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 .
DETAILED ACTION
The amendment filed 08/08/2025 has been entered. Claims 1-26 remain pending in the application.
Priority
Acknowledgment is made of applicant's claim for foreign priority based on an application filed in Europe on 06/19/2023. It is noted, however, that applicant has not filed a certified copy of the EP23180010.3 application as required by 37 CFR 1.55.
Response to Arguments
Applicant’s amendments to the claims are sufficient to overcome the objection to claim 14. Accordingly, the objection has been withdrawn.
Applicant’s amendments to the claims are sufficient to overcome the rejection under 35 U.S.C. 112 (b) of claims 1-15. Accordingly, the rejection has been withdrawn.
Applicant' s arguments with respect to claim(s) 1 and 15 and all subsequent dependent claims have been considered but are moot in view of the references cited in the most current rejection. Examiner now relies upon the prior art references of Kowarski (“Near real-time marine mammal monitoring from gliders”) in view of Groenaas (US 20120120760 A1) with respect to claim 1 and Kowarski in view of Bergler (“An Automatic Killer Whale Sound Detection Toolkit Using Deep Learning”) with respect to claim 15 as detailed below.
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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 1-2 and 12-14 are rejected under 35 U.S.C. 103 as being unpatentable over Kowarski (“Near real-time marine mammal monitoring from gliders”) in view of Groenaas (US 20120120760 A1).
Regarding claim 1, Kowarski teaches a computer implemented method of detecting marine mammals, the method comprising: receiving at a server (ensemble data were received by Teledyne Webb’s server and accessed by JASCO over the internet) acoustic data from one or more hydrophones (“top-mounted hydrophone”). (Abstract, Section II.A, 3. Section. Stage 2: Candidate detection prioritization and data transfer, Fig.1) Kowarski discloses the claimed invention except for a cloud server. It would have been obvious to one having ordinary skill in the art at the time the invention was filled to incorporate a cloud server, since it has been held to be within general skill of a worker in the art to select a known material on the basis of its suitability for the intended use as a matter of obvious design choice.
Kowarski also teaches sampling the acoustic data and transforming the sampled acoustic data to time-frequency image data (“The algorithm applies pre-set Fast Fourier Transform (FFT) settings to create a magnitude spectrogram of length “N” seconds; each frequency in the spectrogram is then median-normalized (Table I).”). (Section II.B.2, Table.1)
Kowarski also teaches to input the image data to at least one model trained to detect the presence or absence of marine mammal vocalizations in the acoustic data (“The FFT parameters, spectral candidate detection threshold, and contour parameters were determined by a trained analyst (K.K.) during the tuning of automated detectors for real-time monitoring of baleen whales in the Gulf of St Lawrence (Table I).”), wherein the model automatically outputs a prediction of whether or not a mammal is present. (Section II.B.2, Table.1, Table.2)
Kowarski also teaches providing output to a user indicating the prediction (Gliders could effectively monitor these buffer regions, reporting when whales are present) via a real time or near real time user interface (near real-time marine mammal monitoring), the user interface displaying, for each model, the time-frequency image data (Table. 1, Figs.3-4) together with a probability score (highest-ranked score, each ensemble had a four-number score) of the presence of a marine mammal (Table. 1, Table.2) for the user to confirm the detection (Human validated near real-time detector performance, human analysts confirmed the occurrence of marine mammal acoustic signals). (Section II, Section IV.C, Table. 1, Table.2, Table.4 Figs.3-4)
Kowarski does not explicitly teach providing output to a remote user and providing user controls to cause some further action upon confirmed detection.
Groenaas teaches providing output to a remote user (18, 106, 110, 112) and providing user controls to cause some further action upon confirmed detection (dynamic control of the survey operations). (Paragraphs 4, 24, 36-37, Fig.3)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate providing output to a remote user and providing user controls to cause some further action upon confirmed detection in order to process the received backscattered acoustic signals to detect the presence of marine mammals.
