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 .
Priority
Acknowledgment is made of applicant’s claim for foreign priority under 35 U.S.C. 119 (a)-(d). The certified copy has been filed in parent Application No. CN202111160570.3, filed on 09/30/2021.
Response to Arguments
Applicant’s argument filed 01/12/2026 have been fully considered but they are not persuasive.
Applicant’s Argument: On page 12 of Applicant’s response to rejections under 35 U.S.C. 101, applicant states that the amended subject matter is integrated into a practical application and provides significantly more than the abstract idea because the predicted land usage property is used to establish supporting facilities.
Examiner’s Response: Applicant’s argument is not persuasive. During examination, the examiner should analyze the "improvements" consideration by evaluating the specification and the claims to ensure that a technical explanation of the asserted improvement is present in the specification, and that the claim reflects the asserted improvement (see MPEP §2106.05(a)). The MPEP (§2106.05(a)(II)) also warns, “it is important to keep in mind that an improvement in the abstract idea itself (e.g. a recited fundamental economic concept) is not an improvement in technology.” Here, the alleged improvement in the form of “a predicted land usage property of the target region at a next time, wherein the predicted land usage property is used to establish supporting facilities corresponding to the predicted land usage property in the target region” is an improvement to the abstract idea of a mental process that can be performed in the human mind with the aid of pen and paper.
An important consideration in determining whether a claim improves technology is the extent to which the claim covers a particular solution to a problem or a particular way to achieve a desired outcome, as opposed to merely claiming the idea of a solution or outcome (see MPEP 2106.05(a)). The amended claims do not provide sufficient details to describe any technological improvement. If the specifications explicitly set forth an improvement but in a conclusory manner (see MPEP 2106.04(d)(1): a bare assertion of an improvement without the detail necessary to be apparent to a person of ordinary skill in the art), the examiner should not determine the claim improves technology.
Applicant’s Argument: On page 14-15 of Applicant’s response to rejections under 35 U.S.C. 103, applicant states that Pijanowski teaches net population growth value for a certain region and does not explicitly reflect the flow of people between two or more regions.
Examiner’s Response: Applicant’s argument is not persuasive. Examiner does not agree with this interpretation, but in order to advance prosecution, Examiner provided an additional reference to explicitly disclose the rate of migration from one region to another. Bright teaches the inflow and outflow migration of the population by county. This teaching directly aligns with Applicant’s interpretation of claim 1 and describes the flow of people between two or more regions. In claim 1, the claim limitation “the human interaction information comprises at least one selected from a flow frequency of human moving from one of the plurality of regions to another one of the plurality of regions at the specified time or a region retrieval frequency of human in one of the plurality of regions retrieving information for another one of the plurality of regions at the specified time” does not specify that the human interaction information is comprises of both elements. The human interaction information is only comprised of one of the two elements because of the recitation of “or” in the claim limitation. Therefore, the references in combination only needs to teach “a flow of frequency of human moving” in order to teach the human interaction information.
Applicant’s Argument: On page 15-17 of Applicant’s response to rejections under 35 U.S.C. 103, applicant states that Censi in combination of Liping fails to teach “updating an initial representation vector of each of the regions according to the human interaction information, so as to obtain an embedding representation vector of each of the regions, wherein for each region, the initial representation vector of the region is calculated according to an initial land usage property of the region”.
Examiner’s Response: Applicant’s argument is not persuasive. Applicant agrees that Censi discloses the target segment embedding is generated independently from the neighborhood embedding based on different information sources. Each embedding represents a different set of information. Thus, the target segment represents the initial representation vector and the neighborhood embeddings is the information used to update the initial representation vector. Censi (pg. 4, Section A, par. 1-2) teaches a graph attention mechanism that is applied prior to the self-attention mechanism. The graph attention mechanism updates the target segment embedding based on the contribution of the spatial neighborhood to generate an aggregated vector. Bright in combination of Censi and Liping is used to teach the entirety of the claim limitation of human interaction information.
