Patent Application 17833210 - MOISTURE AND ORGANIC MATTER PREDICTION USING

Title: MOISTURE AND ORGANIC MATTER PREDICTION USING NEAR INFRARED LIGHT

Application Information

Invention Title: MOISTURE AND ORGANIC MATTER PREDICTION USING NEAR INFRARED LIGHT
Application Number: 17833210
Submission Date: 2025-04-09T00:00:00.000Z
Effective Filing Date: 2022-06-06T00:00:00.000Z
Filing Date: 2022-06-06T00:00:00.000Z
National Class: 702
National Sub-Class: 002000
Examiner Employee Number: 95628
Art Unit: 2863
Tech Center: 2800

Rejection Summary

102 Rejections: 0
103 Rejections: 4

Cited Patents

No patents were cited in this rejection.

Office Action Text

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 .
Response to Arguments
Applicant’s arguments, see page 7, filed 01/29/2025, with respect to claim rejections under 35 U.S.C. 112 have been fully considered, along with amendments, and are persuasive, insofar that it is clearer how the Applicant believes the neural network is improved and that claim 15 has been canceled. The rejections of claims 1-9 and 15 under 35 U.S.C. 112 have been withdrawn.
Applicant's arguments filed 01/29/2025, with respect to rejections of claims 1-16 under 35 U.S.C. 101 have been fully considered but they are not persuasive.
Beginning on page 9, applicant remarks that improving the ability of the neural network system’s ability to predict percentage of organic matter in a soil sample by inputting ambient humidity level cannot be practically performed in the mind. This is not persuasive because “improving” the ability of predicting percentage of organic matter in a soil sample is practically performable in the mind as it is simply analyzing data about the soil, and it is known that providing more relevant data will improve a prediction. Using a neural network in its ordinary capacity is considered mere instructions to apply an exception (see MPEP 2106.05(f)(2)).
Applicant then states that the claimed invention improves the functioning of a computer or improves another technology or technical field and the specification and claim reflect the improvement. This is not persuasive because as stated above, the Neural network itself is not being improved or functioning differently, it is just being provided more data.
Applicant’s remarks about claim 10 being impractical to perform in the human mind are not persuasive, because the claim is merely a computer with a processor and a memory, calling on a neural network to perform a task in its expected capacity. Choosing “unique” inputs is not considered to change or improve how a neural network functions.
Similarly, claim 17 is also merely collecting data and feeding it to a neural network, that performs a task expected of a neural network.
Applicant’s arguments, see page 14, filed 01/29/2025, with respect to the rejection(s) of claim(s) 1-9 under 35 U.S.C. 103 have been fully considered, along with amendments, and are persuasive, insofar that the cited Ge reference does not explicitly teach a neural network that uses humidity. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of Ge et al. (A primer on soil analysis using visible and near-infrared (vis-NIR) and mid-infrared (MIR) spectroscopy, Food and Agriculture Organization of the United Nations, 28 pages, March 2022), hereinafter “Ge”, in view of Alexakis et al. (Integrated Use of Satellite Remote Sensing, Artificial Neural Networks, Field Spectroscopy, and GIS in Estimating Crucial Soil Parameters in Terms of Soil Erosion. Remote Sens. 9 May 2019, 11, 1106. https://doi.org/10.3390/rs11091106), hereinafter “Alexakis”, and Fernandes et al. (Estimation of soil organic matter content by modeling with artificial neural networks, Geoderma, Volume 350, 15 September 2019, Pages 46-51, ISSN 0016-7061, https://doi.org/10.1016/j.geoderma.2019.04.044.), hereinafter “Fernandes”. Alexakis teaches using a neural network to analyze spectral data and determine soil organic material. Fernandes teaches a neural network that uses environmental data such as temperature and humidity to aid in determining soil organic material.
Applicant's arguments, see page 15 filed 01/29/2025, with respect to the rejection(s) of claim(s) 10-14 and 16 under 35 U.S.C. 103 have been fully considered but they are not persuasive.
On page 15 Applicant states that the cited combination of WikiCNN and Ge does not show or suggest a neural network with an ambient stage that receives an ambient humidity and a final stage that receives both the outputs of the spectral and ambient stages. This is not persuasive because the claim under broadest reasonable interpretation still describes a neural network with at least two inputs.
Applicant’s arguments, see page 14, filed 01/29/2025, with respect to the rejection(s) of claim(s) 17-20 under 35 U.S.C. 103 have been fully considered, along with amendments, and are persuasive, insofar that the cited Shen and Ge references do not explicitly teach a neural network that uses humidity. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of Shen et al. (Hyperspectral Estimation of Soil Organic Matter Content using Different Spectral Preprocessing Techniques and PLSR Method. Remote Sens. 2020, 12, 1206. https://doi.org/10.3390/rs12071206), hereinafter “Shen”, in view of Ge et al. (A primer on soil analysis using visible and near-infrared (vis-NIR) and mid-infrared (MIR) spectroscopy, Food and Agriculture Organization of the United Nations, 28 pages, March 2022), hereinafter “Ge”, Alexakis et al. (Integrated Use of Satellite Remote Sensing, Artificial Neural Networks, Field Spectroscopy, and GIS in Estimating Crucial Soil Parameters in Terms of Soil Erosion. Remote Sens. 9 May 2019, 11, 1106. https://doi.org/10.3390/rs11091106), hereinafter “Alexakis”, and Fernandes et al. (Estimation of soil organic matter content by modeling with artificial neural networks, Geoderma, Volume 350, 15 September 2019, Pages 46-51, ISSN 0016-7061, https://doi.org/10.1016/j.geoderma.2019.04.044.), hereinafter “Fernandes”. Shen teaches spectral analysis of undisturbed soil samples. Ge teaches in depth about spectral analysis. Alexakis teaches a neural network that uses spectral input to determine soil organic content. And Fernandes teaches that climate related variable effect soil organic content and measurements.

