Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
Acknowledgment is made of applicant’s claim for foreign priority under 35 U.S.C. 119 (a)-(d).
The certified copy/copies of JP2024-084320 has/have been received on 05/08/2025.
Information Disclosure Statement
The information disclosure statement (IDS) received on 04/24/2025 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Drawings
The drawings are objected to because Figure 8 reads “G: Absolute Value of Complex Morley”, it is suggested to read “G: Absolute Value of Complex Morlet”. Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. The figure or figure number of an amended drawing should not be labeled as “amended.” If a drawing figure is to be canceled, the appropriate figure must be removed from the replacement sheet, and where necessary, the remaining figures must be renumbered and appropriate changes made to the brief description of the several views of the drawings for consistency. Additional replacement sheets may be necessary to show the renumbering of the remaining figures. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Specification
The abstract of the disclosure is objected to because
the abstract fails to “enable the Office and the public generally to determine quickly from a cursory inspection, the nature and gist of the technical disclosure” 37 CFR 1.72(b). Specifically, it is not clear what this technology is used for or what the advantages may include and the abstract is nearly a copy of independent Claim 1.
A corrected abstract of the disclosure is required and must be presented on a separate sheet, apart from any other text. See MPEP § 608.01(b).
The disclosure is objected to because of the following informalities:
In [0022], “Morley wavelet” is suggested to read “Morlet wavelet”.
In [0025], both instances of “Morley wavelet” are suggested to read “Morlet wavelet”.
In [0035], both instances of “Morley wavelet” are suggested to read “Morlet wavelet”.
In [0036], “Morley wavelet” is suggested to read “Morlet wavelet”.
In [0080], both instances of “Morley wavelet” are suggested to read “Morlet wavelet”.
Appropriate correction is required.
Claim Objections
Claims 1, 7-8, and 10 are objected to because of the following informalities:
In Claim 1, “execute image generation processing comprising: …” is suggested to read “execute an image generation process comprising: …” or “execute an image generation method comprising: …”.
In Claim 7, both a “complex Morley wavelet transform” and “complex Morley wavelet” are suggested to read “complex Morlet wavelet transform” and “complex Morlet wavelet”.
In Claim 8, “execute image generation processing comprising: …” is suggested to read “execute an image generation process comprising: …” or “execute an image generation method comprising: …”.
In Claim 10, “execute image generation processing comprising: …” is suggested to read “execute an image generation process comprising: …” or “execute an image generation method comprising: …”.
Appropriate correction is required.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1 and 11 are rejected under 35 U.S.C. 103 as being unpatentable over hossein mobaraki, “How to merge man y time-frequency scalograms into a single image?”, 2024, mathworks, hereinafter referenced as Mobaraki in view of Boufounos et al. (US 20180352249 A1), hereinafter referenced as Boufounos, and in further view of He et al. “FT-FVC: fast transformation-based feature vector concatenation for time series classification”, 2022.
Regarding Claim 1, Mobaraki discloses
generating a multi-dimensional first image representing a frequency characteristic at each time of each piece of time-series data, based on each piece of multi-dimensional time-series data (Mobaraki: [pg. 1], discloses generating a scalogram representing EEG signals using Bonn University data <multi-dimensional first image representing a frequency characteristic at each time of each piece of time series data, the title of the forum post discloses the scalograms represent a time-frequency relationship>, based on each data, therefore creating 500 scalograms <piece of multi-dimensional time-series data>; see below code for creating scalogram data);
PNG
media_image1.png
750
940
media_image1.png
Greyscale
and generating a single second image obtained by combining the multi-dimensional first images (Mobaraki: [pg. 1-2], discloses wanting to create a single image <single second image> representing a scalogram of the entire data <combining the multi-dimensional first images>)
image generation processing (Mobaraki: [pg. 1], discloses a method of image generation from time-series data; see the code in the screenshot above)
Note: While the forum post has a response from 7 Oct 2024, it is not relied upon for this rejection. The contents relied upon are dated 17 Jan 2024 and 18 Jan 2024, both falling before the effective filing date of the claimed invention.
Mobaraki fails to explicitly disclose
A non-transitory recording medium storing a program that causes a computer to execute image generation processing comprising: …
combining … weighted using a random matrix in which a different value is assigned for each frequency.
