Prosecution Insights
Last updated: November 20, 2026
Application No. 18/727,310

A METHOD AND APPARATUS FOR TABLE RECOGNITION

Final Rejection §102§103
Filed
Jul 08, 2024
Priority
Jan 07, 2022 — CN 202210018252.1 +1 more
Examiner
WILLIAMS, REBECCA COLETTE
Art Unit
2677
Tech Center
2600 — Communications
Assignee
Beijing Youzhuju Network Technology Co., Ltd.
OA Round
2 (Final)
50%
Grant Probability
Moderate
3-4
OA Rounds
10m
Est. Remaining
99%
With Interview

Examiner Intelligence

Grants 50% of resolved cases
50%
Career Allowance Rate
7 granted / 14 resolved
-12.0% vs TC avg
Strong +58% interview lift
Without
With
+58.3%
Interview Lift
resolved cases with interview
Typical timeline
3y 3m
Avg Prosecution
21 currently pending
Career history
50
Total Applications
across all art units

Statute-Specific Performance

§101
10.4%
-29.6% vs TC avg
§103
58.3%
+18.3% vs TC avg
§102
16.0%
-24.0% vs TC avg
§112
14.6%
-25.4% vs TC avg
Black line = Tech Center average estimate • Based on career data from 14 resolved cases