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.

The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claims 1-14 and 16-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Specifically, Claim 1 recites:
A method comprising:
receiving a value for a spectral band identified from a soil sample;
inputting the value for the spectral band to a neural network system that predicts a percentage of organic matter in the soil sample using values of spectral bands; and
improving the ability of the neural network system to correctly predict the percentage of organic matter in the soil sample by further inputting an ambient humidity level to the neural network system wherein the humidity level indicates an amount of ambient moisture present when the value for the spectral band was identified and the neural network system uses the ambient humidity level when predicting the percentage of organic matter.
The claim limitations in the abstract idea have been highlighted in bold above; the remaining limitations are “additional elements”.
Under the Step 1 of the eligibility analysis, we determine whether the claims are to a statutory category by considering whether the claimed subject matter falls within the four statutory categories of patentable subject matter identified by 35 U.S.C. 101: Process, machine, manufacture, or composition of matter. The above claim is considered to be in a statutory category (process).
Under the Step 2A, Prong One, we consider whether the claim recites a judicial exception (abstract idea). In the above claim, the highlighted portion constitutes an abstract idea because, under a broadest reasonable interpretation, it recites limitations that fall into/recite an abstract idea exceptions. Specifically, under the 2019 Revised Patent Subject matter Eligibility Guidance, it falls into the grouping of subject matter when recited as such in a claim limitation, that covers mental processes – concepts performed in the human mind including an observation, evaluation, judgement, and/or opinion.
For example, steps of “inputting the value for the spectral band to a neural network system that predicts a percentage of organic matter in the soil sample using values of spectral bands (providing data); and
improving the ability of the neural network system to correctly predict the percentage of organic matter in the soil sample by further inputting an ambient humidity level to the neural network system wherein the humidity level indicates an amount of ambient moisture present when the value for the spectral band was identified and the neural network system uses the ambient humidity level when predicting the percentage of organic matter (providing data)” are treated by the Examiner as belonging to mental process grouping.
Next, under the Step 2A, Prong Two, we consider whether the claim that recites a judicial exception is integrated into a practical application.
In this step, we evaluate whether the claim recites additional elements that integrate the exception into a practical application of that exception.
The above claims comprise the following additional elements:
Claim 1: receiving a value for a spectral band identified from a soil sample; and a neural network.
The additional element receiving a value for a spectral band identified from a soil sample represents a mere data gathering step and only adds an insignificant extra-solution activity to the judicial exception. Inputting the ambient humidity to the already trained neural network system is considered using a neural network in its ordinary capacity is considered mere instructions to apply an exception (see MPEP 2106.05(f)(2)).
Claim 10 is rejected as it merely describes a generic neural network, essentially claiming inputs to obtain an output. The names of the inputs and output on generally link the neural network to its field of use.
Claim 17 is rejected as the claim recites merely data gathering steps, which are also considered mental steps (recording and sharing data), and insignificant extra-solution activity (data gathering). The processing unit (generic processor) is generally recited and not qualified as a particular machine. And inputting the desired inputs to the already trained neural network system is considered using a neural network in its ordinary capacity is considered mere instructions to apply an exception (see MPEP 2106.05(f)(2)).
In conclusion, the above additional elements, considered individually and in combination with the other claim elements do not reflect an improvement to other technology or technical field, and, therefore, do not integrate the judicial exception into a practical application. Therefore, the claims are directed to a judicial exception and require further analysis under the Step 2B.
However, the above claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception (Step 2B analysis).
The claims, therefore, are not patent eligible.
With regards to the dependent claims, claims 2-9, 11-14, 16, and 18-20 provide additional features/steps which are part of an expanded algorithm, so these limitations should be considered part of an expanded abstract idea of the independent claims.

Claim Rejections - 35 USC § 103
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 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.

The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claim(s) 1-7 and 9 is/are rejected under 35 U.S.C. 103 as being unpatentable over Ge et al. (A primer on soil analysis using visible and near-infrared (vis-NIR) and mid-infrared (MIR) spectroscopy, Food and Agriculture Organization of the United Nations, 28 pages, March 2022), hereinafter “Ge”, in view of Alexakis et al. (Integrated Use of Satellite Remote Sensing, Artificial Neural Networks, Field Spectroscopy, and GIS in Estimating Crucial Soil Parameters in Terms of Soil Erosion. Remote Sens. 9 May 2019, 11, 1106. https://doi.org/10.3390/rs11091106), hereinafter “Alexakis”, and Fernandes et al. (Estimation of soil organic matter content by modeling with artificial neural networks, Geoderma, Volume 350, 15 September 2019, Pages 46-51, ISSN 0016-7061, https://doi.org/10.1016/j.geoderma.2019.04.044.), hereinafter “Fernandes”.