Regarding claim 2, Kowarski teaches inputting the image data to each of at least two different models, respectively arranged to detect marine mammal sounds or vocalizations in different frequency ranges corresponding respectively to different mammal sounds or vocalizations. (Section II.B.2, Section II.B.3, Table.1, Table.2)
Regarding claim 12, Kowarski teaches using a sliding window of plural time slices of audio data as input to the models. (Section II.B.2, Table.1)
Regarding claim 13, Kowarski teaches performing prediction pooling on plural successive windows such that plural positive detections output from the model are required for an overall positive detection. (Section II.B.3)
Regarding claim 14, Kowarski teaches automatically or in response to accepting user input, ceasing at least one on board marine activity if the prediction indicates the presence of a marine mammal, wherein the acoustic data and/or image data and prediction are displayed to a user for validation, and optionally the method comprises receiving user input indicating validation of the prediction, wherein the user validation overrides the decision to cease the marine seismic activity. (Section I, Section II.B.4)
Claim(s) 3-5 are rejected under 35 U.S.C. 103 as being unpatentable over Kowarski in view of Groenaas and Bergler (“An Automatic Killer Whale Sound Detection Toolkit Using Deep Learning”).
Regarding claim 3, Kowarski does not explicitly teach wherein a first model comprises a neural network iteratively trained to classify mid frequency acoustic data, wherein the first model trained on training set data comprising acoustic samples and label data indicating whether or not the sound or vocalization of a marine mammal is present in the sample.
Bergler teaches wherein a first model comprises a neural network iteratively trained to classify mid frequency acoustic data, wherein the first model trained on training set data comprising acoustic samples and label data indicating whether or not the sound or vocalization of a marine mammal is present in the sample. (Abstract, Page.13, “Data preprocessing and training” and “Convolutional neural network (CNN)”)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate wherein a first model comprises a neural network iteratively trained to classify mid frequency acoustic data, wherein the first model trained on training set data comprising acoustic samples and label data indicating whether or not the sound or vocalization of a marine mammal is present in the sample in order to retrieve sufficient vocalizations for further analysis and to enable an automated annotation procedure of large bioacoustics databases to marine mammal sounds.
Regarding claim 4, Kowarski does not explicitly teach processing the image data before being input to the model to including one or more of: resizing the image data to suit model input requirements; applying tonal noise reduction to the image data; and standardizing the image data to zero mean and unit variance.
Bergler teaches processing the image data before being input to the model to including one or more of: resizing the image data to suit model input requirements; applying tonal noise reduction to the image data; and standardizing the image data to zero mean and unit variance. (Page.13, “Data preprocessing and training”)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate processing the image before being input to the model to including includes one or more of: resizing the image data to suit model input requirements; applying tonal noise reduction to the image data; and standardizing the image data to zero mean and unit variance in order to be able to compare the validation/test set to multiple models, shorter/longer validation and test signals.
Regarding claim 5, Kowarski does not explicitly teach splitting the image data into at least training data and test data, comprising the steps of training the model on the training data and testing the data on the test data to determine acceptable performance of the model.
Bergler teaches splitting the image data into at least training data and test data, comprising the steps of training the model on the training data and testing the data on the test data to determine acceptable performance of the model. (Page.6, “ORCA-SPOT – training/validation/test set metrics”, Page.14, lines 1-11)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate splitting the image data into at least training data and test data, comprising the steps of training the model on the training data and testing the data on the test data to determine acceptable performance of the model in order to enable an automated annotation procedure of large bioacoustics databases to marine mammal sounds.
Claim(s) 6-11 are rejected under 35 U.S.C. 103 as being unpatentable over Kowarski in view of Groenaas, Bergler and Dugan (“North Atlantic Right Whale Acoustic Signal Processing: Part I. Comparison of Machine Learning Recognition Algorithms”)
Regarding claim 6, Kowarski does not explicitly teach wherein the second model comprises a rule based approach operating on features extracted from the image data applied to low frequency acoustic data, wherein optionally the first model is arranged to detect at least dolphin sounds and the second model is arranged to detect at least whale sounds.
