another of the plurality of images. ([0024] discloses the correlation layer in the adaptive correlation layer computes pixel similarity. )([0025] discloses this results in an output of T2 correlation values for every pixel in f1.)([0036] discloses assign every pixel with probability of having a change. [0037] discloses the correlation map between the feature maps of these two images show lower correlation in the region where the person is present.)
As to claim 4, Gubbi in view of Braman & Rhodes discloses everything as disclosed in claim 3. In addition, Gubbi discloses wherein the change classification is a binary classification. ([0025] discloses a computed correlation map is then passed onto set of convolutional layers to obtain a binary segmentation map.)
As to claim 7, Gubbi in view of Braman & Rhodes discloses everything as disclosed in claim 1. In addition, Gubbi discloses wherein the feature classification scheme is a binary classification scheme configured to classify identified objects as belonging to either a first feature classification or a second feature classification. ([0025, 0033])
As to claim 10, Gubbi in view of Braman & Rhodes discloses everything as disclosed in claim 1. In addition, Rhodes discloses wherein propagating the CNN input data through the CNN encoder to generate the plurality of feature maps includes compressing the CNN input data. ([0028, 0030, 0046, 0068, 0070])
As to claim 13, Gubbi in view of Braman & Rhodes discloses everything as disclosed in claim 1. In addition, Gubbi discloses wherein each of the plurality of images is an image of a common target imaged at respectively different times, such that identifying the one or more differences across the plurality of images is equivalent to identifying one or more changes over time of the subject. ([0021, 0034])
As to claim 30, Gubbi in view of Braman & Rhodes discloses everything as disclosed in claim 1. In addition, Gubbi discloses A non-transitory computer-readable medium comprising computer executable instructions stored thereon which, when executed by a computer, cause the computer to carry out the method of claim 1. ([0039])
Claims 6, 11 & 15 are rejected under 35 U.S.C. 103 as being unpatentable over
Gubbi et al. (U.S. Publication 2021/0065354) in view of Braman et al. (U.S. Publication 2021/0110532) & Rhodes et al. (U.S. Publication 2020/0160528) as applied in claim 1 above further in view of Cha et al. (U.S. Publication 2017/0061217)
As to claim 6, Gubbi in view of Braman & Rhodes discloses everything as disclosed in claim 1 but is silent to wherein the CNN input data includes amplitude data indicative of one or more amplitude values associated with each of the pixels of each of the plurality of images.
However, Cha discloses wherein the CNN input data includes amplitude data indicative of one or more amplitude values associated with each of the pixels of each of the plurality of images. ([0003-0004, 0039])
It would have been obvious to one of ordinary skill in the art at the time of effective filing to modify Gubbi in view of Braman & Rhodes’s disclosure to include the above limitations in order to provide pixel level reflected signal intensity information for detecting image changes based on amplitude differences between corresponding pixels.
As to claim 11, Gubbi in view of Braman & Rhodes discloses everything as disclosed in claim 1 but is silent to wherein the CNN input data further comprises phase data indicative of a respective phase value of each pixel of each image of the plurality of images, and wherein the feature classification of each pixel of each image by its respective feature map is based, at least in part, on said phase data.
However, Cha discloses wherein the CNN input data further comprises phase data indicative of a respective phase value of each pixel of each image of the plurality of images, and wherein the feature classification of each pixel of each image by its respective feature map is based, at least in part, on said phase data. ([0003-0004, 0034, 0043])
It would have been obvious to one of ordinary skill in the art at the time of effective filing to modify Gubbi in view of Braman & Rhodes’s disclosure to include the above limitations in order to allow the neural network change detector to use coherent phase based scattering information for detecting small scale pixel level changes.
As to claim 15, Gubbi in view of Braman & Rhodes discloses everything as disclosed in claim 1 but is silent to wherein each of the plurality of images is coherent with each of the other images.
However, Cha discloses wherein each of the plurality of images is coherent with each of the other images. ([0001-0006, 0041-0045])
It would have been obvious to one of ordinary skill in the art at the time of effective filing to modify Gubbi in view of Braman & Rhodes’s disclosure to include the above limitations in order to detect small scale changes using phase coherence information that would not be available from non-coherent image comparison alone.
