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 .
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.
This action is in response to the application filed on 12/06/2021. Claims 1-20 are pending in the application and have been considered below.
Specification
The title of the invention is not descriptive. A new title is required that is clearly indicative of the invention to which the claims are directed.
Claim Objections
Claim 14 is objected to because of the following informalities: “output output data.” It should recite” output data.” Appropriate correction is required.
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.
Claims 1-2 and 19-20 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Gupta (“Selection of important features and predicting wine quality using machine learning techniques,” hereinafter referred to as Gupta).
As to claim 1, Gupta discloses a learning apparatus comprising a first processor (See 3.2. Neural network: computer includes a processor):
wherein the first processor is configured to
acquire learning input data to be input to a machine learning model for predicting a quality of a product, the learning input data including multi-dimensional physical-property relevance data which is derived from multi-dimensional physical-property data representing a physical property of the product and includes a plurality of items (See Introduction: predicted six geographic wine origins based on neural networks fed with 15 input variables (Examiner “fed with 15 input variables” as 15 input variables as input to the neural networks (machine learning model) for predicting six geographic wine origins);
2. Dataset – Wine dataset is a collection of white and red wines [11]. White wine consists of 4898 samples and red wine contains 1599 samples. Each sample of both types of wine consists of 12 physiochemical variables: fixed acidity, volatile acidity, citric acid, residual sugar, chlorides, free sulfur dioxide, total sulfur dioxide, density, pH, sulphates, alcohol, and quality rating; Examiner interprets the collected “Wine dataset” as the acquire learning input; Examiner further interprets the “12 physiochemical variables” as the derivation of multidimensional physical-property relevance data from multi-dimensional physical property data; Examiner interpretation of “multidimensional physical-property relevance data from multi-dimensional physical property data” is based on paragraphs [0005]-[0006] of the specification);
input the learning input data to the machine learning model, perform learning, and output a temporary machine learning model (See Abstract; usage of machine learning techniques such as linear regression, neural network and support vector machine for product quality in two ways; 3.2. Neural network Fig. 1 Artificial neural network, wherein Examiner interprets the input layer to receive the input data and the output layer to output result. Table 3 represents neural network regression analysis for Red Wine data. In this work, 11-5-1 neural network architecture is used. Table 3 shows the predicted values against original values of quality for Red Wine in both the cases: for all features and for selected features. Examiner interprets the “selected features” in Table 3 as the “input,” the “neural network regression” as the machine learning model, and the “Quality Output” as the “output a temporary machine learning model”);
extract a high-contribution item from the plurality of items of the multi-dimensional physical-property relevance data by using the temporary machine learning model, the high- contribution item being the item of which a contribution to improvement of accuracy of prediction of the quality satisfies a preset condition (abstract; pages. 308-309, 4.1. Determining important features for prediction Table 1 represents the regression summary for dependent variable i.e. quality for Red Wine. This table shows the dependency of quality on all predictors individually. Here, R is the co-relation coefficient which indicates how much dependent variable (quality) is co-related with all predictors as whole; 4.2. Predicting value of dependent variable (Quality) Two different machine learning techniques neural network and SVM have been used to predict the wine quality in this work. The wine quality is predicted using all features of dataset and for selected features (determined from previous subsection) of dataset.
Table 3 represents neural network regression analysis for Red Wine data. In this work, 11-5-1 neural network architecture is used. Table 3 shows the predicted values against original values of quality for Red Wine in both the cases: for all features and for selected features. It is not feasible to show results for all documents, therefore the results are shown for few documents randomly. This table clearly indicates that wine quality is predicted more accurately for selected features in comparison to using all features. The overall summary of neural network; Examiner interpretation of “high-contribution item” is based on paragraph [0020] of the specification); and
selectively input the multi-dimensional physical-property relevance data of the high-contribution item to the machine learning model, perform learning, and output the machine learning model as a learned model to be provided for actual operation (abstract; page. 307, I. 2-3; page 309. abstract; pages. 308-309, 4.1. Determining important features for prediction Table 1 represents the regression summary for dependent variable i.e. quality for Red Wine. This table shows the dependency of quality on all predictors individually. Here, R is the co-relation coefficient which indicates how much dependent variable (quality) is co-related with all predictors as whole; 4.2. Predicting value of dependent variable (Quality) Two different machine learning techniques neural network and SVM have been used to predict the wine quality in this work. The wine quality is predicted using all features of dataset and for selected features (determined from previous subsection) of dataset.
Table 3 represents neural network regression analysis for Red Wine data. In this work, 11-5-1 neural network architecture is used. Table 3 shows the predicted values against original values of quality for Red Wine in both the cases: for all features and for selected features. It is not feasible to show results for all documents, therefore the results are shown for few documents randomly. This table clearly indicates that wine quality is predicted more accurately for selected features in comparison to using all features.. This table clearly indicates that wine quality is predicted more accurately for selected features in comparison to using all features; Examiner interprets the “selected features” in Table 3 as the “selectively input the multi-dimensional physical-property relevance data,” the “neural network regression” as the machine learning model, and the “Quality Output” as the “output the machine learning model as a learned model.”)
