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 .
Claim Status
Claim(s) 1, 8-11, 19-21, 26-27 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Stumpf (“"Uncertainty-guided sampling to improve digital soil maps", 2017; as cited in IDS Filed 03/10/2025).
Claim(s) 2-5 is/are rejected under 35 U.S.C. 103 as being unpatentable over Stumpf, in view of Taghizadeh-Mehrjardi (“Bio-inspired hybridization of artificial neural networks: An application for mapping the spatial distribution of soil texture fractions”, 2021; as cited in IDS filed 5/14/2024).
Claim(s) 12 is/are rejected under 35 U.S.C. 103 as being unpatentable over Stumpf, in view of Barnes (“Remote-and ground-based sensor techniques to map soil properties”, 2003).
Claim(s) 15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Stumpf, in view of Ringrose-Voase (“Four Pillars of digital land resource mapping to address information and capacity shortages in developing countries”; as cited in IDS filed 03/10/2025).
Claim(s) 22 is/are rejected under 35 U.S.C. 103 as being unpatentable over Stumpf, in view of Kulkarni (“Efficient learning of random forest classifier using disjoint partitioning approach”, 2013).
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 05/14/2024 and 03/10/2025 was filed and is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Response to Amendment
The preliminary amendment filed on 09/29/2023 has been entered. Claims 5, 12, 15, 17, 23, 26, 27 were amended. Claims 1-5, 8-12, 15, 17, 19-27 remain pending in the application.
Claim Interpretation
Regarding claims 5, 12, 15, 17, 23 and the usage of “optionally” or the optional limitations. As these are method claims, the method may never reach the “optional” steps. Applicant may choose to amend these limitations so they are no longer optional.
Regarding claim 23 and usage of the 100km threshold for altitude. As one with ordinary skill in the art would know, this boundary is the Karman line. Usually imagery from above this threshold is from satellite and below would be from other aircraft (potentially drones or UAVs).
Claim Rejections - 35 USC § 102
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.
Claim(s) 1, 8-11, 19-21, 26-27 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Stumpf (“"Uncertainty-guided sampling to improve digital soil maps", 2017; as cited in IDS Filed 03/10/2025).
Regarding claim 1, Stumpf teaches A method for training a soil mapping model (Stumpf, pg 31, column 1, 2 paragraphs before the last two lines, reproduced below:
PNG
media_image1.png
288
776
media_image1.png
Greyscale
. DSM is Digital Soil Mapping, from Abstract. Digital Soil Mapping is being interpreted as involving “soil mapping model”), the method performed by a processor and comprising:
Training (Stumpf, pg 31, column 2, Section 2.2, ¶1, reproduced below:
PNG
media_image2.png
296
784
media_image2.png
Greyscale
. “Learner” shows “training” occurs for learning to happen) a set of initial parameters (Stumpf, see Section 2.3 image below, “two initial RF approaches” are being interpreted as “set of initial parameters”) for a soil mapping model relating one or more soil characteristics (Stumpf, see pg 31, column 1 image below: “silt and clay contents” is being interpreted as “one or more soil characteristics”) to one or more covariates over an area of interest (Stumpf, pg 31, column 1, paragraph before Section 2.3, reproduced below:
PNG
media_image3.png
224
780
media_image3.png
Greyscale
. “Terrain attributes as covariates (Table 1)” is being interpreted as “one or more covariates over an area of interest”) based on an initial training dataset (Stumpf, pg 31, column 2, Section 2.1, ¶3, reproduced below:
PNG
media_image4.png
572
1162
media_image4.png
Greyscale
. “Calibration set (LD)” is being interpreted as “initial training dataset”.);
determining an uncertainty map (Stumpf, pg 31, column 1, paragraph before Section 2.3, image provided above, “compute uncertainty maps for both approaches”) for the soil mapping model over at least a portion of the area of interest (Stumpf, pg 31, column 1, paragraph before Section 2.3, reproduced below:
PNG
media_image3.png
224
780
media_image3.png
Greyscale
. “Compute uncertainty maps for both [initial] approaches”) based on the initial parameters (Stumpf, pg 31, column 1, paragraph before Section 2.3, reproduced below:
PNG
media_image3.png
224
780
media_image3.png
Greyscale