Office Action

§102 §103
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 All Information Disclosure Statements filed on or before 01/23/2025 have been considered by examiner. Specification The abstract of the disclosure is objected to because of the inclusion of “(FIG. 1)” at the bottom of the page and “This application discloses” at the top. According to the MPEP the abstract “…may not include other parts of the application or other material” and “the language should be clear and concise and should not repeat information given in the title. It should avoid using phrases which can be implied, such as, "This disclosure concerns," "The disclosure defined by this invention," "This disclosure describes," etc.” (see MPEP 608.01(b) IC Language and Format) A corrected abstract of the disclosure is required and must be presented on a separate sheet, apart from any other text. See MPEP § 608.01(b). Response to Amendment The applicant has amended claims 1, 8, 9, 12, 13, 15-19, 21, and 22. Applicant has cancelled claims 7 and 20. Claims 1-6, 8-10, 12, 13, 15-19, 21, and 22 remain pending. Applicant’s amendments have overcome the previously made 112 (b) rejections. Claim Rejections - 35 USC § 102 In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status. The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action: A person shall be entitled to a patent unless – (a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention. (a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention. Claims 1-3, 8, 9, 12, 13, 15, 16, 21, and 20 are rejected under 35 U.S.C. 102(a)(1) and (a)(2) as being anticipated by Foncubierta (US 20200401798 A1). With respect to claim 1, Foncubierta teaches a method of table recognition (“According to one aspect of the present invention there is provided a computer-implemented method for generating a data structure representing tabular information in a scanned image.” Paragraph 0003), comprising: acquiring an image to be processed that comprises a table (“The method includes storing image data representing a scanned image of a table, processing the image data to identify positions of characters and lines in the image, and mapping locations in the image of information cells, each containing a set of the characters, in dependence on said positions.” Paragraph 0003), and determining information about individual table cells in the image to be processed, the information about the individual table cells comprising positions of bounding boxes of the individual table cells (“The method includes, for each cell, determining cell attribute values, dependent on the cell locations, for a predefined set of cell attributes, and supplying the attribute values as inputs to a machine-learning model trained to pre-classify cells as header cells or data cells in dependence on cell attribute values. The method then applies predefined rules to the pre-classified cells to identify a corner data cell marking a transition between a region containing data cells and edge regions containing header cells in the image, and defines row and column header cells in dependence on location of cells in the edge regions.” Paragraph 0003); obtaining, based on the information about the individual table cells, parent table cells of the individual table cells in a row direction and parent table cells of the individual table cells in a column direction (“The method includes, for each cell, determining cell attribute values, dependent on the cell locations, for a predefined set of cell attributes, and supplying the attribute values as inputs to a machine-learning model trained to pre-classify cells as header cells or data cells in dependence on cell attribute values. The method then applies predefined rules to the pre-classified cells to identify a corner data cell marking a transition between a region containing data cells and edge regions containing header cells in the image, and defines row and column header cells in dependence on location of cells in the edge regions.” Paragraph 0003); and obtaining structural coordinates of the individual table cells based on parent-child relationships of the individual table cells in the row direction and parent-child relationships of the individual table cells in the column direction, wherein a structural coordinate comprises a starting row, a starting column, a terminating row, and a terminating column (“The method further comprises, for each of the set of row header cells and the set of column header cells, defining a hierarchy tree, indicating hierarchical structure of the header cells, in dependence on relative locations of those cells in the image, and using the hierarchy trees to generate a data structure mapping each data cell to corresponding row and column header cells in the hierarchical structure. The resulting data structure is stored to provide a queryable representation of the table.” Paragraph 0003 and figures 10a and 10b and “In step 47, the resulting sets of row header cells and column header cells are processed by hierarchy extraction module 31 to define a hierarchy tree for each set. These hierarchy trees indicate hierarchical structure of the (row or column) header cells. Each hierarchy tree is defined in dependence (at least) on relative locations of the row or column cells in the image as explained in detail below.” Paragraph 0046), wherein the row-direction parent-child relationships form a row forest and the column- direction parent-child relationships form a column forest, each forest comprising at least one tree of nodes corresponding to the individual table cells, the starting and terminating rows being determined from the row forest and the starting and terminating columns being determined from the column forest (“The method further comprises, for each of the set of row header cells and the set of column header cells, defining a hierarchy tree, indicating hierarchical structure of the header cells, in dependence on relative locations of those cells in the image, and using the hierarchy trees to generate a data structure mapping each data cell to corresponding row and column header cells in the hierarchical structure. The resulting data structure is stored to provide a queryable representation of the table.” Paragraph 0003 and figures 10a and 10b and “In step 47, the resulting sets of row header cells and column header cells are processed by hierarchy extraction module 31 to define a hierarchy tree for each set. These hierarchy trees indicate hierarchical structure of the (row or column) header cells. Each hierarchy tree is defined in dependence (at least) on relative locations of the row or column cells in the image as explained in detail below.” Paragraph 0046). With respect to claim 2, Foncubierta teaches the method of claim 1, wherein the information about the individual table cells further comprises: word embedding vectors of a