As to claim 19, Gupta discloses an operation method of a learning apparatus, the operation method comprising:
acquiring learning input data to be input to a machine learning model for predicting a quality of a product, the learning input data including multi-dimensional physical-property relevance data which is derived from multi-dimensional physical-property data representing a physical property of the product and includes a plurality of items (See Introduction: predicted six geographic wine origins based on neural networks fed with 15 input variables (Examiner “fed with 15 input variables” as 15 input variables as input to the neural networks (machine learning model) for predicting six geographic wine origins);
2. Dataset – Wine dataset is a collection of white and red wines [11]. White wine consists of 4898 samples and red wine contains 1599 samples. Each sample of both types of wine consists of 12 physiochemical variables: fixed acidity, volatile acidity, citric acid, residual sugar, chlorides, free sulfur dioxide, total sulfur dioxide, density, pH, sulphates, alcohol, and quality rating; Examiner interprets the collected “Wine dataset” as the acquire learning input; Examiner further interprets the “12 physiochemical variables” as the derivation of multidimensional physical-property relevance data from multi-dimensional physical property data; Examiner interpretation of “multidimensional physical-property relevance data from multi-dimensional physical property data” is based on paragraphs [0005]-[0006] of the specification);
inputting the learning input data to the machine learning model, perform learning, and outputting a temporary machine learning model (See Abstract; usage of machine learning techniques such as linear regression, neural network and support vector machine for product quality in two ways; 3.1. Linear regression; 3.2. Neural network Fig. 1 Artificial neural network, wherein Examiner interprets the input layer to receive the input data and the output layer to output result. Table 3 represents neural network regression analysis for Red Wine data. In this work, 11-5-1 neural network architecture is used. Table 3 shows the predicted values against original values of quality for Red Wine in both the cases: for all features and for selected features. Examiner interprets the “selected features” in Table 3 as the “input,” the “neural network regression” as the machine learning model, and the “Quality Output” as the “output a temporary machine learning model”);
extracting a high-contribution item from the plurality of items of the multi-dimensional physical-property relevance data by using the temporary machine learning model, the high-contribution item being the item of which a contribution to improvement of accuracy of prediction of the quality satisfies a preset condition (abstract; pages. 308-309, 4.1. Determining important features for prediction Table 1 represents the regression summary for dependent variable i.e. quality for Red Wine. This table shows the dependency of quality on all predictors individually. Here, R is the co-relation coefficient which indicates how much dependent variable (quality) is co-related with all predictors as whole; 4.2. Predicting value of dependent variable (Quality) Two different machine learning techniques neural network and SVM have been used to predict the wine quality in this work. The wine quality is predicted using all features of dataset and for selected features (determined from previous subsection) of dataset.
Table 3 represents neural network regression analysis for Red Wine data. In this work, 11-5-1 neural network architecture is used. Table 3 shows the predicted values against original values of quality for Red Wine in both the cases: for all features and for selected features. It is not feasible to show results for all documents, therefore the results are shown for few documents randomly. This table clearly indicates that wine quality is predicted more accurately for selected features in comparison to using all features. The overall summary of neural network; Examiner interpretation of “high-contribution item” is based on paragraph [0020] of the specification); and
selectively inputting the multi-dimensional physical-property relevance data of the high-contribution item to the machine learning model, performing learning, and outputting the machine learning model as a learned model to be provided for actual operation (abstract; page. 307, I. 2-3; page 309. abstract; pages. 308-309, 4.1. Determining important features for prediction Table 1 represents the regression summary for dependent variable i.e. quality for Red Wine. This table shows the dependency of quality on all predictors individually. Here, R is the co-relation coefficient which indicates how much dependent variable (quality) is co-related with all predictors as whole; 4.2. Predicting value of dependent variable (Quality) Two different machine learning techniques neural network and SVM have been used to predict the wine quality in this work. The wine quality is predicted using all features of dataset and for selected features (determined from previous subsection) of dataset.
Table 3 represents neural network regression analysis for Red Wine data. In this work, 11-5-1 neural network architecture is used. Table 3 shows the predicted values against original values of quality for Red Wine in both the cases: for all features and for selected features. It is not feasible to show results for all documents, therefore the results are shown for few documents randomly. This table clearly indicates that wine quality is predicted more accurately for selected features in comparison to using all features.. This table clearly indicates that wine quality is predicted more accurately for selected features in comparison to using all features; Examiner interprets the “selected features” in Table 3 as the “selectively input the multi-dimensional physical-property relevance data,” the “neural network regression” as the machine learning model, and the “Quality Output” as the “output the machine learning model as a learned model.”)
As to claim 20, Gupta discloses a non-transitory computer readable recording medium storing an operation program of a learning apparatus, the program causing a computer to function as:
acquiring learning input data to be input to a machine learning model for predicting a quality of a product, the learning input data including multi-dimensional physical-property relevance data which is derived from multi-dimensional physical-property data representing a physical property of the product and includes a plurality of items (See 2. Dataset – Wine dataset is a collection of white and red wines [11]. White wine consists of 4898 samples and red wine contains 1599 samples. Each sample of both types of wine consists of 12 physiochemical variables: fixed acidity, volatile acidity, citric acid, residual sugar, chlorides, free sulfur dioxide, total sulfur dioxide, density, pH, sulphates, alcohol, and quality rating; ; Examiner interprets the collected “Wine dataset” as the acquire learning input; Examiner further interprets the “12 physiochemical variables” as the derivation of multidimensional physical-property relevance data from multi-dimensional physical property data; Examiner interpretation of “multidimensional physical-property relevance data from multi-dimensional physical property data” is based on paragraphs [0005]-[0006] of the specification);
inputting the learning input data to the machine learning model, performing learning, and outputting a temporary machine learning model (See Abstract; usage of machine learning techniques such as linear regression, neural network and support vector machine for product quality in two ways; 3.1. Linear regression; 3.2. Neural network Fig. 1 Artificial neural network, wherein Examiner interprets the input layer to receive the input data and the output layer to output result. Table 3 represents neural network regression analysis for Red Wine data. In this work, 11-5-1 neural network architecture is used. Table 3 shows the predicted values against original values of quality for Red Wine in both the cases: for all features and for selected features. Examiner interprets the “selected features” in Table 3 as the “input,” the “neural network regression” as the machine learning model, and the “Quality Output” as the “output a temporary machine learning model”
extracting a high-contribution item from the plurality of items of the multi-dimensional physical-property relevance data by using the temporary machine learning model, the high- contribution item being the item of which a contribution to improvement of accuracy of prediction of the quality satisfies a preset condition (abstract; pages. 308-309, 4.1. Determining important features for prediction Table 1 represents the regression summary for dependent variable i.e. quality for Red Wine. This table shows the dependency of quality on all predictors individually. Here, R is the co-relation coefficient which indicates how much dependent variable (quality) is co-related with all predictors as whole; 4.2. Predicting value of dependent variable (Quality) Two different machine learning techniques neural network and SVM have been used to predict the wine quality in this work. The wine quality is predicted using all features of dataset and for selected features (determined from previous subsection) of dataset.