Regarding Claim 1, Ge teaches a method comprising:
receiving a value for a spectral band identified from a soil sample (Ge p. 2, Section 2.4, para. 1, Soil is a complex mixture of a vast array of chemical constituents, and has different physical states in terms of particle sizes, aggregation, surface roughness, and water content. Some of these physical and chemical constituents interact with the visNIR and MIR energy and produce absorption features in the spectra, which is essentially the foundation of soil spectroscopy with vis-NIR and MIR. These “primary” soil properties; including organic matter (or organic carbon), carbonate (or inorganic carbon), total nitrogen, clay minerals, iron content, particle size fractions of clay, silt and sand, and water content, can usually be calibrated from soil vis-NIR and MIR spectra, because absorption bands in the spectra correspond to these soil mineral and organic compositions.); and
inputting the value for the spectral band to a system that predicts a percentage of organic matter in the soil sample using values of spectral bands (Ge p. 5, Section 3.2, para. 1, To obtain the diffuse reflectance of soil samples in vis-NIR, a certified, standard panel with over 99 percent reflectance at all wavelengths is used (effectively considered a perfect, 100 percent diffuse reflector). This process is called white referencing. In MIR, a rough metal surface (such as aluminum) is good enough as a reflectance standard. The spectrometer registers three measurements, DNWhite as the raw spectrum for the reflectance standard (DN stands for digital number), DNDark as the raw spectrum for the dark current, and DNSoil as the raw spectrum for the soil sample. The soil reflectance is calculated as = (DNSoil – DNDark)/ (DNWhite − DNDark ; and this conversion is usually done automatically by the instrument.).
Ge does not explicitly teach a neural network; and
further inputting an ambient humidity level to the neural network system wherein the humidity level indicates an amount of ambient moisture present when the value for the spectral band was identified and the neural network system uses the ambient humidity level when predicting the percentage of organic matter.
Alexakis teaches a neural network to estimate soil organic matter using spectroscopy data (Alexakis p. 9, para. 1, The specific ANN architecture used, as input parameters, the longitude and the latitude [coordinate system: Universal Transverse Mercator (UTM) Zone 35N World Geodetic System (WGS) 1984] coordinates of each soil sample location as well as data from the spectral bands of Landsat 8 for the first ANN and Sentinel-2 for the second. Also see Fig. 4); and
Fernandes teaches inputting an ambient humidity level to the neural network system wherein the humidity level indicates an amount of ambient moisture present when the value for the spectral band was identified and the neural network system uses the ambient humidity level when predicting the percentage of organic matter (Fernandes p. 50, Col. 1: para 3, Li et al. (2013) proposed to estimate SOM content based on 11 climatic and land-use variables in China, using modeling by ANNs and multiple linear regression (MLR). At the end of the study, the authors observed that the ANNs were more precise than MLR in the prediction and that the 11 input variables showed good accuracy in SOM estimation. The high accuracy of estimation obtained by these authors is due to the dependence between climatic and land-use factors such as rainfall, temperature, humidity and radiation and factors of soil, such as terrain slope and content of SOM, directly acting on its decomposition (Balogh et al., 2011).).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Ge in view of Alexakis and Fernandes to explicitly teach inputting an ambient humidity level to the neural network system wherein the humidity level indicates an amount of ambient moisture present when the value for the spectral band was identified and the neural network system uses the ambient humidity level when predicting the percentage of organic matter, because a neural network can be used to determine soil organic material content from spectroscopy data (as shown by Alexakis above), and variables such as temperature and humidity improve the accuracy of ANNs estimating SOM (as shown by Fernandes above). Additionally Ge recognizes how temperature and humidity effect the spectrum of the measured sample (Ge p. 5, Section 3.2, para. 1, White referencing and dark current measurements should be made periodically during a session to ensure the instrument is well calibrated. In the lab, where environmental factors such as temperature and humidity are stable over time, instrument calibration can be done every 15 minutes. However, in the most rigorous and demanding applications, this calibration step is implemented between every sample).

Regarding Claim 2, Ge in view of Alexakis and Fernandes (as stated above) further teaches wherein the system further predicts a moisture content of the soil sample using the values of the spectral bands (Ge p. 2, Section 2.4, para. 1, Soil is a complex mixture of a vast array of chemical constituents, and has different physical states in terms of particle sizes, aggregation, surface roughness, and water content. Some of these physical and chemical constituents interact with the visNIR and MIR energy and produce absorption features in the spectra, which is essentially the foundation of soil spectroscopy with vis-NIR and MIR. These “primary” soil properties; including organic matter (or organic carbon), carbonate (or inorganic carbon), total nitrogen, clay minerals, iron content, particle size fractions of clay, silt and sand, and water content, can usually be calibrated from soil vis-NIR and MIR spectra, because absorption bands in the spectra correspond to these soil mineral and organic compositions.) and the method further comprises improving the ability of the system to correctly predict the moisture content of the soil sample by inputting the humidity level to the system (Fernandes p. 50, Col. 1: para 3, Li et al. (2013) proposed to estimate SOM content based on 11 climatic and land-use variables in China, using modeling by ANNs and multiple linear regression (MLR). At the end of the study, the authors observed that the ANNs were more precise than MLR in the prediction and that the 11 input variables showed good accuracy in SOM estimation. The high accuracy of estimation obtained by these authors is due to the dependence between climatic and land-use factors such as rainfall, temperature, humidity and radiation and factors of soil, such as terrain slope and content of SOM, directly acting on its decomposition (Balogh et al., 2011).).