Bergler teaches wherein the second model comprises an approach operating on features extracted from the image data applied to frequency acoustic data, wherein optionally the first model is arranged to detect at least dolphin sounds and the second model is arranged to detect at least whale sounds. (Abstract, Page.13, “Data preprocessing and training”)
Dugan teaches a rule based approach operating on features extracted from the image data applied to low frequency acoustic data. (Section IV.D, Section II.A)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate wherein the second model comprises a approach operating on features extracted from the image data applied to frequency acoustic data, wherein optionally the first model is arranged to detect at least dolphin sounds and the second model is arranged to detect at least whale sounds as taught by Bergler in order to retrieve sufficient vocalizations for further analysis and to enable an automated annotation procedure of large bioacoustics databases to marine mammal sounds and further modify Kowarski to incorporate a rule based approach operating on features extracted from the image data applied to low frequency acoustic data as taught by Dugan in order to handle mixed feature sets which contain numeric and non-numeric data.
Regarding claim 7, Kowarski does not explicitly teach processing the image data before being input to the model including one or more of: resizing the image data to suit model input requirements; applying tonal noise reduction to the image data; and filtering the spectrogram to expose acoustic artifacts.
Bergler teaches processing the image data before being input to the model including one or more of: resizing the image data to suit model input requirements; applying tonal noise reduction to the image data; and filtering the spectrogram to expose acoustic artifacts. (Page.13, “Data preprocessing and training”)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate processing the image data before being input to the model including one or more of: resizing the image data to suit model input requirements; applying tonal noise reduction to the image data; and filtering the spectrogram to expose acoustic artifacts as taught by Bergler in order to be able to compare the validation/test set to multiple models, shorter/longer validation and test signals.
Regarding claim 8, Kowarski does not explicitly teach extracting at least one acoustic artifact in the image data; generating a plurality of features from the artifact; and using a rules-based classifier to infer whether a marine mammal is present.
Dugan teaches extracting at least one acoustic artifact in the image data; generating a plurality of features from the artifact; and using a rules-based classifier to infer whether a marine mammal is present. (Section III, Section IV.D)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate extracting at least one acoustic artifact in the image data; generating a plurality of features from the artifact; and using a rules-based classifier to infer whether a marine mammal is present as taught by Dugan in order to handle mixed feature sets which contain numeric and non-numeric data.
Regarding claim 9, Kowarski does not explicitly teach drawing a bounding box around the acoustic artifact, and wherein the plurality of features include one or more of: spatial position, including one or more of centroid, minimum x, minimum y, maximum x, maximum y positions, wherein x position in the time axis and y is the position in the frequency axis, and percentage coverage in relative to its bounding box and the whole image.
Dugan teaches drawing a bounding box around the acoustic artifact, and wherein the plurality of features include one or more of: spatial position, including one or more of centroid, minimum x, minimum y, maximum x, maximum y positions, wherein x position in the time axis and y is the position in the frequency axis, and percentage coverage in relative to its bounding box and the whole image. (Table 1, Section I)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate drawing a bounding box around the acoustic artifact, and wherein the plurality of features include one or more of: spatial position, including one or more of centroid, minimum x, minimum y, maximum x, maximum y positions, wherein x position in the time axis and y is the position in the frequency axis, and percentage coverage in relative to its bounding box and the whole image as taught by Dugan in order to handle mixed feature sets which contain numeric and non-numeric data.
Regarding claim 10, Kowarski teaches standardizing the features using a pre-trained scaler. (Section II.B.2, Table I)
Regarding claim 11, Kowarski teaches based on the extracted features, use of unsupervised learning machine learning techniques to cluster acoustic artifacts into labelled biological and non-biological categories. (Section II.A, Table I, Table II)
Kowarski does not explicitly teach wherein at least one model comprises a high frequency model, the method comprising: filtering the audio data to obtain high frequency data; extract features from the filtered audio data to represent echolocation clicks present in the data; validating the clusters; using the labelled and validated data to train a machine learning model, such as a classification algorithm or a deep learning neural network, to learn the patterns and characteristics of echolocation clicks from the extracted features; and using the model to predict the presence of a marine mammal.