Table 3 represents neural network regression analysis for Red Wine data. In this work, 11-5-1 neural network architecture is used. Table 3 shows the predicted values against original values of quality for Red Wine in both the cases: for all features and for selected features. It is not feasible to show results for all documents, therefore the results are shown for few documents randomly. This table clearly indicates that wine quality is predicted more accurately for selected features in comparison to using all features. The overall summary of neural network; Examiner interpretation of “high-contribution item” is based on paragraph [0020] of the specification); and
selectively inputting the multi-dimensional physical-property relevance data of the high-contribution item to the machine learning model, performing learning, and outputting the machine learning model as a learned model to be provided for actual operation (abstract; page. 307, I. 2-3; page 309. abstract; pages. 308-309, 4.1. Determining important features for prediction Table 1 represents the regression summary for dependent variable i.e. quality for Red Wine. This table shows the dependency of quality on all predictors individually. Here, R is the co-relation coefficient which indicates how much dependent variable (quality) is co-related with all predictors as whole; 4.2. Predicting value of dependent variable (Quality) Two different machine learning techniques neural network and SVM have been used to predict the wine quality in this work. The wine quality is predicted using all features of dataset and for selected features (determined from previous subsection) of dataset.
Table 3 represents neural network regression analysis for Red Wine data. In this work, 11-5-1 neural network architecture is used. Table 3 shows the predicted values against original values of quality for Red Wine in both the cases: for all features and for selected features. It is not feasible to show results for all documents, therefore the results are shown for few documents randomly. This table clearly indicates that wine quality is predicted more accurately for selected features in comparison to using all features. This table clearly indicates that wine quality is predicted more accurately for selected features in comparison to using all features; Examiner interprets the “selected features” in Table 3 as the “selectively input the multi-dimensional physical-property relevance data,” the “neural network regression” as the machine learning model, and the “Quality Output” as the “output the machine learning model as a learned model.”)
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 factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Gupta (“Selection of important features and predicting wine quality using machine learning techniques,” hereinafter referred to as Gupta), in view of Kondo et al. (US 2020/0230884 A1, hereinafter referred to as Kondo).
As to claim 2, which incorporates the rejection of claim1, Gupta fails to explicitly teach wherein the learning input data includes production condition data which is set in a production process of the productHowever, Kondo, in combination with Gupta, teaches wherein the learning input data includes production condition data which is set in a production process of the product (paragraphs [0010] and [0015]-[0017], production condition data and product quality data thus monitored in a database).
It would have been obvious to one of ordinary skill in the art, having the teachings of
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the system of Gupta to add production condition data to the system of Gupta, as taught by Kondo above. The modification would have been obvious because one of ordinary skill would be motivated to verify the correctness
of the quality deteriorating factor and the validity of the improvement contents, as suggested by Kondo ([0015]).
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable Gupta (“Selection of important features and predicting wine quality using machine learning techniques,” hereinafter referred to as Gupta), in view of Rizkallah et al. (US 2012/0280146 A1, hereinafter referred to as Rizkallah).
As to claim 3, which incorporates the rejection of claim 1, Gupta teaches spectral measurements (See Introduction) but fails to explicitly teach wherein the multi-dimensional physical-property data includes spectrum data which is detected by performing spectroscopic analysis on the product.
However, Rizkallah, in combination with Gupta, teaches wherein the multi-dimensional physical-property data includes spectrum data which is detected by performing spectroscopic analysis on the product (Abstract; paragraphs [0001] spectroscopic analysis of at least one sample, in particular of a food or drug, implementing a method for analyzing spectroscopic data based on a multi-way statistical model…More generally, the invention can be used in the determination of any quality indicator of a sample, and/or any parameter characterizing a method to which said sample has been subjected; [0005] Multi-way analysis is the natural extension of multivariate analysis when the data is arranged in three-way or more-than-three-way tables. It is based on the use of statistical models such as "PARAFAC" ("Parallel Factor") and NPLS ("N-ways Partial Least Squares regression"). These methods, as well as their use in the analysis of food products, are described in the document [Bro 1998], [0015]-[0020] and [0048]).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the system of Gupta to add spectroscopic analysis to the system of Gupta, as taught by Rizkallah above. The modification would have been obvious because one of ordinary skill would be motivated to provide spectroscopic analysis of the sample by analyzing the spectroscopic data based on the multi-channel statistical model in a simple and rapid manner using simple and less expensive equipment, as suggested by Rizkallah, ([0014]-[0015]).
Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Gupta (“Selection of important features and predicting wine quality using machine learning techniques,” hereinafter referred to as Gupta), in view of Kondo et al. (US 2020/0230884 A1, hereinafter referred to as Kondo), and further in view of Rizkallah et al. (US 2012/0280146 A1, hereinafter referred to as Rizkallah).
As to claim 4, which incorporates the rejection of claim 2, Gupta teaches spectral measurements (See Introduction) but Gupta and Kondo fail to explicitly teach wherein the multi-dimensional physical-property data includes spectrum data which is detected by performing spectroscopic analysis on the product.