Regarding Claim 3, Ge in view of Alexakis and Fernandes (as stated above) does not explicitly further teach improving the ability of the system to correctly predict the percentage of organic matter in the soil sample by further inputting a temperature value to the system wherein the temperature value is an ambient temperature when the value of the spectral band was identified.
Fernandes teaches improving the ability of the system to correctly predict the percentage of organic matter in the soil sample by further inputting a temperature value to the system wherein the temperature value is an ambient temperature when the value of the spectral band was identified (Fernandes p. 50, Col. 1: para 3, Li et al. (2013) proposed to estimate SOM content based on 11 climatic and land-use variables in China, using modeling by ANNs and multiple linear regression (MLR). At the end of the study, the authors observed that the ANNs were more precise than MLR in the prediction and that the 11 input variables showed good accuracy in SOM estimation. The high accuracy of estimation obtained by these authors is due to the dependence between climatic and land-use factors such as rainfall, temperature, humidity and radiation and factors of soil, such as terrain slope and content of SOM, directly acting on its decomposition (Balogh et al., 2011).).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Ge in view of Alexakis and Fernandes (as stated above), further in view of Fernandes to explicitly teach improving the ability of the system to correctly predict the percentage of organic matter in the soil sample by further inputting a temperature value to the system wherein the temperature value is an ambient temperature when the value of the spectral band was identified, because temperature is also known to affect soil organic material content. Additionally Ge recognizes how temperature and humidity effect the spectrum of the measured sample (Ge p. 5, Section 3.2, para. 1, White referencing and dark current measurements should be made periodically during a session to ensure the instrument is well calibrated. In the lab, where environmental factors such as temperature and humidity are stable over time, instrument calibration can be done every 15 minutes. However, in the most rigorous and demanding applications, this calibration step is implemented between every sample).

Regarding Claim 4, Ge in view of Alexakis and Fernandes (as stated above) does not explicitly teach improving the ability of the system to correctly predict the percentage of organic matter in the soil sample by further inputting a geographic area to the system wherein the geographic area describes a location where the soil sample was taken from.
Alexakis further teaches improving the ability of the system to correctly predict the percentage of organic matter in the soil sample by further inputting a geographic area to the system wherein the geographic area describes a location where the soil sample was taken from (Alexakis p. 9, para. 1, The specific ANN architecture used, as input parameters, the longitude and the latitude [coordinate system: Universal Transverse Mercator (UTM) Zone 35N World Geodetic System (WGS) 1984] coordinates of each soil sample location as well as data from the spectral bands of Landsat 8 for the first ANN and Sentinel-2 for the second. Also see Fig. 4).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Ge in view of Alexakis and Fernandes (as stated above), further in view of Alexakis to explicitly teach improving the ability of the system to correctly predict the percentage of organic matter in the soil sample by further inputting a geographic area to the system wherein the geographic area describes a location where the soil sample was taken from, to correlate each soil sample and its spectra to its location.

Regarding Claim 5, Ge in view of Alexakis and Fernandes (as stated above) does not explicitly teach improving the ability of the system to correctly predict the percentage of organic matter in the soil sample by further inputting a geographic area to the system wherein the geographic area describes a location where the soil sample was taken from.
Fernandes teaches improving the ability of the system to correctly predict the percentage of organic matter in the soil sample by further inputting a geographic area to the system wherein the geographic area describes a location where the soil sample was taken from (Fernandes p. 50, Col. 1: para 3, Li et al. (2013) proposed to estimate SOM content based on 11 climatic and land-use variables in China, using modeling by ANNs and multiple linear regression (MLR). At the end of the study, the authors observed that the ANNs were more precise than MLR in the prediction and that the 11 input variables showed good accuracy in SOM estimation. The high accuracy of estimation obtained by these authors is due to the dependence between climatic and land-use factors such as rainfall, temperature, humidity and radiation and factors of soil, such as terrain slope and content of SOM, directly acting on its decomposition (Balogh et al., 2011).).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Ge in view of Alexakis and Fernandes (as stated above), further in view of Fernandes to explicitly teach improving the ability of the system to correctly predict the percentage of organic matter in the soil sample by further inputting a temperature value to the system wherein the temperature value is an ambient temperature when the value of the spectral band was identified, because temperature is also known to affect soil organic material content. Additionally Ge recognizes how temperature and humidity effect the spectrum of the measured sample (Ge p. 5, Section 3.2, para. 1, White referencing and dark current measurements should be made periodically during a session to ensure the instrument is well calibrated. In the lab, where environmental factors such as temperature and humidity are stable over time, instrument calibration can be done every 15 minutes. However, in the most rigorous and demanding applications, this calibration step is implemented between every sample).