. “Compute uncertainty maps for both [initial] approaches”. “Initial approaches” are being interpreted having “initial parameters”), the uncertainty map mapping each of a plurality of locations in the area of interest (Stumpf, see most recent image above: “compute uncertainty maps for both approaches” is being interpreted involving “locations in the area of interest”) to a measure of uncertainty (Stumpf, see most recent image above: “compute uncertainty maps for both approaches”. Uncertainty maps are being interpreted to involve a measure of uncertainty) in at least one of the one or more soil characteristics (Stumpf, see most recent image above: “predict silt and clay contents” is being interpreted as “one or more soil characteristics”);
determining one or more sampling locations (Stumpf, see image directly below: “to identify areas relevant to acquire additional soil data” is being interpreted to involve “determining one or more sampling locations”) in the area of interest based on the uncertainty map (Stumpf, pg 31, column 1, second to last paragraph, reproduced below:
PNG
media_image5.png
238
1182
media_image5.png
Greyscale
. “Spatial uncertainty analysis” is being interpreted to involve “the uncertainty map”, as can be seen in Section 2.2 and 2.3, in the area of interest);
receiving one or more sample measurements (Stumpf, see most recent image above: “using a spatial uncertainty analysis to acquire additional soil data”. “Acquire additional soil data” is being interpreted as “receiving one or more sample measurements”) corresponding to the one or more sampling locations (Stumpf, see most recent image above: “to identify areas relevant to acquire additional soil data”. “Identify areas relevant” is being interpreted as involving “one or more sampling locations”); and
refining the set of initial parameters (Stumpf, see image below, “We refined the initial approaches”. “Initial approaches” are being interpreted as “initial parameters”) of the soil mapping model (Stumpf, see image below, the initial RF approaches for the silt and clay are being interpreted as coming from a digital soil mapping model) based on the one or more sample measurements (Stumpf, see image below: “Sampling design” shows that “one or more sample measurements” were used in refining as they are using the high uncertainty areas) to generate a first set of refined parameters for the soil mapping model (Stumpf, pg 32, starts in column 1, Section 2.3, ¶1, reproduced below:
PNG
media_image6.png
332
788
media_image6.png
Greyscale
.”We refined the initial approaches”).
Regarding claim 8, Stumpf teaches The method according to claim 1 wherein:
the plurality of locations mapped by the uncertainty map (Stumpf, see pg 32 image below: “identifying an area of high uncertainties”. “Area” is being interpreted to involves mapped locations. “Area of high uncertainties” is being interpreted as involving “the uncertainty map”) comprises a plurality of regions in the area of interest (Stumpf, ); and
determining the one or more sampling locations (Stumpf, see pg 32 image below, “sampling design for the additional samples is based on identifying an area of high uncertainties…”. “Area” is being interpreted as involving locations that are determined for sampling) comprises at least one of:
random sampling (Stumpf, see pg 31 image below: “simple random sampling”), uniform sampling, stratified sampling (Stumpf, pg 31, column 2, Section 2.1, ¶3, reproduced below:
PNG
media_image7.png
382
782
media_image7.png
Greyscale
. “Stratified random sampling”), and conditioned Latin hypercube sampling (Stumpf, pg 32, column 2, lines 1-4, reproduced below:
PNG
media_image8.png
126
786
media_image8.png
Greyscale
), and orthogonal sampling in the at least one region.
Regarding claim 9, Stumpf teaches The method according to claim 8 wherein the method determining the one or more sampling locations comprises determining (Stumpf, see pg 32 image below, “determine a subarea of increased uncertainty, and thus to obtain additional samples”. Which is being interpreted as determining one or more sampling locations), for each one of the plurality of regions (Stumpf, see pg 32 image below, “we defined four potential sampling areas” is being interpreted as “each one of the plurality of regions”), a measure of aggregate uncertainty for the region (Stumpf, see pg 32 image below, “> 50%” involves the median, as one with ordinary skill in the art would know, median is being interpreted as involving aggregate uncertainty); and
selecting the at least one region based on the measures of aggregate uncertainty (Stumpf, pg 32, column 2, first 3 paragraphs:
PNG
media_image9.png
738
778
media_image9.png
Greyscale
. “determine a subarea of increased uncertainty, and thus to obtain additional samples” is being interpreted to involve region selection based on aggregate uncertainty).