However, Rizkallah, in combination with Gupta and Kondo, teaches wherein the multi-dimensional physical-property data includes spectrum data which is detected by performing spectroscopic analysis on the product (Abstract; paragraphs [0001] spectroscopic analysis of at least one sample, in particular of a food or drug, implementing a method for analyzing spectroscopic data based on a multi-way statistical model…; [0015]-[0020] and [0048]).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Gupta and Kondo to add spectroscopic analysis to the combination system of Gupta and Kondo, as taught by Rizkallah above. The modification would have been obvious because one of ordinary skill would be motivated to provide spectroscopic analysis of the sample by analyzing the spectroscopic data based on the multi-channel statistical model in a simple and rapid manner using simple and less expensive equipment, as suggested by Rizkallah ([0014]-[0015]).
Claims 5-6 and 9-11 are rejected under 35 U.S.C. 103 as being unpatentable Gupta (“Selection of important features and predicting wine quality using machine learning techniques,” hereinafter referred to as Gupta), in view of Rizkallah et al. (US 2012/0280146 A1, hereinafter referred to as Rizkallah), and further in view of Zabalza et al (“Novel segmented stacked autoencoder for effective dimensionality reduction and feature extraction in hyperspectral imaging,” hereinafter referred to as Zabalza).
As to claim 5, which incorporates the rejection of claim 3, Gupta and Rizkallah teach spectral data but fail to explicitly teach wherein the multi-dimensional physical-property relevance data is a representative value of an intensity derived for each of a plurality of intervals obtained by dividing the spectrum data.
However, Zabalza, in combination with Gupta and Rizkallah, teaches wherein the multi-dimensional physical-property relevance data is a representative value of an intensity derived for each of a plurality of intervals (interpreted by Examiner as segments/regions) obtained by dividing the spectrum data (1. Introduction, we propose a spectral segmentation in the pixels or samples that can divide the complexity and also allow local extraction of features, eventually providing better extraction capability. In this paper, the segmented SAE (S-SAE) method is introduced, where local SAEs are applied to different segments of the spectrum. By locally working in spectral regions, the computational complexity is reduced and, at the same time, the resulting features are improved thus better classification accuracy is obtained thanks to local extraction of information. From our results it is found that, yet with reduced complexity, S-SAE performs better than the conventional SAE implementation and also other state-of-the-art methods in land-cover analysis, which leaves an open door for future investigation and related ideas; page 3 Fig. 4 presents the generic structure of our proposed S-SAE, where the spectral domain of samples p is segmented into K different regions pk; k ϵ [1, K] to which the SAE technique is applied individually)..
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Gupta and Rizkallah to add value of an intensity derived for each of a plurality of intervals obtained by dividing the spectrum data to the combination system of Gupta and Rizkallah, as taught by Zabalza above. The modification would have been obvious because one of ordinary skill would be motivated to have a reduced complexity but improved efficacy of data abstraction and accuracy of data classification, as suggested by Zabalza (Abstract).
As to claim 6, which incorporates the rejection of claim 4, Gupta and Rizkallah fail to explicitly teach wherein the multi-dimensional physical-property relevance data is a representative value of an intensity derived for each of a plurality of intervals obtained by dividing the spectrum data.
However, Zabalza, in combination with Gupta and Rizkallah, teaches wherein the multi-dimensional physical-property relevance data is a representative value of an intensity derived for each of a plurality of intervals (i.e. segments/regions) obtained by dividing the spectrum data (1. Introduction, we propose a spectral segmentation in the pixels or samples that can divide the complexity and also allow local extraction of features, eventually providing better extraction capability. In this paper, the segmented SAE (S-SAE) method is introduced, where local SAEs are applied to different segments of the spectrum. By locally working in spectral regions, the computational complexity is reduced and, at the same time, the resulting features are improved thus better classification accuracy is obtained thanks to local extraction of information. From our results it is found that, yet with reduced complexity, S-SAE performs better than the conventional SAE implementation and also other state-of-the-art methods in land-cover analysis, which leaves an open door for future investigation and related ideas; page 3 Fig. 4 presents the generic structure of our proposed S-SAE, where the spectral domain of samples p is segmented into K different regions pk; k ϵ [1, K] to which the SAE technique is applied individually)..
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Gupta and Rizkallah to add value of an intensity derived for each of a plurality of intervals obtained by dividing the spectrum data to the combination system of Gupta and Rizkallah, as taught by Zabalza above. The modification would have been obvious because one of ordinary skill would be motivated to have a reduced complexity but improved efficacy of data abstraction and accuracy of data classification, as suggested by Zabalza (Abstract).
As to claim 7, which incorporates the rejection of claim 1, Gupta and Rizkallah fail to explicitly teach wherein the multi-dimensional physical-property data includes image data obtained by imaging the product.
However, Zabalza, in combination with Gupta and Rizkallah, teaches wherein the multi-dimensional physical-property data includes image data obtained by imaging the product (1. Introduction Hyperspectral imaging (HSI) is a very motivating field dealing with several different challenges in the last decade. The HIS cameras and devices provide a spatial 2-D image in hundreds of different wavelengths from the electromagnetic spectrum in nature (spectral bands). As a result, a 3-D structure called hypercube is obtained, where each pixel in the 2-D image is represented by an array of spectral values. Obviously, with such amount of information, the use of HSI data for applications including remote classification of image pixels is proving promising, although it demands advanced signal processing applied to stages such as feature extraction or data reduction.