Applicant’s Argument: On page 17-18 of Applicant’s response to rejections under 35 U.S.C. 103, applicant states that Censi in combination of Liping fails to teach the limitations of claim 1 because Censi disclosure of classification is static and Liping does not disclose predictions that rely on feature maps.
Examiner’s Response: Applicant’s argument is not persuasive. In response to applicant's arguments against the references individually, one cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986). Applicant only provides reason how the individual references of Censi and Liping does not teach specific claim limitations of claim 1 and thus, cannot be combined. Applicant does not provide an explanation of how the references fails to be combined together.
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, 2, 5-8, 11, 12, 15-18, and 20-25 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding Claim 1:
Subject Matter Eligibility Analysis Step 1:
Claim 1 recites “A method of determining a regional land usage property, implemented by an electronic device, the method comprising” and is thus a process, one of the four statutory categories of patentable subject matter.
Subject Matter Eligibility Analysis Step 2A Prong 1:
“updating an initial representation vector of each of the regions according to the human interaction information, so as to obtain an embedding representation vector of each of the regions, wherein for each region, the initial representation vector of the region is calculated according to an initial land usage property of the region” (a mathematical calculation, see par. 30, 32 of Specification; The initial representation vector of the region may be combining the weights of sub-regions. The updating of an initial vector consists of aggregating the human interaction information and initial representation vector of each region. The embedding representation vector is calculated by Equation 1 and 2.)
“selecting a target region from the regions, and selecting a plurality of static neighbor regions within a preset range around the target region” (a mental process that can be performed in the human mind, i.e. judgement; Choosing an area to perform the prediction and identifying the neighboring regions can be performed in the human mind.)
“generating a feature map of the target region according to the embedding representation vector of the target region and the embedding representation vectors of the plurality of static neighbor regions” (a mathematical calculation, A feature map is the result of convolution and dot product, which are both mathematical operations. Spec par. 32-33)
“predicting a land usage property of the target region by using the feature map, so as to obtain a predicted land usage property of the target region at a next time, wherein the predicted land usage property is used to establish supporting facilities corresponding to the predicted land usage property in the target region” (a mental process that can be performed in the human mind with the aid of pen and paper, i.e. judgement; A feature map highlights important features and patterns in images and a prediction can be made by analyzing the information.)
“wherein the updating an initial representation vector of each of the regions according to the human interaction information comprises: for each region, adjusting the initial representation vector of the region by using the human interaction information related to the region as a weight to obtain a fusion feature vector of the region” (a mathematical calculation, see par. 31-32 of Specification; The fusion vector is calculated by aggregating the vectors of the human interaction information and the initial representation of each region.)
“performing a weighted summation on the fusion feature vector of the region and the initial representation vector of the region according to a preset coefficient, so as to obtain the embedding representation vector of the region” (a mathematical calculation, see par. 30-32 of Specification; The weighted summation is shown in Equation 2.)
Claim 1 therefore recites an abstract idea.
Subject Matter Eligibility Analysis Step 2A Prong 2:
"acquiring a human interaction information between a plurality of regions at a specified time, wherein the human interaction information comprises at least one selected from a flow frequency of human moving from one of the plurality of regions to another one of the plurality of regions at the specified time or a region retrieval frequency of human in one of the plurality of regions retrieving information for another one of the plurality of regions at the specified time” (This step is directed to data gathering, which is understood to be insignificant extra solution activity - see MPEP 2106.05(g))
The additional elements as disclosed above alone or in combination do not integrate the judicial exception into practical application as they are mere insignificant extra solution activity in combination of generic computer functions being implemented with generic computer elements in a high level of generality to perform the disclosed abstract idea above. Therefore, Claim 1 is directed to the abstract idea.