Regarding claim 10, Stumpf teaches The method according to claim 1 wherein determining one or more sampling locations in the area of interest (Stumpf, see pg 32 image below, “we defined four potential sampling areas”) comprises selecting a first one of a plurality of candidate sampling locations (Stumpf, see pg 32 image below, “we defined four potential sampling areas”. “Potential sampling areas” is being interpreted as involving selecting a candidate sampling location. For example, if the lowest quartile range is chosen, then it would be the first candidate sampling location selected), the first candidate sampling location being mapped to a measure of uncertainty by the uncertainty map (Stumpf, see pg 32 image below, “RFLD_silt/clay” is the uncertainty map that has uncertainty values [measure of uncertainty] for each pixel. For example, if the lowest quartile range is chosen, then it would be the first candidate sampling location selected) at least as great as measures of uncertainty mapped to by any other candidate sampling location (Stumpf, see pg 32 image below, “we defined four potential sampling areas”. “Potential sampling areas” is being interpreted as involving selecting a candidate sampling location. For example, if the highest quartile range is chosen, then it would be at least as great as measures of uncertainty mapped to by any other candidate sampling location) of the plurality of candidate sampling locations (Stumpf, pg 32, column 2, first 3 paragraphs:
PNG
media_image9.png
738
778
media_image9.png
Greyscale
.”we defined four potential sampling areas” is being interpreted as “plurality of candidate sampling locations”).
Regarding claim 11, Stumpf teaches The method according to claim 1 wherein receiving the one or more sample measurements (Stumpf, see pg 31 image below, “acquire additional soil data” is being interpreted to involve receiving the one or more sample measurements”) corresponding to the one or more sampling locations (Stumpf, see pg 31 image below, “to identify areas relevant to acquire additional soil data”. “Areas” are being interpreted as involving “one or more sampling locations”) comprises generating a request for the one or more sample measurements of soil at the one or more sampling locations (Stumpf, pg 31, column 1, two paragraphs before the end, reproduced below:
PNG
media_image10.png
430
1166
media_image10.png
Greyscale
. “identify areas relevant to acquire additional soil data” is being interpreted as involving “generating a request” as the additional soil data does not just suddenly appear).
Regarding claim 19, Stumpf teaches The method according to claim 1 wherein:
the soil mapping model (Stumpf, pg 31, column 1, 3rd to last paragraph, reproduced below:
PNG
media_image11.png
124
778
media_image11.png
Greyscale
. “DSM” is “Digital Soil Mapping”, as shown in the Abstract.) comprises a machine learning model (Stumpf, see image below, “RF” is the acronym for “Random Forest” regression algorithm, as seen in Section 2.2, line 1 on pg 31, which is being interpreted as a machine learning model), the machine learning model (Stumpf, see image below, “RF” is the acronym for “Random Forest” regression algorithm, as seen in Section 2.2, line 1 on pg 31, which is being interpreted as a machine learning model) configured to receive the one or more covariates as input (Stumpf, see image below, “terrain attributes as covariates”) and to generate predictions for the one or more soil characteristics as output (Stumpf, see image below, “predict silt and clay contents”), the predictions (Stumpf, see image below, “predict silt and clay contents”) associated with a measure of confidence (Stumpf, pg 32, column 1, paragraph before Section 2.3, reproduced below:
PNG
media_image12.png
334
1172
media_image12.png
Greyscale
err_var is being interpreted as a measure of confidence.); and
determining an uncertainty map for the soil mapping model comprises (Stumpf, see image above, “we used err_var to compute uncertainty maps”), for each of the plurality of locations (Stumpf, pg 35, Figure 4, which shows the predicted silt and clay contents for a plurality of locations), generating a prediction for the location and determining the measure of uncertainty for the location (Stumpf, see image above: “to computer uncertainty maps”, which is being interpreted to include locations on a map) based on the measure of confidence for the prediction (Stumpf, see image above: “we use err_var to compute uncertainty maps”. Err_var is being interpreted as “measure of confidence”).
Regarding claim 20, Stumpf teaches The method according to claim 19 wherein the machine learning model comprises at least one of:
a neural network and a random forest (Stumpf, pg 31, column 2, Section 2.2, Line 1: “We used the Random Forest (RF) regression algorithm”. “At least one of: a neural network and a random forest” is being interpreted as “a neural network or a random forest”. “Random forest” is being interpreted as a “machine learning model”).