In the last 2–3 decades, a number of methods have been proposed for feature extraction and data reduction in HSI, including both well-known classical techniques and new approaches. These feature extraction and data reduction techniques aim to boost the general data analysis procedures by improving the characterization of features (efficacy) and/or relieving computational complexity (efficiency). For instance, features containing adequate information usually lead to higher classification accuracy of pixels and, in many cases, this can be done along with a reduction in the number of features (feature dimensionality), which in turn increases the overall efficiency).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Gupta and Rizkallah to add value of an intensity derived for each of a plurality of intervals obtained by dividing the spectrum data to the combination system of Gupta and Rizkallah, as taught by Zabalza above. The modification would have been obvious because one of ordinary skill would be motivated to have features containing adequate information usually lead to higher classification accuracy of pixels and, in many cases, this can be done along with a reduction in the number of features (feature dimensionality), which in turn increases the overall efficiency., as suggested by Zabalza (Introduction).
As to claim 8, which incorporates the rejection of claim 2, Gupta and Rizkallah fail to explicitly teach wherein the multi-dimensional physical-property data includes image data obtained by imaging the product.
However, Zabalza, in combination with Gupta and Rizkallah, teaches wherein the multi-dimensional physical-property data includes image data obtained by imaging the product (1. Introduction Hyperspectral imaging (HSI) is a very motivating field dealing with several different challenges in the last decade. The HIS cameras and devices provide a spatial 2-D image in hundreds of different wavelengths from the electromagnetic spectrum in nature (spectral bands). As a result, a 3-D structure called hypercube is obtained, where each pixel in the 2-D image is represented by an array of spectral values. Obviously, with such amount of information, the use of HSI data for applications including remote classification of image pixels is proving promising, although it demands advanced signal processing applied to stages such as feature extraction or data reduction.
In the last 2–3 decades, a number of methods have been proposed for feature extraction and data reduction in HSI, including both well-known classical techniques and new approaches. These feature extraction and data reduction techniques aim to boost the general data analysis procedures by improving the characterization of features (efficacy) and/or relieving computational complexity (efficiency). For instance, features containing adequate information usually lead to higher classification accuracy of pixels and, in many cases, this can be done along with a reduction in the number of features (feature dimensionality), which in turn increases the overall efficiency).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Gupta and Rizkallah to add value of an intensity derived for each of a plurality of intervals obtained by dividing the spectrum data to the combination system of Gupta and Rizkallah, as taught by Zabalza above. The modification would have been obvious because one of ordinary skill would be motivated to have features containing adequate information usually lead to higher classification accuracy of pixels and, in many cases, this can be done along with a reduction in the number of features (feature dimensionality), which in turn increases the overall efficiency., as suggested by Zabalza (Introduction).
As to claim 9, which incorporates the rejection of claim 3, Gupta and Rizkallah fail to explicitly teach
wherein the multi-dimensional physical-property data includes image data obtained by imaging the product.
However, Zabalza, in combination with Gupta and Rizkallah, teaches wherein the multi-dimensional physical-property data includes image data obtained by imaging the product (1. Introduction Hyperspectral imaging (HSI) is a very motivating field dealing with several different challenges in the last decade. The HIS cameras and devices provide a spatial 2-D image in hundreds of different wavelengths from the electromagnetic spectrum in nature (spectral bands). As a result, a 3-D structure called hypercube is obtained, where each pixel in the 2-D image is represented by an array of spectral values. Obviously, with such amount of information, the use of HSI data for applications including remote classification of image pixels is proving promising, although it demands advanced signal processing applied to stages such as feature extraction or data reduction.
In the last 2–3 decades, a number of methods have been proposed for feature extraction and data reduction in HSI, including both well-known classical techniques and new approaches. These feature extraction and data reduction techniques aim to boost the general data analysis procedures by improving the characterization of features (efficacy) and/or relieving computational complexity (efficiency). For instance, features containing adequate information usually lead to higher classification accuracy of pixels and, in many cases, this can be done along with a reduction in the number of features (feature dimensionality), which in turn increases the overall efficiency).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Gupta and Rizkallah to add value of an intensity derived for each of a plurality of intervals obtained by dividing the spectrum data to the system of Gupta, as taught by Zabalza above. The modification would have been obvious because one of ordinary skill would be motivated to have features containing adequate information usually lead to higher classification accuracy of pixels and, in many cases, this can be done along with a reduction in the number of features (feature dimensionality), which in turn increases the overall efficiency., as suggested by Zabalza (Introduction).
As to claim 10, which incorporates the rejection of claim 4, Gupta and Rizkallah fail to explicitly teach wherein the multi-dimensional physical-property data includes image data obtained by imaging the product.
However, Zabalza, in combination with Gupta and Rizkallah, teaches wherein the multi-dimensional physical-property data includes image data obtained by imaging the product (1. Introduction Hyperspectral imaging (HSI) is a very motivating field dealing with several different challenges in the last decade. The HIS cameras and devices provide a spatial 2-D image in hundreds of different wavelengths from the electromagnetic spectrum in nature (spectral bands). As a result, a 3-D structure called hypercube is obtained, where each pixel in the 2-D image is represented by an array of spectral values. Obviously, with such amount of information, the use of HSI data for applications including remote classification of image pixels is proving promising, although it demands advanced signal processing applied to stages such as feature extraction or data reduction.
In the last 2–3 decades, a number of methods have been proposed for feature extraction and data reduction in HSI, including both well-known classical techniques and new approaches. These feature extraction and data reduction techniques aim to boost the general data analysis procedures by improving the characterization of features (efficacy) and/or relieving computational complexity (efficiency). For instance, features containing adequate information usually lead to higher classification accuracy of pixels and, in many cases, this can be done along with a reduction in the number of features (feature dimensionality), which in turn increases the overall efficiency).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Gupta and Rizkallah to add value of an intensity derived for each of a plurality of intervals obtained by dividing the spectrum data to the combination system of Gupta and Rizkallah, as taught by Zabalza above. The modification would have been obvious because one of ordinary skill would be motivated to have features containing adequate information usually lead to higher classification accuracy of pixels and, in many cases, this can be done along with a reduction in the number of features (feature dimensionality), which in turn increases the overall efficiency., as suggested by Zabalza (Introduction).