text in the individual table cells and/or visual features of the individual table cells (“In step 43, cell pre-classification module 29 determines cell attribute values for cells in the cell map. The attribute values comprise values for a predefined set of cell attributes previously used to train a ML model 23 stored in memory 20. This ML model is trained to pre-classify cells as header cells or data cells in dependence on the cell attribute values. Cell attribute values are dependent (at least) on locations of cells in the cell map, and may include other attributes, e.g. relating to format of character strings, as detailed below. In step 44, module 29 supplies the cell attribute values as inputs to the pre-trained model to obtain soft labels which pre-classify individual cells as either header cells or data cells. These preliminary classification labels are stored as metadata in memory 20.” Paragraph 0044). With respect to claim 3, Foncubierta teaches the method of claim 1 or 2, wherein obtaining, based on the information about the individual table cells, parent table cells of the individual table cells in a row direction and parent table cells of the individual table cells in a column direction comprises: inputting the information about the individual table cells into a machine learning model to obtain parent table cells of the individual table cells in the row direction and parent table cells of the individual table cells in the column direction (“In step 44, module 29 supplies the cell attribute values as inputs to the pre-trained model to obtain soft labels which pre-classify individual cells as either header cells or data cells. These preliminary classification labels are stored as metadata in memory 20.” Paragraph 44 and “In step 45, cell analysis module 30 applies a set of predefined rules to the pre- classified cells to identify a corner data cell for the image. Preferred embodiments apply these rules in a probabilistic manner, as detailed below, to identify the corner data cell. The corner data cell marks a transition between a region containing data cells and edge regions containing header cells in the image. The location of the corner data cell is then used in step 46. Here, module 30 defines (labels) the row and column header cells for the image in dependence (at least) on location of cells in the edge regions as detailed below.” Paragraph 45 and “In step 47, the resulting sets of row header cells and column header cells are processed by hierarchy extraction module 31 to define a hierarchy tree for each set.” Paragraph 0046). With respect to claim 8, Foncubierta teaches the method of claim 1, wherein determining the starting and terminating rows from the row forest comprises: determining a starting row coordinate b_i corresponding to a table cell I (“In step 47, the resulting sets of row header cells and column header cells are processed by hierarchy extraction module 31 to define a hierarchy tree for each set. These hierarchy trees indicate hierarchical structure of the (row or column) header cells. Each hierarchy tree is defined in dependence (at least) on relative locations of the row or column cells in the image as explained in detail below.” Paragraph 0046 and “The root cell is a notional cell at a defined reference location in the image. This is illustrated in FIG. 10a for column headers in an exemplary table. Here the root cell location x is defined at a mid-position above the column headers in the image.” Paragraph 58 and figure 10(b) x); determining the number r_i of rows spanned by the table cell I (“In step 47, the resulting sets of row header cells and column header cells are processed by hierarchy extraction module 31 to define a hierarchy tree for each set. These hierarchy trees indicate hierarchical structure of the (row or column) header cells. Each hierarchy tree is defined in dependence (at least) on relative locations of the row or column cells in the image as explained in detail below.” Paragraph 0046 and figure 10(b) hierarchy level 1); and determining, based on the b_i and the r_i, a terminating row coordinate corresponding to the table cell i as b_i+r_i-1 (“In step 47, the resulting sets of row header cells and column header cells are processed by hierarchy extraction module 31 to define a hierarchy tree for each set. These hierarchy trees indicate hierarchical structure of the (row or column) header cells. Each hierarchy tree is defined in dependence (at least) on relative locations of the row or column cells in the image as explained in detail below.” Paragraph 0046 and figure 10(b) hierarchy level 1 last node), wherein if a node corresponding to the table cell i in a tree where the table cell i is located lacks a sub-branch, the r_i is 1; otherwise, the number of rows corresponding to the table cell i is a sum of numbers of rows corresponding to individual sub-branches of the node (“In step 47, the resulting sets of row header cells and column header cells are processed by hierarchy extraction module 31 to define a hierarchy tree for each set. These hierarchy trees indicate hierarchical structure of the (row or column) header cells. Each hierarchy tree is defined in dependence (at least) on relative locations of the row or column cells in the image as explained in detail below.” Paragraph 0046 and figure 10(b)); and if the table cell i corresponds to a header node, the starting row b_i is equal to a terminating row of a previous tree of the tree where the table cell i is located plus 1 (“In step 47, the resulting sets of row header cells and column header cells are processed by hierarchy extraction module 31 to define a hierarchy tree for each set. These hierarchy trees indicate hierarchical structure of the (row or column) header cells. Each hierarchy tree is defined in dependence (at least) on relative locations of the row or column cells in the image as explained in detail below.” Paragraph 0046 and figure 10(b)); if the table cell i corresponds to the kth sub-node of a node q, the starting row b_i is equal to b_q+sum1, wherein b_q is a starting row of the node q, and sum1 is a sum of numbers of rows spanned by (k-1) sub-branches of the node q (“In step 47, the resulting sets of row header cells and column header cells are processed by hierarchy extraction module 31 to define a hierarchy tree for each set. These hierarchy trees indicate hierarchical structure of the (row or column) header cells. Each hierarchy tree is defined in dependence (at least) on relative locations of the row or column cells in the image as explained in detail below.” Paragraph 0046 and figure 10(b)). With respect to claim 9, Foncubierta teaches the method of claim 1, wherein determining the starting and terminating columns from the column forest comprises: determining the number S of columns spanned by the table cell I (see figure 10 (a) and 10(b)); determining a