However, Boufounos discloses
A non-transitory recording medium storing a program that causes a computer to execute (Boufounos: [0019], discloses a non-transitory computer readable storage medium embodied thereon a program executable by a computer for performing a method) comprising: …
measured using a random matrix in which a different value is assigned for each frequency (Boufounos: [0105], discloses a signal measured using a random matrix with i.i.d. elements <i.i.d. stands for independent identical distribution, meaning the elements of the signal, the frequencies, each have different values>; further supported in [0087 and 0099]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply and/or modify the method disclosed by Mobaraki by applying the method to a non-transitory recording medium and using a matrix with an independent identical distribution as taught by Boufounos. One of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to make this modification because a non-transitory recording medium is highly reliable for long-term data archiving and will not be affected by power outages. Further, one of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to make this modification of using an i.i.d. random matrix to simplify mathematical modeling by ensuring that entries share the same probability and have no mutual correlation.
The combination of Mobaraki and Boufounos fail to explicitly disclose
combining … weighted using a random matrix
However, He discloses
combining vector representations of time-series data weighted using a random matrix (He: [1. Introduction, pg. 3], discloses time series transformations combined with random convolutional kernel transform and feature vector concatenation to generate a three-view <combining vectors weighted using random kernels of the ROCKET algorithm, ROCKET algorithm further uses random projection>; see [Fig. 2] below).
PNG
media_image2.png
378
808
media_image2.png
Greyscale
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply and/or modify the non-transitory computer-readable medium disclosed by the combination of Mobaraki and Boufounos by using the ROCKET algorithm as taught by He. One of ordinary skill in the art before the effective filing date of the claimed invention would have been able to make this modification high accuracy and performance when processing time-series data.
Regarding Claim 11, it recites limitations similar to Claim 1, but as a method. As shown in the rejection, the combination of Mobaraki, Boufounos, and He disclose the non-transitory recording medium of Claim 1. The combination of Mobaraki, Boufounos, and He further disclose
An image generation method (Mobaraki: [pgs. 1-2], discloses a solution for creating a single image representing scalogram of the entire data), comprising: …
Claims 2 and 12 are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Mobaraki, Boufounos, and He in view of Meshram, “Audio Separation Using Principal Component Analysis (PCA)”, 2024, sameermeshram.com, hereinafter referenced as Meshram.
Regarding Claims 2 and 12, the combination of Mobaraki, Boufounos, and He disclose the non-transitory recording medium and method of Claims 1 and 11 respectively. The combination of Mobaraki, Boufounos, and He fail to explicitly disclose the limitations of Claims 2 and 12, however, Meshram disclose(s) wherein:
in the generating of the multi-dimensional first image, the multi-dimensional time-series data is converted into time-series data indicating a feature amount mapped onto a multi-dimensional principal component axis by principal component analysis, and the multi-dimensional first image is generated based on the converted time-series data (Meshram: [images 6-10], discloses generating spectrograms <including a multi-dimensional first image> wherein the data undergoes dimensionality reduction through PCA <the multi-dimensional time-series data is converted into time-series data indicating a feature amount mapped onto a multi-dimensional principal component axis by principal component analysis> and then is visualized through a spectrogram ).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply and/or modify the non-transitory recording medium or method disclosed by the combination of Mobaraki, Boufounos, and He by generating a spectrogram after performing PCA for signal analysis as taught by Meshram. One of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to make this modification for precise dimensionality reduction and feature extraction.
Claims 3 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Mobaraki, Boufounos, He, and Meshram in view of Singh et al., “Weighted mixed-norm minimization based joint compressed sensing recovery of multi-channel electrocardiogram signals”, 2016, hereinafter referenced as Singh, and in further view of Takahashi et al. (US 20190012816 A1), hereinafter referenced as Takahashi.