Subject Matter Eligibility Analysis Step 2B:
"acquiring a human interaction information between a plurality of regions at a specified time, wherein the human interaction information comprises at least one selected from a flow frequency of human moving from one of the plurality of regions to another one of the plurality of regions at the specified time or a region retrieval frequency of human in one of the plurality of regions retrieving information for another one of the plurality of regions at the specified time” (This step is directed to transmitting or receiving information, which is understood to be insignificant extra solution activity and well understood, routine and conventional activity of transmitting and receiving data as identified by the court - see MPEP 2106.05(d))
The additional elements as disclosed above alone or in combination do not recite significantly more than the abstract idea itself as they are mere insignificant extra solution activity in combination of generic computer functions being implemented with generic computer elements in a high level of generality to perform the disclosed abstract idea above. Therefore, Claim 1 is subject-matter ineligible.
Regarding Claim 11:
The claim recites a system (“An electronic device, comprising:”) that performs the method as described in claim 1. Therefore, claim 11 is rejected for the same reasons as disclosed for claim 1. The limitations for additional elements of claim 11 are analyzed below.
Subject Matter Eligibility Analysis Step 2A Prong 1:
Please see Step 2A Prong 1 analysis of claim 1
Subject Matter Eligibility Analysis Step 2A Prong 2 & 2B:
“at least one processor” (mere instructions to apply the exception using a generic computer component - see MPEP 2106.05(f))
“a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions, when executed by the at least one processor, cause the at least one processor to at least” (mere instructions to apply the exception using a generic computer component - see MPEP 2106.05(f))
Regarding Claim 20:
The claim recites an article of manufacture that performs the method as described in claim 1. Therefore, claim 20 is rejected for the same reasons as disclosed for claim 1. The limitations for additional elements of claim 20 are analyzed below.
Subject Matter Eligibility Analysis Step 2A Prong 1:
Please see Step 2A Prong 1 analysis of claim 1
Subject Matter Eligibility Analysis Step 2A Prong 2 & 2B:
“A non-transitory computer-readable storage medium having computer instructions stored therein, the instructions, when executed by a computer system, configured to cause the computer system to at least” (mere instructions to apply the exception using a generic computer component - see MPEP 2106.05(f))
Regarding Claims 2, 12, and 21:
Subject Matter Eligibility Analysis Step 2A Prong 1:
“counting, for any region, an initial land usage property of each sub-region in the region” (a mental process, i.e. observations; Identifying the number of different property types in each sub-region.)
“determining a weight for each sub-region in the region according to the initial land usage property of each sub-region in the region and a preset weight for the land usage property” (a mental process, i.e. judgment; see par. 48 in Specification; Assigning preset weights to the identified property types in the region.)
“generating the initial representation vector of the region according to the weight for each sub-region in the region” (a mathematical calculation; see par. 49 in Specification; Converting the determined weights of the land usage properties to a vector format.)
Subject Matter Eligibility Analysis Step 2A Prong 2 & 2B: None
Regarding Claims 5, 15, and 22:
Subject Matter Eligibility Analysis Step 2A Prong 1:
“selecting a region to be predicted for a land usage property from the regions, so as to obtain the target region” (a mental process, i.e. judgement)
“selecting a plurality of random regions within the preset range around the target region, so as to obtain the plurality of static neighbor regions” (a mental process, i.e. judgement)
Subject Matter Eligibility Analysis Step 2A Prong 2 & 2B: None
Regarding Claims 6, 16, and 23:
Subject Matter Eligibility Analysis Step 2A Prong 1:
“stitching the initial representation vectors of the plurality of static neighbor regions, so as to obtain a first static adjacency matrix” (a mathematical calculation; see par. 61 of Specification; Combining vectors together by rows or columns to form a matrix)
“for any region in the plurality of static neighbor regions, stitching the initial representation vectors of other regions in the plurality of static neighbor regions except the region, so as to obtain a second static adjacency matrix” (a mathematical calculation; see par. 64 of Specification; Combining vectors together by rows or columns to form a matrix)
“calculating and comparing a contribution of the first static adjacency matrix and a contribution of the second static adjacency matrix by using a preset efficiency function” (a mathematical calculation, see par. 62-64 of Specification; The efficiency function is used to calculate the influence of the regions.)