starting column coordinate a corresponding to the table cell I (see figure 10 (a) and 10(b)); and determining, based on the s_i and the a_i, a terminating column coordinate corresponding to the table cell i as a_i+s_i-1 (see figure 10 (a) and 10(b)), wherein if a node corresponding to the table cell i in a tree where the table cell i is located lacks a sub-branch, the S i is 1; otherwise, the number of columns corresponding to the table cell i is a sum of numbers of columns corresponding to induvial individual sub-branches of the node (see figure 10 (a) and 10(b)); and if the table cell i corresponds to a header node, the starting column a is equal to a terminating column of a header node of a previous tree of the tree where the table cell i is located plus 1 (see figure 10 (a) and 10(b)); if the table cell i corresponds to the jth sub-node of a node p, the starting column a is equal to a_p+sum2, wherein a_p is a starting column of the node p, and sum2 is a sum of numbers of columns spanned by (j-1) sub-branches of the node p (see figure 10 (a) and 10(b)). With respect to claim 12, Foncubierta teaches all claim limitations in consideration of claim 1, due to the substantial similarities between claims 1 and 12 with claim 12 being directed towards a device that performs the method of claim 1. Foncubierta additionally teaches a device that comprises a processor and a memory (“A further aspect of the invention provides a computer program product comprising a computer readable storage medium embodying program instructions, executable by a computing apparatus, to cause the computing apparatus to perform a method for generating a data structure as described above.” Paragraph 0004) and that the processor is configured to execute instructions stored in the memory to cause the device to perform acts (“A further aspect of the invention provides a computer program product comprising a computer readable storage medium embodying program instructions, executable by a computing apparatus, to cause the computing apparatus to perform a method for generating a data structure as described above.” Paragraph 0004). With respect to claim 13, Foncubierta teaches all claim limitations in consideration of claim 1, due to the substantial similarities between claim 1 and claim 13, with claim 13 being directed towards a storage medium that contains the instructions for the claim 1 method. Foncubierta additionally teaches a nontransitory computer-readable storage medium comprising instructions (“A further aspect of the invention provides a computer program product comprising a computer readable storage medium embodying program instructions, executable by a computing apparatus, to cause the computing apparatus to perform a method for generating a data structure as described above.” Paragraph 0004) , the instructions indicating a device to perform acts (“A further aspect of the invention provides a computer program product comprising a computer readable storage medium embodying program instructions, executable by a computing apparatus, to cause the computing apparatus to perform a method for generating a data structure as described above.” Paragraph 0004). With respect to claim 15, Foncubierta teaches the device of claim 12, and all additional limitations in consideration of claim 2, due to the substantial similarities between claim 15 and claim 2, with claim 15 being directed towards the device that performs the method of claim 2. With respect to claim 16, Foncubierta teaches the device of claim 12, and all additional limitations in consideration of claim 3, due to the substantial similarities between claim 16 and claim 3, with claim 16 being directed towards the device that performs the method of claim 3. With respect to claim 21, Foncubierta teaches the device of claim 12, and all additional limitations in consideration of claim 8, due to the substantial similarities between claim 21 and claim 8, with claim 21 being directed towards the device that performs the method of claim 8. With respect to claim 22, Foncubierta teaches the device of claim 12, and all additional limitations in consideration of claim 9, due to the substantial similarities between claim 22 and claim 9, with claim 22 being directed towards the device that performs the method of claim 9. 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 4 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Foncubierta. With respect to claim 4, with dependency from claim 1 or claim 2, Foncubierta teaches the method of claim 3. Foncubierta additionally teaches the rest of the claim 4 limitations in consideration of a combination of its own embodiments. Foncubierta discloses a flexible model based system for its hierarchal detection strategy (“It will be appreciated that various changes and modifications can be made to the exemplary embodiments described above. By way of example, other machine learning models, such as neural networks or support vector machine models, may be employed for pre-classification of cells. While MLN models are preferred for CDC detection, other rule-based systems, e.g. fixed rule-sets or Bayesian networks/other probabilistic graph models, could be employed.” Paragraph 0065). It details two induvial models, an RF model for feature (Attribute) extraction, and an MLN for rule based cell position detection. As supported by the flexibility illuminated in Foncubierta’s specification, it would be obvious for a person of ordinary skill to combine the functionality of these two models into a singular model or model pipeline, with the expectation being that doing so would yield predictable results. Such combination would comprise: a feature extracting module (“The attribute set comprised a plurality of attributes dependent on location of cells in the image. These attributes were selected from: number of neighboring cells; number of neighboring cells to the right; number of neighboring cells to the left; number of neighboring cells above; number of neighboring cells below; having a line on the right; having a line on the left; having a line above; having a line below; and the area occupied by the cell. In preferred embodiments, the attribute set also included a subset of attributes dependent on character strings in cells. These attributes were selected from: characters are bold; characters are italic; proportion (e.g. %) of numeric characters; proportion of alphabetic characters; and proportion of non-alphanumeric characters (special characters, symbols, etc.).” paragraph 0052 and “Pre-classification module 29 determines attribute values, corresponding to each of the above attributes, for all cells in the cell map.” Paragraph 0053), and a first decision module and a second decision module (“The soft labels are supplied to cell analysis module 30 which implements a Markov logic network 61, described further below, to identify the corner data