Regarding Claims 3 and 13, the combination of Mobaraki, Boufounos, He, and Meshram disclose the non-transitory recording medium and method of Claims 2 and 12 respectively. The combination of Mobaraki, Boufounos, He, and Meshram further disclose(s) wherein:
… of first images among the multi-dimensional first images (Mobaraki: [pg. 1], discloses images <first images> of scalograms to be merged <where the set scalograms is interpreted as an equal subset of itself>)
each frequency (Mobaraki: [pg. 1], discloses time-series data represented as scalograms with a feature of “SamplingFrequencies”)
The combination of Mobaraki, Boufounos, He, and Meshram fail to explicitly disclose
the random matrix is a matrix using, as an element, a value obtained in a manner that a sparse matrix, in which a weight for predetermined pieces (Singh: [3. Results and discussions, pg. 6], discloses a random sparse matrix in which d <predetermined pieces> non zero values at random positions and every other value is 0 <binary>),
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply and/or modify the non-transitory recording medium or method disclosed by Mobaraki, Boufounos, He, and Meshram by using random sampling and binary weighting for a sparse random matrix as taught by Singh. One of ordinary skill in the art before the effective filing date of the claimed invent would have been motivated to make this modification to reduce computational load and storage complexity while keeping key properties of the data.
The combination of Mobaraki, Boufounos, He, Meshram, and Singh fail to explicitly disclose
a value is multiplied by a contribution degree obtained by the principal component analysis, at each frequency
and a value after multiplication is normalized for each frequency
PNG
media_image3.png
874
458
media_image3.png
Greyscale
However, Takahashi discloses
a value is multiplied by a contribution degree obtained by the principal component analysis, at each principal component (Takahashi: [0057-0059], discloses a coefficient cj representing the contribution rate of a principal component Vj <for each principal component, each one has a different contribution rate>; see [Mathematical formula 2] on left illustrating cj Vj <interpreted as a value multiplied by a contribution degree>),
and a value after multiplication is normalized for each principal component (Takahashi: [0057-0059], discloses the size <a value> is standardized <normalized> for each principal component <calculating principal components requires matrix multiplication so standardization of principal components would have to happen after a multiplication operation>).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply and/or modify the non-transitory recording medium or method disclosed by the combination of Mobaraki, Boufounos, He, Meshram, and Singh by utilizing weighting and standardization as taught by Takahashi. One of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to make this modification to improve data quality, model accuracy, and fair statistical analysis.
Claims 4 and 14 are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Mobaraki, Boufounos, and He in view of Thamm et al., “Random matrix analysis of deep neural network weight matrices”, 2022, hereinafter referenced as Thamm.
Regarding Claims 4 and 14, the combination of Mobaraki, Boufounos, and He disclose the non-transitory recording medium and method of Claims 1 and 11 respectively. The combination of Mobaraki, Boufounos, and He further disclose(s) wherein:
the generating of the second image includes merging many time-frequency scalograms (Mobaraki: [pg. 1], discloses wherein generating a single image <a second image> includes merging multiple time-frequency scalograms)
The combination of Mobaraki, Boufounos, and He fail to explicitly disclose
performing weighting on the multi-dimensional first image by using a different random matrix at each time
However, Thamm discloses
performing weighting on the multi-dimensional first image by using a different random matrix at each time (Thamm: [II. Main Results, pg. 2], discloses weighting on the input image < images are interpreted as multi-dimensional because they are 2D arrays> by using an initial weight matrix that is random at each layer <at each time the initial image passes through a layer, it will be weighted using a different random matrix each time>).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply and/or modify the non-transitory recording medium or method disclosed by the combination of Mobaraki, Boufounos, and He by weighting the input image using different random matrices for the initial weight matrices at each layer as taught by Thamm. One of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to make this modification to allow a model to learn distinct features of inputs.
Claims 5 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Mobaraki, Boufounos, He, and Thamm in view of Łuczak et al., “Cloud Based Fault Diagnosis by Convolutional Neural Network as Time–Frequency RGB Image Recognition of Industrial Machine Vibration with Internet of Things Connectivity”, 2023, hereinafter referenced as Łuczak.