Regarding Claim 6, Ge in view of Alexakis and Fernandes (as stated above) does not explicitly teach improving the ability of the system to correctly predict the moisture content and organic matter of the soil sample by further inputting a geographic area to the system wherein the geographic area describes a location where the soil sample was taken from .
Alexakis further teaches improving the ability of the system to correctly predict the moisture content and organic matter of the soil sample by further inputting a geographic area to the system wherein the geographic area describes a location where the soil sample was taken from (Alexakis p. 9, para. 1, The specific ANN architecture used, as input parameters, the longitude and the latitude [coordinate system: Universal Transverse Mercator (UTM) Zone 35N World Geodetic System (WGS) 1984] coordinates of each soil sample location as well as data from the spectral bands of Landsat 8 for the first ANN and Sentinel-2 for the second. Also see Fig. 4).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Ge in view of Alexakis and Fernandes (as stated above), further in view of Alexakis to explicitly teach improving the ability of the system to correctly predict the percentage of organic matter in the soil sample by further inputting a geographic area to the system wherein the geographic area describes a location where the soil sample was taken from, to correlate each soil sample and its spectra to its location.

Regarding Claim 7, Ge in view of Alexakis and Fernandes (as stated above) further teaches inputting a plurality of respective values of different spectral bands to the system (Ge p. 2, section 2.4, para. 1, These fundamental vibrational absorption bands are usually strong and well-defined; a reason why MIR models are superior to vis-NIR models in predicting soil properties such as organic matter and clay. The overtones and combinations of these fundamental bands appear in the NIR region. Scanning inherently produces multiple spectral bands to receive and distinguish results).

Regarding Claim 9, Ge in view of Alexakis and Fernandes (as stated above) does not explicitly teach wherein the soil sample is unprocessed before the value for the spectral band is identified from the soil sample.
However, Ge teaches that there are instruments designed to perform measurements on undisturbed soil (Ge p. 5, section 3.2, paragraph 3, The contact probe is portable, more flexible than the muglight, and used more often to scan samples in a bag, or soil in natural states).
And Alexakis teaches field spectral measurements (Alexakis p. 18, para. 1, In this context, efficient spatial simulations of the crucial soil parameters, SOM and CaCO3, were carried out using field spectral VIS-NIR measurements and satellite remote sensing observations (Sentinel-2 and Landsat 8) combined with non-linear (ANNs) approaches. Also see Fig. 4, Field Spectroscopy)
Therefore, it would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Ge in view of Alexakis and Fernandes (as stated above), further in view of Ge and Alexakis to explicitly teach wherein the soil sample is unprocessed before the value for the spectral band is identified from the soil sample, by using a contact probe on soil in a natural state and ensuring the instrument is calibrated to its measuring environmental factors, such as temperature and humidity (as stated in Ge), or in cases where satellite/remote sensing observations are taken (as stated in Alexakis).

Claim(s) 8 is/are rejected under 35 U.S.C. 103 as being unpatentable over Ge in view of Alexakis and Fernandes (as stated above), further in view of Chen et al. (Convolutional neural network model for soil moisture prediction and its transferability analysis based on laboratory Vis-NIR spectral data,International Journal of Applied Earth Observation and Geoinformation, Volume 104, 2021, 102550, ISSN 1569-8432, https://doi.org/10.1016/j.jag.2021.102550.), hereinafter “Chen”.

Regarding Claim 8, Ge in view of Alexakis and Fernandes (as stated above) further teaches wherein the system comprises a (Ge p. 9, section 3.5, para. 3, More recently, a large family of so-called “machine learning” methods have become widespread in soil vis-NIR and MIR spectroscopy. Examples of these methods are Random Forest (RF), Support Vector Regression, Cubist, Artificial Neural Networks (ANN), etc. (Minasny and McBratney, 2008; Viscarra Rossel et al., 2016). Models trained on these machine learning methods usually show higher predictive accuracy than PLSR and PCR, especially when the number of training samples is large. One rationale for their superior performance is that they can more effectively model the non-linear relationships between spectral data and soil properties.).
Ge in view of Alexakis and Fernandes (as stated above) is not relied upon to explicitly teach a convolution neural network model.
Chen teaches a convolution neural network model (Chen p. 7, section 5, CNN models have advantages for mining the relationship between spectral data and soil properties especially using a large number of samples.)
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Ge in view of Alexakis and Fernandes (as stated above), further in view of Chen to explicitly teach a convolution neural network model, because Ge discloses to the use of neural networks and Chen discloses the specific use of convolutional neural networks.

Claim(s) 10-14 and 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Wikipedia ("Convolutional neural network." Wikipedia, The Free Encyclopedia. Wikipedia, The Free Encyclopedia, 24 May. 2022.), hereinafter “WikiCNN”, in view of Ge et al. (A primer on soil analysis using visible and near-infrared (vis-NIR) and mid-infrared (MIR) spectroscopy, Food and Agriculture Organization of the United Nations, 28 pages, March 2022), hereinafter “Ge”.