“determining a land usage property of the region as an explanation for the predicted land usage property of the target region, in response to the contribution of the first static adjacency matrix being not equal to the contribution of the second static adjacency matrix” (a mental process, i.e. judgement)
Subject Matter Eligibility Analysis Step 2A Prong 2 & 2B:
“acquiring initial representation vectors of the plurality of static neighbor regions” (This step is directed to data gathering, which is understood to be insignificant extra solution activity (2106.05(g) in step 2A prong 2) and well understood, routine and conventional activity of transmitting and receiving data as identified by the court (2106.05(d) in step 2B))
Regarding Claims 7, 17, and 24:
Subject Matter Eligibility Analysis Step 2A Prong 1:
“determining regions with a human interaction with the target region at the specified time, so as to obtain dynamic neighbor regions” (a mental process, i.e. judgement)
“stitching the initial representation vectors of the dynamic neighbor regions, so as to obtain a first dynamic adjacency matrix” (a mathematical calculation; see par. 61 and 68 of Specification; Combining vectors together by rows or columns to form a matrix)
“for any region in the dynamic neighbor regions, stitching the initial representation vectors of other regions in the dynamic neighbor regions except the region, so as to obtain a second dynamic adjacency matrix” (a mathematical calculation; see par. 64 and 69 of Specification; Combining vectors together by rows or columns to form a matrix)
“calculating and comparing a contribution of the first dynamic adjacency matrix and a contribution of the second dynamic adjacency matrix by using a preset efficiency function” (a mathematical calculation, see par. 62-64, 72 of Specification; The efficiency function is used to calculate the influence of the regions.)
“determining a land usage property of the region as an explanation for the predicted land usage property of the target region, in response to the contribution of the first dynamic adjacency matrix being not equal to the contribution of the second dynamic adjacency matrix” (a mental process, i.e. judgement)
Subject Matter Eligibility Analysis Step 2A Prong 2 & 2B: None
Regarding Claims 8, 18, and 25:
Subject Matter Eligibility Analysis Step 2A Prong 1: None
Subject Matter Eligibility Analysis Step 2A Prong 2 & 2B:
“wherein the predicting a land usage property of the target region by using the feature map, so as to obtain a predicted land usage property of the target region at a next time comprise analyzing the feature sub-map by using a pre-trained graph convolution network, so as to obtain the predicted land usage property of the target block at the next time, wherein the pre-trained graph convolution network is a network model trained using a historical land usage property” (mere instructions to apply the exception using a generic computer component - see MPEP 2106.05(f))
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, 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, 2, 5-8, 11, 12, 15-18, and 20-25 are rejected under 35 U.S.C. 103 as being unpatentable over Censi “Attentive Spatial Temporal Graph CNN for Land Cover Mapping From Multi Temporal Remote Sensing Data” in view of Liping “Monitoring And Predicting Land Use And Land Cover Changes Using Remote Sensing And GIS Techniques – A Case Study Of A Hilly Area, Jiangle, China” and Bright (US20150213160A1).
Regarding claim 1, Censi teaches:
“A method of determining a regional land usage property, implemented by an electronic device, the method comprising” (abstract; pg. 6, Section D, par. 1, Experiments are conducted to determine land cover classification. It is implied that the architecture of the CNN is executed on an electronic device such as a computer.)
“acquiring a ” ([pg. 2, col. 1, par. 2; pg. 3, col. 2, par. 3-4], A region adjacency graph can be extracted from satellite images to show objects spatial interactions. A RAG represents a plurality of regions of a geographical area at a particular time instance.)