Regarding claim 21, Stumpf teaches The method according to claim 1 wherein:
the soil mapping model comprises an ensemble of soil sub-models (Stumpf, pg 31, column 2, Section 2.2, ¶1, reproduced below:
PNG
media_image13.png
430
1172
media_image13.png
Greyscale
. “Multiple randomized decision tree models” are being interpreted as including “an ensemble of soil sub-models”);
determining an uncertainty map for the soil mapping model (Stumpf, pg 32, column 1, paragraph before Section 2.3, reproduced below:
PNG
media_image14.png
336
1168
media_image14.png
Greyscale
. “Computer uncertainty maps”) comprises,
for each of the plurality of locations (Stumpf, see pg 32 image above, “uncertainty maps” are being interpreted to include the plurality of locations),
generating a prediction for the one or more soil characteristics at the location (Stumpf, see pg 32 image above, “predict silt and clay contents”) by each of the soil sub-models (Stumpf, see pg 31 image above, “Multiple randomized decision tree models” are being interpreted as including “an ensemble of soil sub-models”),
thereby generating a plurality of predictions for the location (Stumpf, see pg 32 image above, “predict silt and clay contents”; pg 33, Figure 2—which shows predictions for the location), and
determining the measure of uncertainty for the location (Stumpf, see pg 32 image above, “compute uncertainty maps”, which is being interpreted as involving determine “the measure of uncertainty”) based on the plurality of predictions (Stumpf, see pg 32 image above, “predict silt and clay contents” is being interpreted as involving “plurality of predictions”).
Regarding claim 26, Stumpf teaches A method for mapping soil characteristics (Stumpf, pg 32, starts in column 1, Section 2.3, ¶1, reproduced below:
PNG
media_image6.png
332
788
media_image6.png
Greyscale
.”We refined the initial approaches”. “Initial RF approaches” and “soil properties” are being interpreted as “method for mapping soil characteristics” because of the “digital soil mapping” from the prior art), the method (Stumpf, see image above, “RF approaches”, or Random Forest approaches) performed by a processor (Stumpf, see image above, Random Forest is being interpreted as an algorithm that is performed by a processor) and comprising:
mapping at least one location in an area of interest (Stumpf, see image above, “identifying an area” is being interpreted to involve “mapping at least one location in an area of interest”) to at least one predicted value of at least one soil characteristics (Stumpf, see image below, “observations of silt and clay contents” are being interpreted as “at least one predicted value of at least one soil characteristics”) by a soil mapping model (Stumpf, pg 31, column 2, Section 2.2, ¶1, reproduced below:
PNG
media_image15.png
298
784
media_image15.png
Greyscale
”Random Forest regression algorithm” is being interpreted as part of a soil mapping model),
the at least one predicted value based on a set of refined parameters (Stump, see Section 2.3 image above: “we refined the initial approaches” is being interpreted as “refined parameters”. “Initial approaches” is being interpreted as “at least one predicted value”) for the soil mapping model trained according to claim 1 (Stumpf, pg 31, column 1, 2 paragraphs before the last two lines, reproduced below:
PNG
media_image1.png
288
776
media_image1.png
Greyscale
. DSM is Digital Soil Mapping, from Abstract. Digital Soil Mapping is being interpreted as involving “soil mapping model”).
Regarding claim 27, Stumpf teaches A computer system comprising:
one or more processors (Stumpf, pg 31, column 2, Section 2.2, line 1: “We used the Random Forest (RF) regression algorithm”. Algorithm is being interpreted to use one or more processors); and
a memory storing instructions which cause the one or more processors to perform operations comprising the method of claim 1 (Stumpf, pg 31, column 2, Section 2.2, line 1: “We used the Random Forest (RF) regression algorithm”. Algorithm is being interpreted as using a memory storing instructions, the instructions being involved with the algorithm).
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.
Claim(s) 2-5 is/are rejected under 35 U.S.C. 103 as being unpatentable over Stumpf, in view of Taghizadeh-Mehrjardi (“Bio-inspired hybridization of artificial neural networks: An application for mapping the spatial distribution of soil texture fractions”, 2021; as cited in IDS filed 5/14/2024).