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Gubbi et al. (U.S. Publication 2021/0065354) in view of Braman et al. (U.S. Publication 2021/0110532) & Rhodes et al. (U.S. Publication 2020/0160528) as applied in claim 1 above further in view of Du et al. (U.S. Publication 2020/0160533)
As to claim 8, Gubbi in view of Braman & Rhodes discloses everything as disclosed in claim 7 but is silent to wherein the training data used to train the neural network comprises data representative of both the first and second feature classifications, and wherein the data representative of the first feature classification within the training data is scarce relative to the data representative of the second feature classification.
However, Du discloses wherein the training data used to train the neural network comprises data representative of both the first and second feature classifications, and wherein the data representative of the first feature classification within the training data is scarce relative to the data representative of the second feature classification. ([0084, 0102])
It would have been obvious to one of ordinary skill in the art at the time of effective filing to modify Gubbi in view of Braman & Rhodes’s disclosure to include the above limitations in order to compensate for underrepresentation of the scarce class during training and improve the learned binary feature classification.
Claims 14 & 16-19 are rejected under 35 U.S.C. 103 as being unpatentable over
Gubbi et al. (U.S. Publication 2021/0065354) in view of Braman et al. (U.S. Publication 2021/0110532) & Rhodes et al. (U.S. Publication 2020/0160528) as applied in claim 1 above further in view of Asner (U.S. Publication 2013/0216103)
As to claim 14, Gubbi in view of Braman & Rhodes discloses everything as disclosed in claim 1 but is silent to wherein the plurality of images comprises successive images and the method further comprises: propagating the ConvLSTM input data through the ConvLSTM network and convolving the ConvLSTM input data respectively associated with each of the successive images with the ConvLSTM input data associated with a respectively preceding image to generate successive change maps, wherein each successive change map is representative of a change between one of the plurality of images and a successive image.
However, Asner discloses wherein the plurality of images comprises successive images and the method further comprises: propagating the ConvLSTM input data through the ConvLSTM network and convolving the ConvLSTM input data respectively associated with each of the successive images with the ConvLSTM input data associated with a respectively preceding image to generate successive change maps, wherein each successive change map is representative of a change between one of the plurality of images and a successive image. ([0013, 0038-0044])
It would have been obvious to one of ordinary skill in the art at the time of effective filing to modify Gubbi in view of Braman & Rhodes’s disclosure to include the above limitations in order to output a separate temporal change representation for each successive image pair.
As to claim 16, Gubbi in view of Braman & Rhodes discloses everything as disclosed in claim 1 but is silent to wherein each of the plurality of images is an image of an area of 10 square kilometres or more, 50 square kilometres or more, 100 square kilometres or more, 1000 square kilometres or more, 5000 square kilometres or more, or 10 000 square kilometres or more.
However, Asner discloses wherein each of the plurality of images is an image of an area of 10 square kilometres or more, 50 square kilometres or more, 100 square kilometres or more, 1000 square kilometres or more, 5000 square kilometres or more, or 10 000 square kilometres or more. ([0020-0021, 0103])
It would have been obvious to one of ordinary skill in the art at the time of effective filing to modify Gubbi in view of Braman & Rhodes’s disclosure to include the above limitations in order to perform regional scale geographic change detection over large remote sensing image collections.
As to claim 17, Gubbi in view of Braman, Rhodes & Asner discloses everything as disclosed in claim 14 but is silent to wherein the change map is configured to resolve spatial features with a size of 50 metres or less, 10 metres or less, 5 metres or less, or 1 metre or less.
However, Asner discloses wherein the change map is configured to resolve spatial features with a size of 50 metres or less, 10 metres or less, 5 metres or less, or 1 metre or less. ([0018-0022, 0038-0044])
It would have been obvious to one of ordinary skill in the art at the time of effective filing to modify Gubbi in view of Braman & Rhodes’s disclosure to include the above limitations in order to detect and map remote sensing change features at a spatial resolution finer than 50 meters.
As to claim 18, Gubbi in view of Braman & Rhodes discloses everything as disclosed in claim 1 but is silent to wherein: each of the plurality of images is an image of a geographical area, and the feature classification scheme includes (a) a first feature classification indicating that a pixel classified as such is representative of the presence of a predetermined geographical feature and (b) a second feature classification indicating that a pixel classified as such is representative of the absence of the predetermined geographical feature, and the method further comprises identifying areas where the presence/absence of the predetermined geographical feature changes based on the identified differences across the plurality of images.