Regarding Claim 10, WikiCNN teaches a computing device comprising:
a memory containing parameters representing a neural network (An artificial neural network is a computational machine learning model requiring a processor and memory);
a processor using the parameters of the neural network stored in the memory to form the neural network (An artificial neural network is a computational machine learning model requiring a processor and memory) such that the neural network comprises:
at least one input (WikiCNN p. 1, Architecture: convolutional neural network consists of an input layer, hidden layers and an output layer.); and
at least one other input (WikiCNN p. 1, Architecture: convolutional neural network consists of an input layer, hidden layers and an output layer.); and
a final stage receiving both the inputs output based on the inputs (WikiCNN p. 1, Architecture: convolutional neural network consists of an input layer, hidden layers and an output layer.).
WikiCNN does not explicitly teach the specific inputs of at least one spectral stage receiving spectral band values determined from a soil sample and providing spectral stage outputs based on the spectral band values; and
at least one ambient stage receiving an ambient humidity value representing ambient moisture present when the spectral band values were determined,
providing ambient stage outputs based on the ambient humidity value (WikiCNN p. 1, Architecture: convolutional neural network consists of an input layer, hidden layers and an output layer. This limitation is simply outputting the input with no processing); and
the output of organic matter percentage for the soil sample based on the spectral stage outputs and the ambient stage outputs.
However, WikiCNN teaches the well-known purpose of convolutional neural networks for image recognition and processing (WikiCNN p. 1, Definition). Claim 1 as written, merely claims a neural network with specific inputs for a desired output. It is well known that a neural network is trained on chosen inputs processed through the neural networks layers, to produce a desired output. The specific naming of the stages/variables does not change the function of the neural network itself.
Additionally Ge teaches spectroscopy is used to determine organic content of soil (Ge p. 2, Section 2.4, para. 1, Soil is a complex mixture of a vast array of chemical constituents, and has different physical states in terms of particle sizes, aggregation, surface roughness, and water content. Some of these physical and chemical constituents interact with the visNIR and MIR energy and produce absorption features in the spectra, which is essentially the foundation of soil spectroscopy with vis-NIR and MIR. These “primary” soil properties; including organic matter (or organic carbon), carbonate (or inorganic carbon), total nitrogen, clay minerals, iron content, particle size fractions of clay, silt and sand, and water content, can usually be calibrated from soil vis-NIR and MIR spectra, because absorption bands in the spectra correspond to these soil mineral and organic compositions.), and that environmental factors such as temperature and humidity effect the measured value of organic matter content (Ge p. 5, Section 3.2, para. 1, White referencing and dark current measurements should be made periodically during a session to ensure the instrument is well calibrated. In the lab, where environmental factors such as temperature and humidity are stable over time, instrument calibration can be done every 15 minutes. However, in the most rigorous and demanding applications, this calibration step is implemented between every sample).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application to modify WikiCNN in view of Ge to explicitly teach specific inputs of spectral band values determined from a soil sample and an ambient humidity value representing ambient moisture present when the spectral band values were determined, and the output of an organic matter percentage for a soil sample, because the point of the Neural network is to learn to recognize features of and between the inputs to arrive at the desired conclusion/output, and it is known that factors such as temperature and humidity effect the value of the organic matter content.

Regarding Claim 11, WikiCNN in view of Ge (as stated above) further teaches wherein the input layer further receives ambient temperature values representing a temperature present when the spectral bands were determined (Ge p. 5, Section 3.2, para. 1, White referencing and dark current measurements should be made periodically during a session to ensure the instrument is well calibrated. In the lab, where environmental factors such as temperature and humidity are stable over time, instrument calibration can be done every 15 minutes. However, in the most rigorous and demanding applications, this calibration step is implemented between every sample. Since temperature effects measured organic matter content, it would also be obvious to use as an input.).

Regarding Claim 12, WikiCNN in view of Ge (as stated above) does not explicitly teach wherein the input layer further receives a geographic area where the soil sample was obtained from.
However, Ge teaches building spectral libraries (Ge p. 14, Chapter 4.) and using modeling schemes that account for local/similar soil profiles (Ge p. 14, Chapter 4, para. 2, when it comes to applications, the need to estimate soil properties often occurs at the local scale (for example, in precision agriculture to estimate soil fertility and its in-field variation). To use the spectral libraries more efficiently, there is a need to choose and optimize the subset of samples for model calibration. To address this problem, more advanced machine learning methods (for example, spectral similarity search and memory-based learning) and modeling schemes (for example, spiking with extra weight) are developed to improve the performance of prediction to soil properties using spectra that are spectrally similar, instead of using all spectral data in a library. It is also usual to test whether or not, the new sample one wishes to predict, effectively falls within the space covered by the spectra in the library.).
Therefore, it would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify WikiCNN in view of Ge (as stated above) to explicitly teach wherein the input layer further receives a geographic area where the soil sample was obtained from, because a simple identifier such as the geographic region where a soil sample was taken allows for more accurate soil property evaluation by the neural network.

Regarding Claim 13, WikiCNN in view of Ge (as stated above) does not explicitly teach wherein the output layer further indicates a moisture content of the soil sample.
However WikiCNN clearly shows that the output layer can comprise multiple outputs (WikiCNN p. 1, Fig. 1) and Ge discloses that water content can also be obtained through measuring spectra of soil (Ge p. 2, section 2.4, These “primary” soil properties; including organic matter (or organic carbon), carbonate (or inorganic carbon), total nitrogen, clay minerals, iron content, particle size fractions of clay, silt and sand, and water content, can usually be calibrated from soil vis-NIR and MIR spectra, because absorption bands in the spectra correspond to these soil mineral and organic compositions.).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify WikiCNN in view of Ge (as stated above) to explicitly teach wherein the output layer further indicates a moisture content of the soil sample, because it is known that soil moisture content can also be determined from spectra measurements, just as organic content, and that a neural network can determine both outputs with its different interpretations and processing of the inputs.