Regarding claim 2, Stumpf teaches The method according to claim 1 comprising mapping at least one location (Stumpf, pg 31, column 1, second to last paragraph, reproduced below:
PNG
media_image5.png
238
1182
media_image5.png
Greyscale
. “Spatial uncertainty analysis” is being interpreted to involve “the uncertainty map”, as can be seen in Section 2.2 and 2.3, in the area of interest that contains at least one location) in the area of interest (Stumpf, see previous image, “Spatial uncertainty analysis” is being interpreted to involve “the uncertainty map”, as can be seen in Section 2.2 and 2.3, in the area of interest that contains at least one location) to at least one predicted value of at least one of the one or more soil characteristics, the at least one predicted value based on at least one of:
However, Stumpf does not appear to specifically teach the first set of refined parameters and a further set of refined parameters based on the first set of refined parameters for the soil mapping model.
Pertaining to the same field of endeavor, Taghizadeh-Mehrjardi teaches
the first set of refined parameters (Taghizadeh-Mehrjardi, see image below, “Each iteration” shows that a first set of refined parameters occurs after the first PSF [particle size fraction] map generated during the first iteration) and a further set of refined parameters based on the first set of refined parameters for the soil mapping model (Taghizadeh-Mehrjardi, pg 10, paragraph before Section 3, reproduced below:
PNG
media_image16.png
144
908
media_image16.png
Greyscale
. “Each iteration” shows that subsequent iterations use a further set of refined parameters, otherwise no iteration would be done. As stated in the Abstract: “soil texture and particle size fractions (PSFs) are a critical characteristic of soil”, which are being interpreted to involve parameters.).
Stumpf and Taghizadeh-Mehrjardi are considered to be analogous art because they are directed to digital soil mapping with uncertainty-guided refinement. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method and system for digital soil mapping with uncertainty-guided refinement (or a first iteration) (as taught by Stumpf) to include multiple iterations of refinement (as taught by Taghizadeh-Mehrjardi) because the combination provides a reduction in uncertainty (Taghizadeh-Mehrjardi, Abstract).
Regarding claim 3, Stumpf teaches The method according to claim 1 comprising iteratively generating one or more further sets of refined parameters (Stumpf, pg 32, starts in column 1, Section 2.3, ¶1, reproduced below:
PNG
media_image6.png
332
788
media_image6.png
Greyscale
.”We refined the initial approaches”. Which was done with at least one iteration), each set of refined parameters generated by:
However, Stumpf does not appear to specifically teach determining a further uncertainty map.
Pertaining to the same field of endeavor, Taghizadeh-Mehrjardi
determining a further uncertainty map (Taghizadeh-Mehrjardi, pg 10, paragraph before Section 3, reproduced below:
PNG
media_image16.png
144
908
media_image16.png
Greyscale
. “PSF maps generated by each iteration of the models” is being interpreted to involve “a further uncertainty map”) for the soil mapping model (Taghizadeh-Mehrjardi, see image above, PSF is being interpreted as part of the “soil mapping model”) based on at least one of the first set of refined parameters (Taghizadeh-Mehrjardi, see image above, the second iteration and beyond would refine based on at least one of the first set of refined parameters, the parameters involving what creates the PSF maps) and the one or more further sets of refined parameters (Taghizadeh -Mehrjardi, see image above, the second iteration and beyond would refine based on at least one of the first set of refined parameters, the parameters involving what creates the PSF maps);
determining one or more further sampling locations in the area of interest (Taghizadeh-Mehrjardi, pg 7, Section 2.4.2, ¶1, reproduced below:
PNG
media_image17.png
398
1214
media_image17.png
Greyscale
”select the parents” is being interpreted as “determining one or more further sampling locations”. The genetic algorithm is used as part of the digital soil mapping model, as seen in pg 5, Section 2.4, lines 1-2) based on the further uncertainty map (Taghizadeh-Mehrjardi, see pg 10 image above, uncertainty analysis is being interpreted as uncertainty map);
receiving one or more further sample measurements (Taghizadeh-Mehrjardi, see pg 10 image above, each iteration a PSF map is generated, which requires further sample measurements to create the PSF map) corresponding to the one or more further sampling locations (Taghizadeh-Mehrjardi, see Section 2.4.2 image above, the parent selection is being interpreted as involving a sampling location as each sampling location allows for the creation of a PSF map); and
refining at least one of the first set of refined parameters (Taghizadeh-Mehrjardi, see pg 10 image above. Each iteration, the PSF maps are refined. After the first iteration, a first set of refined parameters are refined) and at least one further sets of refined parameters based on the one or more sample measurements (Taghizadeh-Mehrjardi, see pg 10 image above, each iteration a PSF map is generated, which requires further sample measurements to create the PSF map. Figure 4 contains the covariates used to create the PSF maps) to generate a further set of refined parameters for the soil mapping model (Taghizadeh-Mehrjardi, see pg 10 image above. Each iteration, the parameters are refined that are used to create the PSF maps which are part of the soil mapping model).