“updating an initial representation vector of each of the regions according to the ” ([pg. 4, col. 1, par. 1-2; Figure 3], Satellite image time series (SITS) data is processed by CNN models to extract information from the target segment and the spatial neighborhood set surrounding the target segment. The target segment embedding and the neighborhood embedding are generated. The two embeddings are combined together to form the combined embeddings to create an updated set of information for the region to be used in land cover classification. The initial representation vector is generated based on the SITS data.)
“selecting a target region from the regions, and selecting a plurality of static neighbor regions within a preset range around the target region” ([pg. 3, col. 2, par. 3-4; pg. 4, col. 1, par. 1; pg. 7, col. 2, par. 2; Figure 3 & 6], A RAG may represent the target region as a red node and the neighborhood regions as blue nodes. For each node in the RAG, there can be a different number of neighbors that are surrounding the particular region. The neighbors are determined by identifying whether a region is within the spatial neighborhood of the target segment.)
“generating a feature map of the target region according to the embedding representation vector of the target region and the embedding representation vectors of the plurality of static neighbor regions” ([pg. 5, Section C, par. 1; pg. 6, col. 1, par. 3; Figure 3], The 1-D CNN may generate a feature map for the target region and the adjacent neighboring regions. In addition, a softmax function is applied during the classification step. A softmax function is an activation function that performs non-linear mapping of features of a vector.)
“predicting a land usage property of the target region by using the feature map, so as to obtain a predicted land usage property of the target region ” ([pg. 5, col. 2, par. 1; Figure 3], The model uses the combined embedding to perform land cover classification of the target segment.)
“wherein the updating an initial representation vector of each of the regions according to the ” ([pg. 4, Section A, par. 1-2; Figure 3], The target segment embedding (initial representation vector) employs graph attention mechanism to determine how much influence a neighborhood has on the target segment. The target segment embedding is adjusted using the attention coefficient that weights the importance of the neighbors on a target segment. The result is an aggregation of the target segment embedding and neighborhood embedding.)
“performing a weighted summation on the fusion feature vector of the region and the initial representation vector of the region according to a preset coefficient, so as to obtain the embedding representation vector of the region” ([pg. 4-5, Section B, par. 1; Figure 3], A self-attention mechanism is employed to combine the target segment embedding and the spatial neighborhood embedding. The self-attention mechanism is a weighted summation and the contribution of the features from the target segment and its spatial neighborhood are automatically weighted. The result of the attention-based aggregation is the final embedding.)
Censi does not explicitly disclose an implementation of “acquiring a human interaction information ... wherein the human interaction information comprises at least one selected from a flow frequency of human moving from one of the plurality of regions to another one of the plurality of regions at the specified time or a region retrieval frequency of human in one of the plurality of regions retrieving information for another one of the plurality of regions at the specified time”, “updating an initial representation vector of each of the regions according to the human interaction information”, and “obtain a predicted land usage property of the target region at a next time, wherein the predicted land usage property is used to establish supporting facilities corresponding to the predicted land usage property in the target region”. However, Liping discloses in the same field of endeavor:
“acquiring a human interaction information between a plurality of regions at a specified time ...” ([pg. 3, par. 5; pg. 4, par. 1], Land use data is collected for the experiments during different periods of time. Land use data shows the interactions between humans and the physical environment.)
“updating an initial representation vector of each of the regions according to the human interaction information ...” ([pg. 3, par. 5; pg. 4-6, par. 1-2], Land use data shows the interactions between humans and the physical environment. Markov chain analysis is used to simulate complex processes of the land use change. Markov chain analysis used land use type number, land use status, and time point to determine the transition from a current state to the next state.)
“predicting a land usage property of the target region wherein the predicted land usage property is used to establish supporting facilities corresponding to the predicted land usage property in the target region” ([pg. 8, par. 1-2, pg. 14-15, par 2; Figure 8 & 9], The model makes land use predictions for 2025 and 2036 based on the data in 2003 and 2014. The results show the transfer of woodland and farmland into construction land. Figure 8 shows the percentage of area where buildings can be established based on the prediction of the land cover changes.)