Regarding Claims 5 and 15, the combination of Mobaraki, Boufounos, He, and Thamm disclose the non-transitory recording medium and method of Claims 4 and 15 respectively. The combination of Mobaraki, Boufounos, He, and Thamm further disclose(s) wherein:
and the generating of the second image includes merging many time-frequency scalograms (Mobaraki: [pg. 1], discloses wherein generating a single image <a second image> includes merging multiple time-frequency scalograms)
The combination of Mobaraki, Boufounos, He, and Thamm fails to explicitly disclose
embedding a weighted value for the multi-dimensional first image in a first component among an R component, a G component, and a B component of an RGB image, embedding, in a second component, different values in stages in accordance with a frequency, and embedding, in a third component, different values in stages in accordance with elapse of time
However, Łuczak discloses
embedding a weighted value for the multi-dimensional first image in a first component among an R component, a G component, and a B component of an RGB image, embedding, in a second component, different values in stages in accordance with a frequency, and embedding, in a third component, different values in stages in accordance with elapse of time (Łuczak: [7. Recognition of a Time-Frequency RGB Image of Vibration, pg. 13-14], discloses generating an image by embedding data samples <different values in stages> from the X-axis into a R channel, samples <different values in stages> from the Y axis into a G channel, and samples from the Z-axis <different values in stages> into a B channel <where each channel is interpreted as a respective first/second/third component and each axis represents features>; [Fig. 14], illustrates the X-axis represents time, the Y-axis represents frequency, and the Z-axis, represented as color, is a density value, a weight <this is interpreted as the Z-axis embedded among a B channel, a first component; the Y-axis embedded into the G channel, the second component; and the X-axis embedded into a R channel, the third component>; see [Fig. 16] below).
PNG
media_image4.png
104
269
media_image4.png
Greyscale
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply and/or modify the non-transitory recording medium or method disclosed by the combination of Mobaraki, Boufounos, He, and Thamm by using feature embedding with RGB channels as taught by Łuczak. One of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to make this modification for efficient data visualization, enabling data interpretability by viewing many (more than 2) dimensions in a 2D space.
Claims 6 and 16 are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Mobaraki, Boufounos, and He in view of Zontone et al., “Convolutional Neural Networks Using Scalograms for Stress Recognition in Drivers”, 2023, hereinafter referenced as Zontone, and in further view of Łuczak.
Regarding Claims 6 and 16, the combination of Mobaraki, Boufounos, and He disclose the non-transitory recording medium and method of Claims 1 and 11 respectively. The combination of Mobaraki, Boufounos, and He further disclose(s) wherein:
and the generating of the second image includes merging many time-frequency scalograms (Mobaraki: [pg. 1], discloses wherein generating a single image <a second image> includes merging multiple time-frequency scalograms)
The combination of Mobaraki, Boufounos, and He fail to explicitly disclose
the generating of the multi-dimensional first image includes generating two types of the multi-dimensional first images representing two different types of frequency characteristics for each piece of time-series data
the generating of the
However, Zontone discloses
the generating of the multi-dimensional first image includes generating two types of the multi-dimensional first images representing two different types of frequency characteristics for each piece of time-series data (Zontone: [A. The datasets, pg. 2-3], discloses generating SQR and HR scalograms each representing types of frequency characteristics at each time of each piece of time series data for dataset 1 <based on> of multi-dimensional time-series data; see the SPR scalograms and the HR scalograms of [Figs. 2-4] below <includes generating a first multi-dimensional image),
PNG
media_image5.png
864
662
media_image5.png
Greyscale
PNG
media_image6.png
662
1078
media_image6.png
Greyscale
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply and/or modify the non-transitory recording medium or method disclosed by the combination of Mobaraki, Boufounos, and He by generating different types of scalograms as taught by Zontone. One of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to make this modification for complete data characterization and accurate anomaly detection.
The combination of Mobaraki, Boufounos, He, and Zontone fail to explicitly disclose
the generating of the
However, Łuczak discloses
the generating of the (Łuczak: [7. Recognition of a Time-Frequency RGB Image of Vibration, pg. 13-14], discloses generating an image by embedding data samples <different values in stages> from the X-axis into a R channel, samples <different values in stages> from the Y axis into a G channel, and samples from the Z-axis <different values in stages> into a B channel <where each channel is interpreted as a respective first/second/third component and each axis represents features>; [Fig. 14], illustrates the X-axis represents time, the Y-axis represents frequency, and the Z-axis, represented as color, is a density value, a weight <this is interpreted as the Z-axis embedded among a B channel, a first component; the Y-axis embedded into the G channel, the second component; and the X-axis embedded into a R channel, the third component>; see [Fig. 16]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply and/or modify the non-transitory recording medium or method disclosed by the combination of Mobaraki, Boufounos, He, and Zontone by using feature embedding with RGB channels as taught by Łuczak. One of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to make this modification for efficient data visualization, enabling data interpretability by viewing many (more than 2) dimensions in a 2D space.
Claim 7 is rejected under 35 U.S.C. 103 as being unpatentable over the combination of Mobaraki, Boufounos, He, Zontone, and Łuczak in view of Kosunen et al. (US 20230181902 A1), hereinafter referenced as Kosunen.