Stumpf and Taghizadeh-Mehrjardi are considered to be analogous art because they are directed to digital soil mapping with uncertainty-guided refinement. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method and system for digital soil mapping with uncertainty-guided refinement (or a first iteration) (as taught by Stumpf) to include determining a further uncertainty map (as taught by Taghizadeh-Mehrjardi) because the combination provides a reduction in uncertainty (Taghizadeh-Mehrjardi, Abstract).
Regarding claim 4, Stumpf teaches The method according to claim 3
However, Stumpf does not appear to specifically teach further sets of refined parameters for the soil mapping model.
Pertaining to the same field of endeavor, Taghizadeh-Mehrjardi
comprising mapping at least one location in the area of interest to at least one predicted value (Taghizadeh-Mehrjardi, see pg 10 image below, PSF maps are being interpreted as involving predicting, as a hybridized ANN is involved that does predictions for at least one location in the area of interest. As pg 4, Figure 1 shows a sampling points for areas of interest in a map) of at least one of the one or more soil characteristics (Taghizadeh-Mehrjardi, Abstract: “soil texture and particle size fractions (PSFs) are a critical characteristic of soil”), the at least one predicted value (Taghizadeh-Mehrjardi, pg 10, paragraph before Section 3, reproduced below:
PNG
media_image16.png
144
908
media_image16.png
Greyscale
. “PSF maps generated by each iteration of the models” are being interpreted to involve “at least one predicted value”) based on at least one of the further sets of refined parameters for the soil mapping model (pg 7, Section 2.4.2, reproduced below:
PNG
media_image18.png
250
760
media_image18.png
Greyscale
. “Optimizing the weights and bias of neural network connections” is being interpreted as “further sets of refined parameters” that occur after a number of iterations).
Stumpf and Taghizadeh-Mehrjardi are considered to be analogous art because they are directed to digital soil mapping with uncertainty-guided refinement. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method and system for digital soil mapping with uncertainty-guided refinement (or a first iteration) (as taught by Stumpf) to include further sets of refined parameters for the soil mapping model (as taught by Taghizadeh-Mehrjardi) because the combination provides a reduction in uncertainty (Taghizadeh-Mehrjardi, Abstract).
Regarding claim 5, Stumpf teaches The method according to claim 3 wherein iteratively generating the one or more further sets of refined parameters (Taghizadeh-Mehrjardi, pg 10, paragraph before Section 3, reproduced below:
PNG
media_image16.png
144
908
media_image16.png
Greyscale
. “PSF maps generated by each iteration of the models” is being interpreted as “iteratively generating the one or more further sets of refined parameters”) comprises determining a measure of model uncertainty based on at least one of the uncertainty maps (Taghizadeh-Mehrjardi, pg 12, last paragraph, reproduced below:
PNG
media_image19.png
304
1054
media_image19.png
Greyscale
. “Evaluation of the uncertainty estimates” is being interpreted as “measure of model uncertainty”); and halting generation of further refined parameters (pg 7, Section 2.4.1, ¶1: “for adjusting the weights and bias until at least one stopping criteria is reached”. “weights and bias” are being interpreted as “refined parameters”) based on the measure of model uncertainty (pg 7, Section 2.4.1., ¶1: “The maximum number of epochs and the MSE of the network output for each target PSFs were the two stopping criteria”. MSE, mean squared error, is being interpreted as being based on model uncertainty);
Examiner notes that the following limitations are optional for a method claim. The method may never reach the optional limitations (though Stumpf does teach average of uncertainty measures through variance, which requires an average, as one with ordinary skill in the art would know), and therefore will optional limitations will not be considered.
optionally wherein the measure of model uncertainty comprises an average of measures of uncertainty for the plurality of locations in the area of interest; or
optionally wherein halting generation of further refined parameters based on the measure of model uncertainty comprises determining that the measure of model uncertainty is less than a threshold uncertainty value.