It would be obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention to incorporate the teaching of “acquiring a human interaction information”, “updating an initial representation vector of each of the regions according to the human interaction information”, and “obtain a predicted land usage property of the target region at a next time, wherein the predicted land usage property is used to establish supporting facilities corresponding to the predicted land usage property in the target region” from Liping into the teaching of Censi. Doing so can improve the predictions for future land use and land change trends by applying advanced analysis techniques on the dynamic changes in land use patterns. (Liping, abstract).
Censi in view of Liping does not explicitly disclose an implementation of “wherein the human interaction information comprises at least one selected from a flow frequency of human moving from one of the plurality of regions to another one of the plurality of regions at the specified time or a region retrieval frequency of human in one of the plurality of regions retrieving information for another one of the plurality of regions at the specified time”. However, Bright discloses in the same field of endeavor:
“... wherein the human interaction information comprises at least one selected from a flow frequency of human moving from one of the plurality of regions to another one of the plurality of regions at the specified time or a region retrieval frequency of human in one of the plurality of regions retrieving information for another one of the plurality of regions at the specified time” ([0016, 0018-0020, 0023, 0029-0030, 0034-0035, 0044], Migration data (human interaction information) can be obtained from IRS data warehouse. The year-to-year address change from filed tax returns provides information for the number of inflow and outflow of the population in each county of the US. The migration rate by county is derived by equation 5. The prediction of population forecasts of the future may be used for regional planning to identify areas for new physical construction. The system may include several variables to identify areas of potential growth such as excluding certain lands, slope of land covers, roads, and city limits. The systems and processes described in Bright can be combined with systems that model and project land cover changes. Thus, it would be obvious to a person having ordinary skills in the arts to implement a system that can model changes to urban and non-urban areas at the county level, same day migration, and population loss patterns.)
It would be obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention to incorporate the teaching of “wherein the human interaction information comprises at least one selected from a flow frequency of human moving from one of the plurality of regions to another one of the plurality of regions at the specified time or a region retrieval frequency of human in one of the plurality of regions retrieving information for another one of the plurality of regions at the specified time” from Bright into the teaching of Censi in view of Liping. Doing so can improve the planning of urban areas in future development based on population changes at a county level. (Bright, abstract, par. 44).
Regarding claim 11:
Claim 11 recites a system that performs the same process as described in Claim 1. Therefore claim 11 is rejected under the same reasons mention for claim 1. The additional elements of claim 11 is addressed below by Censi:
“An electronic device, comprising: at least one processor” ([pg. 8, section VI], The experiments of executing the different methods would have been conducted on a computer with a processor and memory component.)
“a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions, when executed by the at least one processor, cause the at least one processor to at least” ([pg. 8, section VI], The experiments of executing the different methods would have been conducted on a computer with a processor and memory component.)
Regarding claim 20:
Claim 20 recites an article of manufacture that performs the same process as described in Claim 1. Therefore claim 20 is rejected under the same reasons mention for claim 1. The additional elements of claim 20 is addressed below by Censi:
“A non-transitory computer-readable storage medium having computer instructions stored therein, the instructions, when executed by a computer system, configured to cause the computer system to at least” ([pg. 8, section VI], The experiments of executing the different methods would have been conducted on a computer with a processor and memory component.)
Regarding claims 2, 12, and 21, Censi teaches:
“counting, for any region, an initial land usage property of each sub-region in the region” ([pg. 3, col. 2, par. 2-4; pg. 6, col. 1-2, par. 4-5; Figure 4; Table 3], The experiment uses satellite images and some images may have ground truth data associated with the region to indicate the number of objects in the region.)