Regarding Claim 14, WikiCNN in view of Ge (as stated above) further teaches the neural network comprises a convolution neural network (WikiCNN p. 1, Definition, Convolutional neural networks are a specialized type of artificial neural networks that use a mathematical operation called convolution in place of general matrix multiplication in at least one of their layers.[13] They are specifically designed to process pixel data and are used in image recognition and processing.. WikiCNN is specifically about the structure and functions of convolutional neural networks).

Regarding Claim 16, WikiCNN in view of Ge (as stated above) does not explicitly teach wherein the spectral band values determined from the soil sample are determined without processing the soil sample before determining the spectral values..
However, Ge teaches that there are instruments designed to perform measurements on undisturbed soil (Ge p. 5, section 3.2, paragraph 3, The contact probe is portable, more flexible than the muglight, and used more often to scan samples in a bag, or soil in natural states).
Therefore, it would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify WikiCNN in view of Ge (as stated above), further in view of Ge to explicitly teach wherein the soil sample is unprocessed before the value for the spectral band is identified from the soil sample, by using a contact probe on soil in a natural state and ensuring the instrument is calibrated to its measuring environmental factors, such as temperature and humidity (as stated above for claim 1).

Claim(s) 17-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Shen et al. (Hyperspectral Estimation of Soil Organic Matter Content using Different Spectral Preprocessing Techniques and PLSR Method. Remote Sens. 2020, 12, 1206. https://doi.org/10.3390/rs12071206), hereinafter “Shen”, in view of Ge et al. (A primer on soil analysis using visible and near-infrared (vis-NIR) and mid-infrared (MIR) spectroscopy, Food and Agriculture Organization of the United Nations, 28 pages, March 2022), hereinafter “Ge”, Alexakis et al. (Integrated Use of Satellite Remote Sensing, Artificial Neural Networks, Field Spectroscopy, and GIS in Estimating Crucial Soil Parameters in Terms of Soil Erosion. Remote Sens. 9 May 2019, 11, 1106. https://doi.org/10.3390/rs11091106), hereinafter “Alexakis”, and Fernandes et al. (Estimation of soil organic matter content by modeling with artificial neural networks, Geoderma, Volume 350, 15 September 2019, Pages 46-51, ISSN 0016-7061, https://doi.org/10.1016/j.geoderma.2019.04.044.), hereinafter “Fernandes”.

Regarding Claim 17, Shen teaches a method comprising:
placing an unprocessed soil sample in a spectroradiometer (Shen p. 4, section 2.2, para. 1, Analytical Spectral Devices (ASD) FieldSpec 4 High-Res spectroradiometer was used to perform the indoor spectral measurements on the undisturbed soil samples, which were stored in aluminum boxes and were not pretreated (Figure 2).);
recording values for spectral bands generated by the spectroradiometer from the unprocessed soil sample (Shen p. 4, section 2.2, para. 1, Each soil sample was measured four times—the aluminum box was rotated by 90° after each measurement for a total of three rotations. A total of 10 spectral curves were collected automatically during each measurement and the arithmetic average of the curves was used as the spectral data. Standard whiteboard calibration was performed before each measurement.).
Shen does not explicitly teach recording a humidity level for air proximate the spectroradiometer; and
providing the recorded values for the spectral bands and the recorded value for the humidity level to a processing unit;
the processing unit executing a neural network that uses both the values for the spectral bands and the humidity level to provide a percentage of organic material in the unprocessed soil sample.
However Shen teaches standard whiteboard calibration was performed before each measurement (Shen p. 4, section 2.2, para. 1).
And Ge teaches that the measuring instrument should be calibrated regularly since environment factors effect output (Ge p. 5, Section 3.2, para. 1, White referencing and dark current measurements should be made periodically during a session to ensure the instrument is well calibrated. In the lab, where environmental factors such as temperature and humidity are stable over time, instrument calibration can be done every 15 minutes. However, in the most rigorous and demanding applications, this calibration step is implemented between every sample).
Alexakis teaches executing a neural network using the values for the spectral bands to provide a percentage of organic material in the unprocessed soil sample (Alexakis p. 6, section 3.1, spectral signatures were collected (using field spectroscopy) from all the soil samples, also see p. 9, para. 1, The estimated output parameters of the model were the SOM, the CaCO3, and the K-factor, as determined from soil analysis, See Fig. 4).
And Fernandes teaches a neural network that uses the humidity level to provide a percentage of organic material (Fernandes p. 50, Col. 1: para 3, Li et al. (2013) proposed to estimate SOM content based on 11 climatic and land-use variables in China, using modeling by ANNs and multiple linear regression (MLR). At the end of the study, the authors observed that the ANNs were more precise than MLR in the prediction and that the 11 input variables showed good accuracy in SOM estimation. The high accuracy of estimation obtained by these authors is due to the dependence between climatic and land-use factors such as rainfall, temperature, humidity and radiation and factors of soil, such as terrain slope and content of SOM, directly acting on its decomposition (Balogh et al., 2011).).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Shen in view of Ge, Alexakis, and Fernandes to explicitly teach recording a humidity level for air proximate the spectroradiometer; and
providing the recorded values for the spectral bands and the recorded value for the humidity level to a processing unit to obtain a percentage of organic material in the soil sample, because a neural network is a known technique for analyzing spectra of a soil sample to determine organic material content (as taught by Alexakis), and environmental factors such as temperature and humidity are known to effect spectral measurements and organic material content and should be considered so that the neural network can provide more accurate results (as taught by Fernandes), especially in cases when the measurement system is not calibrated for the environmental factors (as taught by Shen and Ge).