cell. All cells are then classified as either a row header cell, a column header cell, or a data cell.” Paragraph 0053 and “It will be appreciated that various changes and modifications can be made to the exemplary embodiments described above. By way of example, other machine learning models, such as neural networks or support vector machine models, may be employed for pre-classification of cells. While MLN models are preferred for CDC detection, other rule-based systems, e.g. fixed rule-sets or Bayesian networks/other probabilistic graph models, could be employed.” Paragraph 0065 and “During inference, the rules are applied to the pre-classified cells by applying the MLN, with probabilities P assigned to predicates for each cell, to determine the most probable CDC for the table. In particular, the MLN will identify the cell associated with the lowest “rule-breaking cost” for the network. Some rules will generally be broken for all cells, and a higher cost is associated with higher rule weights, whereby the cell associated with the minimum total cost is the most likely CDC.” Paragraph 0056, dual column and row rulesets ); wherein the feature extracting module is configured to process the information about the individual table cells to obtain a feature sequence (“The attribute set comprised a plurality of attributes dependent on location of cells in the image. These attributes were selected from: number of neighboring cells; number of neighboring cells to the right; number of neighboring cells to the left; number of neighboring cells above; number of neighboring cells below; having a line on the right; having a line on the left; having a line above; having a line below; and the area occupied by the cell. In preferred embodiments, the attribute set also included a subset of attributes dependent on character strings in cells. These attributes were selected from: characters are bold; characters are italic; proportion (e.g. %) of numeric characters; proportion of alphabetic characters; and proportion of non-alphanumeric characters (special characters, symbols, etc.).” paragraph 0052 and “Pre-classification module 29 determines attribute values, corresponding to each of the above attributes, for all cells in the cell map.” Paragraph 0053) ; the first decision module is configured to obtain, based on the feature sequence, parent table cells of the individual table cells in the row direction (“The soft labels are supplied to cell analysis module 30 which implements a Markov logic network 61, described further below, to identify the corner data cell. All cells are then classified as either a row header cell, a column header cell, or a data cell.” Paragraph 0053 and “It will be appreciated that various changes and modifications can be made to the exemplary embodiments described above. By way of example, other machine learning models, such as neural networks or support vector machine models, may be employed for pre-classification of cells. While MLN models are preferred for CDC detection, other rule-based systems, e.g. fixed rule-sets or Bayesian networks/other probabilistic graph models, could be employed.” Paragraph 0065 and “During inference, the rules are applied to the pre-classified cells by applying the MLN, with probabilities P assigned to predicates for each cell, to determine the most probable CDC for the table. In particular, the MLN will identify the cell associated with the lowest “rule-breaking cost” for the network. Some rules will generally be broken for all cells, and a higher cost is associated with higher rule weights, whereby the cell associated with the minimum total cost is the most likely CDC.” Paragraph 0056, dual column and row rulesets ); and the second decision module is configured to obtain, based on the feature sequence, parent table cells of the individual table cells in the column direction (“The soft labels are supplied to cell analysis module 30 which implements a Markov logic network 61, described further below, to identify the corner data cell. All cells are then classified as either a row header cell, a column header cell, or a data cell.” Paragraph 0053 and “It will be appreciated that various changes and modifications can be made to the exemplary embodiments described above. By way of example, other machine learning models, such as neural networks or support vector machine models, may be employed for pre-classification of cells. While MLN models are preferred for CDC detection, other rule-based systems, e.g. fixed rule-sets or Bayesian networks/other probabilistic graph models, could be employed.” Paragraph 0065 and “During inference, the rules are applied to the pre-classified cells by applying the MLN, with probabilities P assigned to predicates for each cell, to determine the most probable CDC for the table. In particular, the MLN will identify the cell associated with the lowest “rule-breaking cost” for the network. Some rules will generally be broken for all cells, and a higher cost is associated with higher rule weights, whereby the cell associated with the minimum total cost is the most likely CDC.” Paragraph 0056, dual column and row rulesets ). With respect to claim 17, Foncubierta teaches the device of claim 16, and all additional limitations in consideration of claim 4, due to the substantial similarities between claim 17 and claim 4, with claim 17 being directed towards the device that performs the method of claim 4. Claims 5, 6, 18, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Foncubierta as applied to claims 4 and 17 above, and further in view of Wang (Zhiruo Wang, Haoyu Dong, Ran Jia, Jia Li, Zhiyi Fu, Shi Han, and Dongmei Zhang. 2021. TUTA: Tree-based Transformers for Generally Structured Table Pre-training. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining (KDD '21). Association for Computing Machinery, New York, NY, USA, 1780–1790.). With respect to claim 5, Foncubierta teaches the method of claim 4, with respect to both claim 1 and claim 2 dependent embodiments, but doesn’t explicitly teach the further limitations. Wang further teaches wherein the feature extracting module is an encoder of a Transformer model (“…our encoder with tree attention…” page 1784 3.3 Tree attention lines 27-28 and See figure 3 Embedding layers and Transformer). Wang is analogous art in the same field of endeavor as the claimed invention. Wang is directed towards tree-based table understanding (“… we propose TUTA, a structure-aware method for Table Understanding with Tree-based Attention” page 1781 left column lines 5-6). A person of ordinary skill in the art before the effective filing date of the claimed invention would have found it obvious to combine the tree based hierarchy of Foncubierta with the tree-based TUTA model of Wang, by utilizing Wang’s TUTA model, through simple substitution, with the expectation that doing so would lead to predictable