“determining a weight for each sub-region in the region according to the initial land usage property of each sub-region in the region and a preset weight for the land usage property” ([pg. 6, col. 1-2, par. 4], For the satellite images, a normalized difference vegetation index and a normalized difference water index are calculated to identify the different land properties. These indexes are calculated from the satellite images at different times and contributes to the land cover classification in helping to predict the changes over time. The satellite images were preprocessed to consider a range of bands and the calculated indices are applied to the satellite data.)
“generating the initial representation vector of the region according to the weight for each sub-region in the region” ([pg. 4, col. 2, par. 1-4], A graphical representation is created from satellite image time series data and the information is used to compute the target segment embedding and neighborhood embeddings.)
Regarding claims 5, 15, and 22, Censi teaches:
“selecting a region to be predicted for a land usage property from the regions, so as to obtain the target region” ([pg. 3, col. 2, par. 3-4; pg. 4, col. 1, par. 1; Figure 2 & 3], Satellite images are segmented to generate the region adjacency graph. A target segment is selected and represented as the red node in the RAG.)
“selecting a plurality of random regions within the preset range around the target region, so as to obtain the plurality of static neighbor regions” ([pg. 4, col. 1, par. 1; pg. 7, col. 2, par. 2; Figure 6], A spatial neighborhood set around the target segment is selected and represented by blue nodes. For each node in the RAG, there can be a different number of neighbors that are surrounding the particular region. The neighbors are determined by identifying whether a region is within the spatial neighborhood of the target segment.)
Regarding claims 6, 16, and 23, Censi teaches:
“acquiring initial representation vectors of the plurality of static neighbor regions” ([pg. 4, col. 2, par. 1-4, Figure 3], A graphical representation is created from satellite image time series data and the information is used to compute the neighborhood embeddings. A 1-D CNN model processes the neighborhood segment time series data to create the initial embeddings that represents the neighboring regions of the target segment.)
“stitching the initial representation vectors of the plurality of static neighbor regions, so as to obtain a first static adjacency matrix” ([pg. 4, col. 2, par. 1-4, Figure 3], The individual neighborhood embeddings are aggregate together using a graph attention mechanism. The region adjacency graph is input into the architecture and the graph structure defines the initial relationship between nodes and its neighboring nodes.)
“for any region in the plurality of static neighbor regions, stitching the initial representation vectors of other regions in the plurality of static neighbor regions except the region, so as to obtain a second static adjacency matrix” ([pg. 4, col. 2, par. 1-4], Each node from the graph structure aggregates the features from its neighbors to update its representation. The neighbor segment embeddings are multiplied by the attention coefficient to determine a weighted adjacency matrix. Each neighboring node is assigned an attention coefficient indicating the importance of that neighbor’s features for the feature update of a node.)
“calculating and comparing a contribution of the first static adjacency matrix and a contribution of the second static adjacency matrix by using a preset efficiency function” ([pg. 4, col. 2, par. 1-4], Graph attention mechanism use attention functions to assign different importance weights to neighboring nodes, which is a key component for their operation and can be seen as an efficiency function in terms of selectively focusing on relevant information.)
“determining a land usage property of the region as an explanation for the predicted land usage property of the target region, in response to the contribution of the first static adjacency matrix being not equal to the contribution of the second static adjacency matrix” ([pg. 4, col. 2, par. 1-4; pg. 11, col. 2, par. 1; Figure 9], The graph attention mechanism needs to discriminate against the features of a particular node with the features of its neighboring nodes. The features between a node and its neighboring nodes may be different. When the neighboring nodes’ features are more heavily weighted, it can influence the classification of node. For example, when neighboring nodes shows features that represent vineyards, the classification of the target node may be corrected to vineyard if the initial representation did not contain features that represent vineyards.)
Regarding claims 7, 17, and 24, Censi teaches:
“determining regions with a ” ([pg. 4, col. 2, par. 1-4, Figure 3], A graphical representation is created from satellite image time series data and the information is used to compute the neighborhood embeddings. A 1-D CNN model processes the neighborhood segment time series data to create the initial embeddings that represents the neighboring regions of the target segment.)