Regarding Claim 18, Shen in view of Ge, Alexakis, and Fernandes (as stated above) is not relied upon to further teach recording a temperature value for air proximate the spectroradiometer; and providing the temperature value to the processing unit when providing the recorded values for the spectral bands and the recorded value for the humidity level to the processing unit.
Fernandes teaches recording a temperature value for air proximate the spectroradiometer; and providing the temperature value to the processing unit when providing the recorded values for the spectral bands and the recorded value for the humidity level to the processing unit (Fernandes p. 50, Col. 1: para 3, Li et al. (2013) proposed to estimate SOM content based on 11 climatic and land-use variables in China, using modeling by ANNs and multiple linear regression (MLR). At the end of the study, the authors observed that the ANNs were more precise than MLR in the prediction and that the 11 input variables showed good accuracy in SOM estimation. The high accuracy of estimation obtained by these authors is due to the dependence between climatic and land-use factors such as rainfall, temperature, humidity and radiation and factors of soil, such as terrain slope and content of SOM, directly acting on its decomposition (Balogh et al., 2011).).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Shen in view of Ge, Alexakis, and Fernandes to explicitly teach recording a temperature value for air proximate the spectroradiometer; and providing the temperature value to the processing unit when providing the recorded values for the spectral bands and the recorded value for the humidity level to the processing unit, because environmental factors such as temperature and humidity are known to effect spectral measurements and organic material content and should be considered so that the neural network can provide more accurate results (as taught by Fernandes), especially in cases when the measurement system is not calibrated for the environmental factors (as taught by Shen and Ge).

Regarding Claim 19, Shen in view of Ge, Alexakis, and Fernandes (as stated above) does not explicitly teach providing an area where the soil sample was obtained from to the processing unit when providing the recorded values for the spectral bands and the recorded value for the humidity level to the processing unit.
Alexakis teaches providing an area where the soil sample was obtained from to the processing unit when providing the recorded values for the spectral bands and the recorded value for the humidity level to the processing unit (Alexakis p. 9, para. 1, The specific ANN architecture used, as input parameters, the longitude and the latitude [coordinate system: Universal Transverse Mercator (UTM) Zone 35N World Geodetic System (WGS) 1984] coordinates of each soil sample location as well as data from the spectral bands of Landsat 8 for the first ANN and Sentinel-2 for the second. Also see Fig. 4).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Ge in view of Alexakis and Fernandes (as stated above), further in view of Alexakis to explicitly teach providing an area where the soil sample was obtained from to the processing unit when providing the recorded values for the spectral bands and the recorded value for the humidity level to the processing unit, to correlate each soil sample and its spectra to its location.

Regarding Claim 20, although Shen in view of Ge, Alexakis, and Fernandes (as stated above) further teaches that spectral information is related to both soil organic matter content and soil moisture (Shen p. 2, section 1, Soil spectral information is not only related to the characteristics of the chemical components of the soil, such as its SOM content, iron oxide content, and soil moisture but also to the physical properties of the soil such as its particle size, density, and surface roughness [9]), Shen in view of Ge (as stated above) is not relied upon to teach providing the recorded values for the spectral bands and the recorded value for the humidity level to the processing unit to obtain a moisture content in the soil sample.
Ge teaches providing the recorded values for the spectral bands and the recorded value for the humidity level to the processing unit to obtain a moisture content in the soil sample (Ge p. 2, Section 2.4, para. 1, Soil is a complex mixture of a vast array of chemical constituents, and has different physical states in terms of particle sizes, aggregation, surface roughness, and water content. Some of these physical and chemical constituents interact with the visNIR and MIR energy and produce absorption features in the spectra, which is essentially the foundation of soil spectroscopy with vis-NIR and MIR. These “primary” soil properties; including organic matter (or organic carbon), carbonate (or inorganic carbon), total nitrogen, clay minerals, iron content, particle size fractions of clay, silt and sand, and water content, can usually be calibrated from soil vis-NIR and MIR spectra, because absorption bands in the spectra correspond to these soil mineral and organic compositions.).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Shen in view of Ge, Alexakis, and Fernandes (as stated above), further in view of Ge, to explicitly teach providing the recorded values for the spectral bands and the recorded value for the humidity level to the processing unit to obtain a moisture content in the soil sample, because it is well known that spectral information of a soil sample is used to determine soil moisture/water content, along with organic matter content.

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 CHRISTIAN T BRYANT whose telephone number is (571)272-4194. The examiner can normally be reached Monday-Thursday and Alternate Fridays 7:00-4:30.
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, LISA CAPUTO can be reached on (571) 272-2388. 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.

/CHRISTIAN T BRYANT/Examiner, Art Unit 2863 04/03/2025

/LISA M CAPUTO/Supervisory Patent Examiner, Art Unit 2863