As to claim 11, which incorporates the rejection of claim 5, Gupta and Rizkallah fail to explicitly teach wherein the multi-dimensional physical-property data includes image data obtained by imaging the product.
However, Zabalza, in combination with Gupta and Rizkallah, teaches wherein the multi-dimensional physical-property data includes image data obtained by imaging the product (1. Introduction Hyperspectral imaging (HSI) is a very motivating field dealing with several different challenges in the last decade. The HIS cameras and devices provide a spatial 2-D image in hundreds of different wavelengths from the electromagnetic spectrum in nature (spectral bands). As a result, a 3-D structure called hypercube is obtained, where each pixel in the 2-D image is represented by an array of spectral values. Obviously, with such amount of information, the use of HSI data for applications including remote classification of image pixels is proving promising, although it demands advanced signal processing applied to stages such as feature extraction or data reduction.
In the last 2–3 decades, a number of methods have been proposed for feature extraction and data reduction in HSI, including both well-known classical techniques and new approaches. These feature extraction and data reduction techniques aim to boost the general data analysis procedures by improving the characterization of features (efficacy) and/or relieving computational complexity (efficiency). For instance, features containing adequate information usually lead to higher classification accuracy of pixels and, in many cases, this can be done along with a reduction in the number of features (feature dimensionality), which in turn increases the overall efficiency).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Gupta and Rizkallah to add value of an intensity derived for each of a plurality of intervals obtained by dividing the spectrum data to the combination system of Gupta and Rizkallah, as taught by Zabalza above. The modification would have been obvious because one of ordinary skill would be motivated to have features containing adequate information usually lead to higher classification accuracy of pixels and, in many cases, this can be done along with a reduction in the number of features (feature dimensionality), which in turn increases the overall efficiency., as suggested by Zabalza (Introduction).
As to claim 13, which incorporates the rejection of claim 1, Gupta and Rizkallah fail to explicitly teach wherein the first processor is further configured to derive the multi-dimensional physical-property relevance data by applying at least a part of an autoencoder to the multi- dimensional physical-property data.
However, Zabalza, in combination with Gupta and Rizkallah, teaches wherein the first processor is further configured to derive the multi-dimensional physical-property relevance data by applying at least a part of an autoencoder to the multi- dimensional physical-property data (abstract; 1. Introduction To this end, we propose a spectral segmentation in the pixels or samples that can divide the complexity and also allow local extraction of features, eventually providing better extraction capability.
In this paper, the segmented SAE (S-SAE) method is introduced, where local SAEs are applied to different segments of the spectrum. By locally working in spectral regions, the computational complexity is reduced and, at the same time, the resulting features are improved thus better classification accuracy is obtained thanks to local extraction of information; page 2, 3. Stacked autoencoders SAEs can be defined expanding this concept and simply introducing several layers between the input and the output. Therefore, final features are obtained through progressive abstraction levels.
In Fig. 2, a SAE with two layers is shown, where usually F < L.).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Gupta and Rizkallah to add an autoencoder data to the combination system of Gupta and Rizkallah, as taught by Zabalza above. The modification would have been obvious because one of ordinary skill would be motivated to have features containing adequate information usually lead to higher classification accuracy of pixels and, in many cases, this can be done along with a reduction in the number of features (feature dimensionality), which in turn increases the overall efficiency., as suggested by Zabalza (Introduction).
As to claim 16, which incorporates the rejection of claim 13, Gupta fails to explicitly teach wherein the multi-dimensional physical-property data includes image data of a spectrum which is represented by spectrum data detected by performing spectroscopic analysis on the product.
However, Rizkallah, in combination with Gupta, teaches wherein the multi-dimensional physical-property data includes spectrum data which is detected by performing spectroscopic analysis on the product (Abstract; paragraphs [0001] spectroscopic analysis of at least one sample, in particular of a food or drug, implementing a method for analyzing spectroscopic data based on a multi-way statistical model…; [0015]-[0020] and [0048]).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the system of Gupta to add spectroscopic analysis to the system of Gupta, as taught by Rizkallah above. The modification would have been obvious because one of ordinary skill would be motivated to provide a simpler, faster
analysis method requiring simple and less expensive equipment, as suggested by Rizkallah ([0014]).
As to claim 17, which incorporates the rejection of claim 16, Gupta and Rizkallah teach spectral data but fail to explicitly teach wherein the first processor is configured to derive the multi-dimensional physical- property relevance data for each of a plurality of intervals obtained by dividing the spectrum data.
However, Zabalza, in combination with Gupta and Rizkallah, teaches wherein the multi-dimensional physical-property relevance data is a representative value of an intensity derived for each of a plurality of intervals (i.e. segments/regions) obtained by dividing the spectrum data (1. Introduction, we propose a spectral segmentation in the pixels or samples that can divide the complexity and also allow local extraction of features, eventually providing better extraction capability.
In this paper, the segmented SAE (S-SAE) method is introduced, where local SAEs are applied to different segments of the spectrum. By locally working in spectral regions, the computational complexity is reduced and, at the same time, the resulting features are improved thus better classification accuracy is obtained thanks to local extraction of information. From our results it is found that, yet with reduced complexity, S-SAE performs better than the conventional SAE implementation and also other state-of-the-art methods in land-cover analysis, which leaves an open door for future investigation and related ideas; page 3 Fig. 4 presents the generic structure of our proposed S-SAE, where the spectral domain of samples p is segmented into K different regions pk; k ϵ [1, K] to which the SAE technique is applied individually)..