Regarding Claim 7, the combination of Mobaraki, Boufounos, He, Zontone, and Łuczak disclose the non-transitory recording medium of Claim 6. The combination of Mobaraki, Boufounos, He, Zontone, and Łuczak fail to explicitly disclose the limitations of Claim 7, however Kosunen disclose(s) wherein:
the two types of frequency characteristics are set as a real part and an imaginary part of complex Morley wavelet transform, or set as an absolute value of a complex Morley wavelet and a Ricker wavelet (Kosunen: [0140], discloses cosine and imaginary cosine <sin> waves are set as real and imaginary parts in the complex Morlet wavelet transform <sin and cosine are types of waves, interpreted as frequency characteristics>; see the bottom graph on right side of [Fig. 3] below).
PNG
media_image7.png
324
720
media_image7.png
Greyscale
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply and/or modify the non-transitory recording medium disclosed by the combination of Mobaraki, Boufounos, He, Zontone, and Łuczak by modelling real and imaginary parts of a complex Morlet wavelet transform as taught by Kosunen. One of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to make this modification to decouple features for instantaneous feature extraction, enabling robust signal analysis.
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Mobaraki in view of Shen et al. (US 20190122394 A1), hereinafter referenced as Shen.
Regarding Claim 8, Mobaraki discloses
generating a multi-dimensional first image representing a frequency characteristic at each time of each piece of time-series data, based on each piece of multi-dimensional time-series data (Mobaraki: [pg. 1], discloses generating a scalogram representing EEG signals using Bonn University data <multi-dimensional first image representing a frequency characteristic at each time of each piece of time series data, the title of the forum post discloses the scalograms represent a time-frequency relationship>, based on each data, therefore creating 500 scalograms <piece of multi-dimensional time-series data>);
and generating a single second image obtained by combining the multi-dimensional first images in a manner that uses code (Mobaraki: [pg. 1-2], discloses wanting to create a single image <single second image> representing a scalogram of the entire data <combining the multi-dimensional first images> in a manner that uses code)
Mobaraki fails to explicitly disclose
A non-transitory recording medium storing a program that causes a computer to execute image generation processing comprising:
the generated multi-dimensional first image is input to an encoder obtained by training a self-encoder to convert the multi-dimensional first image into a single image by using the multi-dimensional first image as training data.
However, Shen discloses
A non-transitory recording medium storing a program that causes a computer to execute image generation processing (Shen: [Claims 20 and 11], discloses a non-transitory machine-readable storage medium storing machine-readable instruction codes, wherein the instruction codes, when being read and executed by a computer, cause the computer to perform an image processing method) comprising:
the generated multi-dimensional first image is input to an encoder obtained by training a self-encoder to convert the multi-dimensional first image into a single image by using the multi-dimensional first image as training data (Shen: [0068-0069, Fig. 12], discloses a second training phase where the training images <digital images are interpreted as multi-dimensional because they are 2D arrays> are input into a self-encoder obtained by a training, with the first training phase, the self-encoder to convert training images into reconstructed images <training images from the first phase of training are re-used in the second phase of training the autoencoder>).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply and/or modify the method disclosed by Mobaraki by applying the method to a non-transitory recording medium and re-using training data as taught by Shen. One of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to make this modification because a non-transitory recording medium is highly reliable for long-term data archiving and will not be affected by power outages. Further, one of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to make this modification for data efficiency, domain adaption, allowing the model to deeply learn and compress the data distribution.
Claim 9 is rejected under 35 U.S.C. 103 as being unpatentable over the combination of Mobaraki and Shen in view of Daragahi et al., “NDE Data Correlation Using Encoder-Decoder Networks with Wavelet Scalogram Images”, 2022, hereinafter referenced as Daragahi.