results (see Foncubierta “It will be appreciated that various changes and modifications can be made to the exemplary embodiments described above. By way of example, other machine learning models, such as neural networks or support vector machine models, may be employed for pre-classification of cells. While MLN models are preferred for CDC detection, other rule-based systems, e.g. fixed rule-sets or Bayesian networks/other probabilistic graph models, could be employed.” Paragraph 0065). With respect to claim 6, Foncubierta teaches the method of claim 4, with respect to both claim 1 and claim 2 dependent embodiments, but doesn’t explicitly teach the further limitations. Wang further teaches wherein both the first decision module and the second decision module are self-attention modules (See figure 3 Transformer and “Motivated by the general idea of graph attention networks [38], we inject the tree structure into the attention mechanism by performing masked attention” page 1784 3.3 Tree attention lines 13-15, masked self attention). Wang is analogous art in the same field of endeavor as the claimed invention. Wang is directed towards tree-based table understanding (“… we propose TUTA, a structure-aware method for Table Understanding with Tree-based Attention” page 1781 left column lines 5-6). A person of ordinary skill in the art before the effective filing date of the claimed invention would have found it obvious to combine the tree based hierarchy of Foncubierta with the tree-based TUTA model of Wang, by utilizing Wang’s TUTA model, through simple substitution, with the expectation that doing so would lead to predictable results (see Foncubierta “It will be appreciated that various changes and modifications can be made to the exemplary embodiments described above. By way of example, other machine learning models, such as neural networks or support vector machine models, may be employed for pre-classification of cells. While MLN models are preferred for CDC detection, other rule-based systems, e.g. fixed rule-sets or Bayesian networks/other probabilistic graph models, could be employed.” Paragraph 0065). With respect to claim 18, Foncubierta teaches the device of claim 17, and in view of Wang, teaches all additional limitations in consideration of claim 5, due to the substantial similarities between claim 18 and claim 5, with claim 18 being directed towards the device that performs the method of claim 5. With respect to claim 19, Foncubierta teaches the device of claim 17, and in view of Wang, teaches all additional limitations in consideration of claim 6, due to the substantial similarities between claim 19 and claim 6, with claim 19 being directed towards the device that performs the method of claim 6. Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Foncubierta as applied to claim 1 above, and further in view of Yuan (CN 113139445 A). With respect to claim 10, Foncubierta teaches the method of claim 1,but doesn’t teach the additional limitations. Yuan teaches wherein the method further comprises: generating a target table based on the structural coordinates of the individual table cells and texts within the individual table cells (“When a recognition command is detected, the table image to be recognized is acquired, and the table image is processed by extracting the frame lines to obtain a binary image of line segments. The line segments in the binary image of line segments are then spliced together to obtain a set of spliced line segments. The splicing segment set is processed by cell extraction to obtain a first table, and the first table is checked to determine whether there are any missed lines in the first table; If there are any missed lines in the first table, the cells of the first table are divided into two parts to obtain the second table. The cells of the second table are sliced, and text information is extracted from the sliced cells. The second table and the text information are then standardized and output to obtain the target table.” Translated Yuan pages 7-8 paragraphs 0006-0009). Yuan is analogous art in the same field of endeavor as the claimed invention. Yuan is directed towards table recognition (“When a recognition command is detected, the table image to be recognized is obtained.” Translated Yuan page 40 paragraph 2). A person of ordinary skill before the expected filing date of the claimed invention would have found it obvious to combine the queryable table extraction teachings of Foncubierta with the target table output of Yuan, with the expectation that doing so would enable further user interaction (“after the system identifies the table image and outputs the editable table, obtaining the operation instruction for editing the table; according to the obtained operation instruction, editing the table, editing the content comprises the character information in the table the cell for example, changing and deleting the character information in the the cell adding and so on; if the character information in the the cell is number, also can be inserted in the formula to calculate the number in the the cell Further, the table style can be set according to the operation instruction of the user, for example, filling the the cell in the table, changing the font style of the character information, so as to satisfy the individual setting of different users.” Page 12 (last paragraph) -13 (first line)). Response to Arguments Applicant’s arguments filed 08/17/2026 have been fully considered. First, on page 9 of applicant remarks, the applicant mentions the objection to specification indication on the previous office action. The examiner recognizes this and has provided the previously missing corresponding objection (See above). Second, in regards to applicant’s arguments against the previously used combination of Yuan and Wang, upon further search and consideration and in light of the newly amended claims a new prior art reference was found and used (See updated rejection). Accordingly, applicant’s arguments that Yuan, Wang, or a combination of Wang and Yuan fail to teach all limitations are rendered moot. Additionally, applicant’s arguments against the combination of Yuan and Wang are rendered moot, due to them no longer being combined. Conclusion Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a). A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action. Any inquiry concerning this communication or earlier communications from the examiner should be directed to REBECCA C WILLIAMS whose telephone number is (571)272-7074. The examiner can normally be reached M-F 7:30am - 4:00pm. 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 W Bee can be reached at (571)270-5183. 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. /REBECCA COLETTE WILLIAMS/Examiner, Art Unit 2677 /ANDREW W BEE/Supervisory Patent Examiner, Art Unit 2677
Read full office action