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Gupta and Rizkallah to add value of an intensity derived for each of a plurality of intervals obtained by dividing the spectrum data to the combination system of Gupta and Rizkallah, as taught by Zabalza above. The modification would have been obvious because one of ordinary skill would be motivated to have a reduced complexity but improved efficacy of data abstraction and accuracy of data classification, as suggested by Zabalza (Abstract).
Claim 12 is rejected under 35 U.S.C. 103 as being unpatentable Gupta (“Selection of important features and predicting wine quality using machine learning techniques,” hereinafter referred to as Gupta), in view of Rizkallah et al. (US 2012/0280146 A1, hereinafter referred to as Rizkallah), and further in view of Sebastian “Perspective Article: Flow Synthesis of Functional Materials,” hereinafter referred to as Sebastian),
As to claim 12, which incorporates the rejection of claim 1, Gupta and Rizkallah fail to explicitly teach wherein the product is produced by using a flow synthesis method.
However, Sebastian, in combination with Gupta and Rizkallah, teaches wherein the product is produced by using a flow synthesis method (1. Introduction Advanced functional materials occupy a prominent place in the day-to-day life of a significant portion of the global population. The impact of these materials on the quality of human life is seen either in the direct form (viz., communication devices, energy storage devices, light-emitting devices, reflecting mirrors, currency notes, precious metals and their various forms and compositions, cosmetics, diagnostics, paints and coatings, stain resistant fabric, biomimetic colorants, etc.) or indirectly (viz, catalysts, display technologies, toughened surfaces of high speed machining components, drag-reducing lubricants, functionalized nano silica for precise chemical separations, light-weight materials, nanocomposites in conducing inks, etc.). In general, functional materials can be classified based on their functionality and quality of performance. A variety of these functional materials is used over a very wide range of quantities (from a few mg to few tons).
The functionality of these materials depends on the dimensions of the material,
viz., size and shape for particulate matters, pore size and surface area in the case of
porous materials, and thickness for films. This implies that attaining the desired
dimensions through controlled synthesis is the key to retain the properties that make
these materials functional in true sense. A few such examples that highlight the
impact of size on the specific property and hence application are given in Figure 1.
The only way to achieve such consistency in the properties is through wet chemical
synthesis or through biological routes. Among the two, the later approach is precise
yet unreliable due to the scarcity and purity of specific biological moieties needed
for certain activity and the former option of chemical synthesis becomes a more
reliable approach, albeit only if the synthetic recipe allows consistent product
quality at all scales of production. Continuous-flow synthesis of the functional
materials not only paves the way to achieve consistency in properties but it is
also, scalable and yet decentralized, giving it a flexibility of on-site-on-demand
production. This article aims to provide developing a broader perspective of the
utility and relevance of flow synthesis for the manufacture of a variety of
functional materials and also gazes in to the crystal ball, developing a map for
exploration in as yet new areas, which might rise to prominence in the coming
years).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Gupta and Rizkallah to add flow synthesis to the combination system of Gupta and Rizkallah, as taught by Sebastian above. The modification would have been obvious because one of ordinary skill would be motivated to use flow synthesis because “not only it paves the way to achieve consistency in properties but it is also scalable and yet decentralized, giving it a flexibility of on-site-on-demand production, as suggested by Sebastian (Introduction).
Claims 14-15 are rejected under 35 U.S.C. 103 as being unpatentable over Gupta (“Selection of important features and predicting wine quality using machine learning techniques,” hereinafter referred to as Gupta), in view of Zabalza et al (“Novel segmented stacked autoencoder for effective dimensionality reduction and feature extraction in hyperspectral imaging,” hereinafter referred to as Zabalza), and further in view of Japanese Patent Publication MASUDA et al. (No. JP 2016091359 A , hereinafter referred to as to as MASUDA), which is cited in Applicant's information disclosure statement filed on February, 04,2022. Reference is made herein to the provided machine-translation of MASUDA).
As to claim 14, which incorporates the rejection of claim 13, Gupta and Rizkallah fail to explicitly teach wherein the autoencoder is learned by inputting the multi-dimensional physical- property data of the product of which the quality is higher than a preset level, and
wherein the first processor is configured to:
input the multi-dimensional physical-property data to the autoencoder and output data, and
derive the multi-dimensional physical-property relevance data based on difference data between the multi-dimensional physical-property data which is input to the autoencoder and the output data.
MASUDA, in combination Gupta and Rizkallah, teaches wherein the autoencoder is learned by inputting the multi-dimensional physical-property data of the product of which the quality is higher than a preset level, and
wherein the first processor is configured to:
input the multi-dimensional physical-property data to the autoencoder and output data, and derive the multi-dimensional physical-property relevance data based on difference data between the multi-dimensional physical-property data which is input to the autoencoder and the output data (see page 4, MASUDA teaches that multidimensional physical property data for a product that meets a predetermined standard is used to train the dimension reduction means and the dimension restoration means, i.e. the autoencoders. The auto-encoder is thus learned by inputting multidimensional physical-property data of a product of which the quality is higher than a preset level, i.e. that meets the predetermined standard. Like further noted above, pages 5 and 19, MASUDA also teaches that an error calculation unit calculates a multidimensional error vector based on the difference between the multi-dimensional physical-property data input to the autoencoders and the multidimensional data output from the autoencoders. As noted above, this error data is interpreted as multidimensional physical property-relevance data/learning input data like claimed.