Stumpf and Taghizadeh-Mehrjardi are considered to be analogous art because they are directed to digital soil mapping with uncertainty-guided refinement. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method and system for digital soil mapping with uncertainty-guided refinement (or a first iteration) (as taught by Stumpf) to include determining a measure of model uncertainty based on at least one of the uncertainty maps (as taught by Taghizadeh-Mehrjardi) because the combination provides a reduction in uncertainty (Taghizadeh-Mehrjardi, Abstract).
Claim(s) 12 is/are rejected under 35 U.S.C. 103 as being unpatentable over Stumpf, in view of Barnes (“Remote-and ground-based sensor techniques to map soil properties”, 2003).
Regarding claim 12, Stumpf teaches The method according to claim 1 comprising measuring soil at the one or more sampling locations (Stumpf, see image below, “to identify areas relevant”) to generate the one or more sample measurements (Stumpf, pg 31, column 1, two paragraphs before the end, reproduced below:
PNG
media_image10.png
430
1166
media_image10.png
Greyscale
. “acquire additional soil data” is being interpreted as involving “generate the one or more sample measurements”):
Examiner notes that the following limitations are optional for a method claim. The method may never reach the optional limitations, and therefore will optional limitations will not be considered. However, if applicant chooses to amend the limitations so that they are no longer optional, then the following claim mapping may apply.
However, Stumpf does not appear to explicitly teach optional steps of using proximal sensors to collect soil data.
Pertaining to the same field of endeavor, Barnes teaches
optionally wherein measuring soil at the one or more sampling locations (Barnes, see image below: “which allows for the on-the-go simultaneous collection of GPS referenced horizontal and vertical EM38 signal data.” “GPS” shows sampling locations) comprises generating the one or more sample measurements by proximal sensing (Barnes, see image below, “EC sensors” show “proximal sensing”. “Assessing irrigation, drainage, and salinity management using conductivity survey data” shows “generating the one or more sample measurements by proximal sensing”);
further optionally wherein generating the one or more sample measurements by proximal sensing (Barnes, see image below, “EC sensors” show “proximal sensing”. “Assessing irrigation, drainage, and salinity management using conductivity survey data” shows “generating the one or more sample measurements by proximal sensing”) comprises inserting a proximal sensor into soil at the one or more sampling locations (Barnes, pg 6, column 1 to column 2 lines 1-5, reproduced below:
PNG
media_image20.png
789
814
media_image20.png
Greyscale
”an electrode-based sensor requiring soil contact” is being interpreted to involve “inserting a proximal sensor into soil”).
Stumpf and Barnes are considered to be analogous art because they are directed to digital soil mapping. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method and system for digital soil mapping with uncertainty-guided refinement (as taught by Stumpf) to include optional steps of using proximal sensors to collect soil data (as taught by Barnes) because the combination provides an improvement to soil monitoring and mapping (Barnes, Abstract).
Claim(s) 15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Stumpf, in view of Ringrose-Voase (“Four Pillars of digital land resource mapping to address information and capacity shortages in developing countries”; as cited in IDS filed 03/10/2025).
Regarding claim 15, Stumpf teaches The method according to claim 1
However, Stumpf does not appear to explicitly teach soil carbon content.
Pertaining to the same field of endeavor, Ringrose-Voase teaches
wherein the one or more soil characteristics (Ringrose-Voase, see image below, “range of soil parameters”) comprise a measure of soil carbon content (Ringrose-Voase, see pg 306, Section 3.3 ¶1 image below, “organic carbon” is being interpreted as “soil carbon content”);
optionally wherein the measure of soil carbon content comprises a measure of soil organic carbon content (Ringrose-Voase, pg 306, Section 3.3 ¶1, reproduced below:
PNG
media_image21.png
668
1174
media_image21.png
Greyscale
”Organic carbon” is being interpreted as “a measure of soil organic carbon content”, as “Organic C” is a “Soil property in Table 5 on the same page. More broadly speaking, this is also “a measure of soil carbon content”).