Regarding Claim 9, the combination of Mobaraki and Shen disclose the non-transitory recording medium of Claim 8. The combination of Mobaraki and Shen fail to explicitly disclose the limitations of Claim 9, however, Dargahi disclose(s) wherein:
the encoder is the encoder in the self-encoder trained to convert a multi-dimensional restoration image of the multi-dimensional first image obtained by inputting the multi-dimensional first image obtained by converting the multi-dimensional time-series data to the self-encoder, which includes an encoder and a decoder, into multi-dimensional restoration time-series data, and to minimize an error between the multi-dimensional time-series data and the multi-dimensional restoration time-series data (Dargahi: [Fig. 1, pg. 4], illustrates an encoder that is the encoder of an autoencoder <self-encoder> that converts <is trained to convert> a reconstructed low level feature map <multi-dimensional restoration image> of the input scalogram <multi-dimensional first image> obtains by converting the waveform data <multi-dimensional time-series data> and inputting it into the autoencoder, which includes an encoder and decoder, into a multi-dimensional restoration time-series data <outputted restored scalogram>; [4.4 Autoencoder Implementation], discloses the autoencoder was trained to perform the process; [3.2.2 Autoencoder Architecture], discloses minimizing error between the reconstructed NDE data and the ground truth data <between the multi-dimensional time-series data and the multi-dimensional restoration time-series data>; see [Fig. 1] below).
PNG
media_image8.png
738
848
media_image8.png
Greyscale
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply and/or modify the non-transitory recording medium disclosed by the combination of Mobaraki and Shen by using the autoencoder approach of time-series data representation reconstruction as taught by Dargahi. One of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to make this modification identify shared latent features automatically utilizing a powerful framework for analyzing complex, time-varying data with accurate feature extraction and improved latent space quality.
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Zontone in view of Shen.
Regarding Claim 10, Zontone discloses
generating a plurality of types of multi-dimensional first images representing a plurality of types of different frequency characteristics at each time of each piece of time-series data, based on each piece of multi-dimensional time-series data (Zontone: [A. The datasets, pg. 2-3], discloses generating SQR and HR scalograms representing types of frequency characteristic at each time of each piece of time series data for dataset 1 <based on> of multi-dimensional time-series data; see the SPR scalograms and the HR scalograms of [Figs. 2-4] below);
PNG
media_image5.png
864
662
media_image5.png
Greyscale
PNG
media_image6.png
662
1078
media_image6.png
Greyscale
and generating output concatenation obtained by combining the plurality of types of multi-dimensional first images (Zontone: [Fig. 4], discloses generating an output concatenation obtained by combining the plurality of SQR and HR scalograms plurality of types of multi-dimensional first images).
Zontone fails to explicitly disclose
A non-transitory recording medium storing a program that causes a computer to execute image generation processing comprising: …
and generating a single second image obtained by …
However, Shen discloses
A non-transitory recording medium storing a program that causes a computer to execute image generation processing (Shen: [Claims 20 and 11], discloses a non-transitory machine-readable storage medium storing machine-readable instruction codes, wherein the instruction codes, when being read and executed by a computer, cause the computer to perform an image processing method) comprising: …
and generating a single second image obtained by inputting the training image into a self-encoder (Shen: [0068], discloses generating a reconstructed image <second image> of the training image obtained by inputting the training image into a self-encoder)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply and/or modify the method disclosed by Zontone by applying the method to a non-transitory recording medium and generating a second image as taught by Shen. One of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to make this modification because a non-transitory recording medium is highly reliable for long-term data archiving and will not be affected by power outages. Further, one of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to make this modification for precise anomaly detection.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Andrzejak et al., “Indications of nonlinear deterministic and finite-dimensional structures in times series of brain electrical activity: Dependence of recording region and brain state”, 2001 discloses the EEG Bonn University dataset.
Dempster et al., “ROCKET: Exceptionally fast and accurate time series classification using random convolutional kernels”, 2019 discloses the state-of-the-art ROCKET method.
Kannan et al., “Analysis and Comparison of Different Wavelet Transform Methods Using Benchmarks for Image Fusion”, 2020 discloses wavelet based image fusion.
Zaman et al., “Fault Diagnosis in Centrifugal Pumps: A Dual-Scalogram Approach with Convolution Autoencoder and Artificial Neural Network”, 2020 discloses a convolutional autoencoder that processes scalograms.
Amer et al. (US 20250213197 A1) discloses processing signals for anomaly detection.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ISABELLA OCHSNER whose telephone number is (571)272-9322. The examiner can normally be reached 9:30 - 6:00 PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Devona Faulk can be reached at (571) 272-7515. 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.
/I.O./Examiner, Art Unit 2618
/DEVONA E FAULK/Supervisory Patent Examiner, Art Unit 2618
Prosecution Insights