Prosecution Timeline

Jul 08, 2024
Application Filed
May 15, 2026
Non-Final Rejection mailed — §102, §103
Aug 07, 2026
Response Filed
Nov 13, 2026
Final Rejection mailed — §102, §103 (current)

Precedent Cases

Applications granted by this same examiner with similar technology

Patent 12762729
FLOORPLAN-AWARE CAMERA POSE REFINEMENT METHOD AND SYSTEM
2y 1m to grant Granted Oct 13, 2026
Patent 12718365
INTELLIGENT PLAN OPTIMIZATION METHOD AND SYSTEM
2y 4m to grant Granted Aug 25, 2026
Patent 12705526
Image processing using photonic quantum computing
3y 7m to grant Granted Aug 11, 2026
Patent 12633080
SYSTEMS AND METHODS FOR INSPECTION OF GAS PLUME USING OBJECT DETECTION AND SEGMENTATION MODELS
1y 5m to grant Granted May 19, 2026
Patent 12626335
IMAGE PROCESSING METHOD, ELECTRONIC DEVICE, AND STORAGE MEDIUM
2y 10m to grant Granted May 12, 2026
Study what changed to get past this examiner. Based on 5 most recent grants.

Strategy Recommendation AI-generated — please review before filing

Get a prosecution strategy drawn from examiner precedents, rejection analysis, and claim mapping.
Typically takes 5-10 seconds — AI-generated, attorney review required before filing

Prosecution Projections

3-4
Expected OA Rounds
50%
Grant Probability
99%
With Interview (+58.3%)
3y 3m (~10m remaining)
Median Time to Grant
Moderate
PTA Risk
Based on 14 resolved cases by this examiner. Grant probability derived from career allowance rate.

Sign in with your work email

Enter your email to receive a magic link. No password needed.

Personal email addresses (Gmail, Yahoo, etc.) are not accepted.

Free tier: 3 strategy analyses per month