MASUDA thus further teaches inputting the multi-dimensional physical property data to the autoencoder, and deriving the multi-dimensional physical property relevance data, i.e. the error data, based on difference data between the multi-dimensional physical property data which is input to the autoencoder and the output data output from the autoencoder).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Gupta and Rizkallah to add difference data to the combination system of Gupta and Rizkallah, as taught by MASUDA above. The modification would have been obvious because one of ordinary skill would be motivated to use error data to identify whether the object to be identified is a product that satisfies, for example, a standard, as suggested by MASUDA (page 4).
As to claim 15, which incorporates the rejection of claim 13, Gupta and Rizkallah fail to explicitly teach wherein the first processor is configured to input the multi-dimensional physical-property data to the autoencoder and output feature data from an encoder network of the autoencoder, and derive the multi-dimensional physical-property relevance data based on the feature data.
MASUDA, in combination Gupta and Rizkallah, teaches wherein the first processor is configured to input the multi-dimensional physical-property data to the autoencoder and output feature data from an encoder network of the autoencoder, and derive the multi-dimensional physical-property relevance data based on the feature data (see e.g. page 06), MASUDA teaches that an error calculation unit calculates a multidimensional error vector based on the difference between the multidimensional physical-property data input to the auto-encoders and the multidimensional data output from the autoencoders. As noted above, this error data is interpreted as multidimensional physical property-relevance data/learning input data like claimed; page 07, MASUDA thus further teaches inputting the multi-dimensional physical property data to the autoencoder, outputting feature data from an encoder network of the autoencoder, and deriving the multi-dimensional physical property relevance data, i.e. the error vector, based on the feature data).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Gupta and Rizkallah to add difference data to the combination system of Gupta and Rizkallah, as taught by MASUDA above. The modification would have been obvious because one of ordinary skill would be motivated to use error data to identify whether the object to be identified is a product that satisfies, for example, a standard, as suggested by MASUDA (page 4).
Claim 18 are rejected under 35 U.S.C. 103 as being unpatentable over Japanese Patent Publication MASUDA et al. (No. JP 2016091359 A, hereinafter referred to as to as MASUDA), which is cited in Applicant's information disclosure statement filed on February, 04,2022. Reference is made herein to the provided machine-translation of MASUDA), in view of Chidlovskii et al. (US 2017/0220951 A1, hereinafter referred to as Chidlovskii)
As to claim 18, MASUDA teaches an operating apparatus comprising [a second processor;
wherein the second processor] is configured to
acquire (Abstract: data acquisition unit) the learned model which is output from the first processor of the learning apparatus according to claim 1 (Abstract an output unit that outputs an identification result obtained by the identification unit.);
acquire multi-dimensional physical-property relevance data (page 3, the information processing apparatus 10 identifies whether or not the object is a product that satisfies, for example, a standard based on multidimensional data regarding the physical characteristics of the object. Note that the multidimensional data regarding the physical characteristics of the object is an example of the first data; i.e. error data) for prediction which is data of a product of which a quality is unknown (page 2, evaluate unexpected (interpreted by Examiner as unknown ) defective (interpreted by Examiner as quality) products);
input the multi-dimensional physical-property relevance data for prediction which is data of the product of which the quality is unknown to the learned model (e.g. to an identification processing unit implementing a support vector machine, SVM)
and predict the quality ((i.e. input the multi-dimensional physical-property relevance
data for prediction to the learned model and predict the quality); and control outputting of a prediction result of the quality by the learned model (i.e. of the quality).
MASUDA, however, does not disclose that these tasks are implemented on an
operating apparatus comprising a second processor, which acquires the learned model from the first processor of the learning apparatus, as required by claim 18.
Training a machine learning model on a first device and then providing the machine
learning model to a second device to perform classification on input data is nevertheless taught in the art.
Chidlovskii describes a system comprising a first apparatus (i.e. a machine learning device) that acquires learning input data (i.e. training instances), inputs the learning input data to a machine learning model (i.e. a classifier), performs learning, and then outputs the machine learning model as a learned model to be provided for actual operation (paragraphs [0015]- [0020]).
Chidlovskii further teaches that a second apparatus comprising a second processor can then: (i) acquire the learned model which is output by the processor of the first apparatus; (ii) acquire multi-dimensional data for prediction which is data of which a classification is unknown (i.e. acquire an unlabeled input instance comprising a feature vector); (iii) input the multi-dimensional data for prediction to the learned model and predict the classification (i.e. a label); and (iv) output the prediction result by the learned model (see paragraphs [0015] and [0020]-[0021] The classifier 40 operates on the input feature vector x,n to generate (i.e. predict) a label 44 for the input instance 42.).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the system of MASUDA to add an apparatus comprising a second processor to the system of MASUDA, as taught by Chidlovskii, above. The modification would have been obvious because one of ordinary skill would be motivated to enable more suitable computers to perform the training the classifier (learning phase)
and classification tasks, as suggested by Chidlovskii ([0015]).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Patents and patent related publications are cited in the Notice of References Cited (Form PTO-892) attached to this action to further show the state of the art with respect to the invention.
Ikeda et al. (US 12, 406, 208 B2) teach a feature value generation apparatus e.g. abnormality detection apparatus, has detection unit normalized or standardized by generation unit and detects abnormality based on numeric vector and learning result for every numeric vector.
NAKATSUJI et al. (US-20220100932-A1) teach a design support method for assisting design of metal material with desired characteristic by computer, involves presenting component composition and manufacturing conditions among design conditions corresponding to desired characteristics
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ABABACAR SECK whose telephone number is (571)270-7146. The examiner can normally be reached Monday-Friday 8:00 A.M.-6:00 P.M..
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, Lamardo Viker can be reached at 571-270-5871. 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.
/ABABACAR SECK/Examiner, Art Unit 2147
/VIKER A LAMARDO/Supervisory Patent Examiner, Art Unit 2147
Prosecution Insights