“stitching the initial representation vectors of the ” ([pg. 4, col. 2, par. 1-4, Figure 3], The individual neighborhood embeddings are aggregate together using a graph attention mechanism. The region adjacency graph is input into the architecture and the graph structure defines the initial relationship between nodes and its neighboring nodes.)
“for any region in the ” ([pg. 4, col. 2, par. 1-4], Each node from the graph structure aggregates the features from its neighbors to update its representation. The neighbor segment embeddings are multiplied by the attention coefficient to determine a weighted adjacency matrix. Each neighboring node is assigned an attention coefficient indicating the importance of that neighbor’s features for the feature update of a node.)
“calculating and comparing a contribution of the first ” ([pg. 4, col. 2, par. 1-4], Graph attention mechanism use attention functions to assign different importance weights to neighboring nodes, which is a key component for their operation and can be seen as an efficiency function in terms of selectively focusing on relevant information.)
“determining a land usage property of the region as an explanation for the predicted land usage property of the target region, in response to the contribution of the first ” ([pg. 4, col. 2, par. 1-4; pg. 11, col. 2, par. 1; Figure 9], The graph attention mechanism needs to discriminate against the features of a particular node with the features of its neighboring nodes. The features between a node and its neighboring nodes may be different. When the neighboring nodes’ features are more heavily weighted, it can influence the classification of node. For example, when neighboring nodes shows features that represent vineyards, the classification of the target node may be corrected to vineyard if the initial representation did not contain features that represent vineyards.)
Censi does not explicitly disclose an implementation of “a human interaction”, “dynamic neighbor regions”, and “dynamic adjacency matrix”. Censi teaches the spatial neighborhood of a generic target node is fixed over time. However, Liping discloses in the same field of endeavor:
“determining regions with a human interaction with the target region at the specified time, so as to obtain dynamic neighbor regions” ([pg. 3, par. 5; pg. 4, par. 1; pg. 6, par. 1], Land use data is collected for the experiments during different periods of time. Land use data shows the interactions between humans and the physical environment. Cellular automata is a dynamic model that can adaptively change the neighborhood’s size based on changing influence and interactions.)
“stitching the initial representation vectors of the dynamic neighbor regions, so as to obtain a first dynamic adjacency matrix” ([pg. 6, par. 1-2], Cellular automata consist of cell, and its neighbors in a grid that can be represented as an adjacency matrix to show the connections between cells.)
It would be obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention to incorporate the teaching of “a human interaction”, “dynamic neighbor regions”, and “dynamic adjacency matrix” from Liping into the teaching of Censi. Doing so can improve the predictions for future land use and land change trends by applying advanced analysis techniques on the dynamic changes in land use patterns. (Liping, abstract).
Regarding claims 8, 18, and 25, Censi teaches:
“wherein the predicting a land usage property of the target region by using the feature map, so as to obtain a predicted land usage property of the target region at a next time comprise analyzing the feature sub-map by using a pre-trained graph convolution network, so as to obtain the predicted land usage property of the target block at the next time, wherein the pre-trained graph convolution network is a network model trained using a historical land usage property” ([pg. 5, col. 2, par. 1; pg. 6, col. 1, par. 1-4], The model uses the combined embedding to perform land cover classification of the target segment. A classifier processes the combined embeddings that contains the feature maps from the 1-dimensional CNN to classify the target segment. The STEGON architecture is trained using satellite image time series from the Reunion Island and Dordogne dataset, which contains historical land usage data of the regions.)
Conclusion
THIS ACTION IS MADE FINAL. 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 GARY MAC whose telephone number is (703)756-1517. The examiner can normally be reached Monday - Friday 8:00 AM - 5:00 PM.
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, Abdullah Kawsar can be reached at (571) 270-3169. 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.
/GARY MAC/Examiner, Art Unit 2127
/ABDULLAH AL KAWSAR/Supervisory Patent Examiner, Art Unit 2127