However, Asner discloses wherein: each of the plurality of images is an image of a geographical area, and the feature classification scheme includes (a) a first feature classification indicating that a pixel classified as such is representative of the presence of a predetermined geographical feature and (b) a second feature classification indicating that a pixel classified as such is representative of the absence of the predetermined geographical feature, and the method further comprises identifying areas where the presence/absence of the predetermined geographical feature changes based on the identified differences across the plurality of images.
([0034, 0039, 0041])
It would have been obvious to one of ordinary skill in the art at the time of effective filing to modify Gubbi in view of Braman & Rhodes’s disclosure to include the above limitations in order to identify where the presence or absence of a geographical feature changes over time.
As to claim 19, Gubbi in view of Braman, Rhodes & Asner discloses everything as disclosed in claim 18 but is silent to wherein: the first feature classification is a forest classification indicating that a pixel classified as such is representative of forested land, the second feature classification is a non-forest classification indicating that a pixel classified as such is representative of land that is not forested, and the method further comprises identifying changes in sizes of areas of deforestation around forested land based on the identified differences across the plurality of images.
However, Asner discloses wherein: the first feature classification is a forest classification indicating that a pixel classified as such is representative of forested land, the second feature classification is a non-forest classification indicating that a pixel classified as such is representative of land that is not forested, and the method further comprises identifying changes in sizes of areas of deforestation around forested land based on the identified differences across the plurality of images. ([0039, 0041, 0104])
It would have been obvious to one of ordinary skill in the art at the time of effective filing to modify Gubbi in view of Braman, Rhodes & Asner’s disclosure to include the above limitations in order to identify and quantify changes in forested versus non-forested land over time.
Claims 20-22 are rejected under 35 U.S.C. 103 as being unpatentable over
Gubbi et al. (U.S. Publication 2021/0065354) in view of Braman et al. (U.S. Publication 2021/0110532) & Rhodes et al. (U.S. Publication 2020/0160528) as applied in claim 1 above further in view of Edinger et al. (U.S. Publication 2019/0025422)
As to claim 20, Gubbi in view of Braman & Rhodes discloses everything as disclosed in claim 1 but is silent to wherein each of the plurality of images is generated by synthetic aperture radar imaging.
However, Edinger discloses wherein each of the plurality of images is generated by synthetic aperture radar imaging. ([0002-0015, 0038-0044])
It would have been obvious to one of ordinary skill in the art at the time of effective filing to modify Gubbi in view of Braman & Rhodes’s disclosure to include the above limitations in order to provide radar based image data suitable for all weather remote sensing change detection.
As to claim 21, Gubbi in view of Braman & Rhodes discloses everything as disclosed in claim 1 but is silent to wherein each of the plurality of images is a generated from data acquired by a satellite.
However, Edinger discloses wherein each of the plurality of images is a generated from data acquired by a satellite. ([0010-0018, 0018-0022, 0040-0044])
It would have been obvious to one of ordinary skill in the art at the time of effective filing to modify Gubbi in view of Braman & Rhodes’s disclosure to include the above limitations in order to obtain repeat remote sensing image data for change detection over geographically distributed regions.
As to claim 22, Gubbi in view of Braman, Rhodes & Edinger discloses everything as disclosed in claim 21 but is silent to wherein each of the images is generated from data acquired by a satellite in a low-earth orbit.
However, Edinger discloses wherein each of the images is generated from data acquired by a satellite in a low-earth orbit. ([0026-0032])
It would have been obvious to one of ordinary skill in the art at the time of effective filing to modify Gubbi in view of Braman, Rhodes & Edinger’s disclosure to include the above limitations in order to satellite acquired SAR image data with sufficient coverage and signal quality for remote sensing change detection.
CONCLUSION
No prior art has been found for claim 9 in its current form.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Stephen P Coleman whose telephone number is (571)270-5931. The examiner can normally be reached Monday-Thursday 8AM-5PM.
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, Andrew Moyer can be reached at (571) 272-9523. 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.
Stephen P. Coleman
Primary Examiner
Art Unit 2675
/STEPHEN P COLEMAN/Primary Examiner, Art Unit 2675
Prosecution Insights