Stumpf and Ringrose-Voase are considered to be analogous art because they are directed to digital soil mapping. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method and system for digital soil mapping (as taught by Stumpf) to include soil carbon content (as taught by Ringrose-Voase) because the combination provides an improvement to soil surveys (Ringrose-Voase, Abstract).
Claim(s) 22 is/are rejected under 35 U.S.C. 103 as being unpatentable over Stumpf, in view of Kulkarni (“Efficient learning of random forest classifier using disjoint partitioning approach”, 2013).
Regarding claim 22, Stumpf teaches The method according to claim 21 wherein a first soil sub-model of the ensemble (Stumpf, pg 31, column 2, Section 2.2, reproduced below:
PNG
media_image22.png
442
1178
media_image22.png
Greyscale
. “Randomized decision tree models” is being interpreted to include “a first soil sub-model of the ensemble”) comprises initial parameters trained over a first training dataset (Stumpf, see image below: “form the calibration set” is being interpreted as “a first training dataset”. Where the random forest regression algorithm would have initial parameters, as one with ordinary skill in the art would know) and a second soil sub-model of the ensemble (Stump, see image above, “multiple randomized decision tree models” that is part of the “non-parametric ensemble learner” is being interpreted to include a second soil sub-model that uses initial parameters) comprises initial parameters trained over a second training dataset (Stumpf, pg 31, column 2, Section 2.1, ¶3, reproduced below:
PNG
media_image23.png
572
1162
media_image23.png
Greyscale
. “Second calibration set”),
However, Stumpf does not appear to explicitly disjoint datasets.
Pertaining to the same field of endeavor, Kulkarni teaches
the first and second datasets being disjoint (Kulkarni, pg 2, column 2, Section III, ¶1, partially reproduced below:
PNG
media_image24.png
1010
768
media_image24.png
Greyscale
“Disjoint sets of data samples” are being interpreted to include a first and second dataset).
Stumpf and Kulkarni are considered to be analogous art because they are directed to ensemble learning. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method and system for random forest regression algorithm for digital soil mapping (as taught by Stumpf) to include disjoint datasets (as taught by Kulkarni) because the combination provides an improvement to learning time of Random Forest (Kulkarni, Abstract).
Allowable Subject Matter
Claims 23 and 25 is/are allowed.
The following is an examiner’s statement of reasons for allowance:
Regarding claim 23: The cited prior art fails to disclose, teach, or suggest: “The method according to claim 22 wherein the first training dataset comprises images of the area of interest captured from an altitude of no more than 100 km and the second training dataset comprises images of the area of interest captured from an altitude of no less than 100 km; optionally wherein the first training dataset comprises images captured by aircraft and the second training dataset comprises images captured by satellite” in the context of the claim as a whole.
Regarding claim 25: The cited prior art fails to disclose, teach, or suggest: “The method according to claim 22 wherein the first training dataset comprises images having a first spatial density and the second training dataset comprises images having a second spatial density less than the first spatial density, such that an element of the second training dataset corresponds spatially to a plurality of elements of the first training dataset” in the context of the claim as a whole.
The cited prior art includes Gray et al (“Integrating Drone Imagery into High Resolution Satellite Remote Sensing Assessments of Estuarine Environments”, 2018) discloses combining greater than 100 km altitude imagery (satellite, which may capture a different spatial resolution, or density, than the other) and less than 100 km imagery (UAS or drone or unoccupied aircraft systems; which may capture ultra-high spatial resolution or high spatial density) for habitat monitoring. However, Gray does not appear to specifically teach disjoint datasets for training as they are combined and soil mapping. Further, there appears to be no motivation to combine with the cited prior art.
Any comments considered necessary by applicant must be submitted no later than the payment of the issue fee and, to avoid processing delays, should preferably accompany the issue fee. Such submissions should be clearly labeled “Comments on Statement of Reasons for Allowance.”
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOHNNY B DUONG whose telephone number is (571)272-1358. The examiner can normally be reached Monday - Thursday 10a-9p (ET).
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, Matthew Bella can be reached at (571)272-7778. 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.
/J.B.D./Examiner, Art Unit 2667
/MATTHEW C BELLA/Supervisory Patent Examiner, Art Unit 2667