Dugan teaches wherein at least one model comprises a frequency model, the method comprising: filtering the audio data to obtain high frequency data. (Section II.A, Section II.B)
Dugan also teaches extract features from the filtered audio data to represent echolocation clicks present in the data. (Section III, Section IV.D)
Dugan also teaches validating the clusters; using the labelled and validated data to train a machine learning model, such as a classification algorithm or a deep learning neural network, to learn the patterns and characteristics of echolocation clicks from the extracted features. (Section IV.C, Section IV.F, Section I, Table II)
Dugan also teaches using the model to predict the presence of a marine mammal. (Section II.B.3, Section IV.C)
Bergler teaches at least one model comprises a high frequency model. (Page.2)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate wherein at least one model comprises a frequency model, the method comprising: filtering the audio data to obtain high frequency data; extract features from the filtered audio data to represent echolocation clicks present in the data; validating the clusters; using the labelled and validated data to train a machine learning model, such as a classification algorithm or a deep learning neural network, to learn the patterns and characteristics of echolocation clicks from the extracted features; and using the model to predict the presence of a marine mammal as taught by Dugan in order to be able to compare the validation/test set to multiple models, shorter/longer validation and test signals.
Claim(s) 15-19 and 26 are rejected under 35 U.S.C. 103 as being unpatentable over Kowarski in view of Bergler.
Regarding claim 15, Kowarski teaches a system for detection of marine mammals, the system comprising: a processing device and memory holding processor executable instructions. (Abstract, Section II.A, Section IV.A, Fig.1)
Kowarski also teaches an input interface configured to receive input acoustic data from one or more hydrophones(“top-mounted hydrophone”). (Abstract, Section II.A, Fig.1)
Kowarski also teaches a transformation module to sample the acoustic data and transform the sampled acoustic data to time-frequency image data (“The algorithm applies pre-set Fast Fourier Transform (FFT) settings to create a magnitude spectrogram of length “N” seconds; each frequency in the spectrogram is then median-normalized (Table I).”). (Section II.B.2, Table.1)
Kowarski also teaches a model module trained to detect the presence or absence of marine mammal vocalizations in the acoustic data (“The FFT parameters, spectral candidate detection threshold, and contour parameters were determined by a trained analyst (K.K.) during the tuning of automated detectors for real-time monitoring of baleen whales in the Gulf of St Lawrence (Table I).”). (Section II.B.2, Table.1, Table.2)
Kowarski also teaches an output interface to cause a positive prediction of whether or not a marine mammal is present to be displayed to a user by a display device or communicated to a remote user. (Gliders could effectively monitor these buffer regions, reporting when whales are present). (Section I. INTRODUCTION, Section II.B.3, Section IV.C)
Kowarski does not explicitly teach wherein a model of the model module is trained on training data comprises positive examples of marine mammal vocalizations and negative examples of other noise sources to enable the model to learn to distinguish between sources of noise in a marine environment, wherein the noise sources include one or more of boat noise and onboard operation noise, and wherein the processor executable instructions cause the processor to divide the acoustic data into overlapping windows before being input to the model and wherein a positive prediction is based on pooling a plurality of successive predictions of the model that a marine mammal is present based on successive instances of the divided acoustic data.
Bergler teaches wherein a model of the model module is trained on training data comprises positive examples of marine mammal vocalizations and negative examples of other noise sources to enable the model to learn to distinguish between sources of noise in a marine environment, wherein the noise sources include one or more of boat noise and onboard operation noise. (Page.4, “Orchive annotation catalog (OAC)” and “Automatic extracted orchive tape data (AEOTD)”, Page.13, “Convolutional neural network (CNN)” and “Data preprocessing and training”, Figs.5-6, Table 3)
Bergler also teaches wherein the processor executable instructions cause the processor to divide the acoustic data into overlapping windows before being input to the model and wherein a positive prediction is based on pooling a plurality of successive predictions of the model that a marine mammal is present based on successive instances of the divided acoustic data. (Page.5, “Results”, Page.8, “Orchive”, Table 3 Figs.4-6)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate wherein a model of the model module is trained on training data comprises positive examples of marine mammal vocalizations and negative examples of other noise sources to enable the model to learn to distinguish between sources of noise in a marine environment, wherein the noise sources include one or more of boat noise and onboard operation noise, and wherein the processor executable instructions cause the processor to divide the acoustic data into overlapping windows before being input to the model and wherein a positive prediction is based on pooling a plurality of successive predictions of the model that a marine mammal is present based on successive instances of the divided acoustic data as taught by Bergler in order to be able to compare the validation/test set to multiple models, shorter/longer validation and test signals and enable an automated annotation procedure of large bioacoustics databases to marine mammal sounds.
Regarding claim 16, Kowarski teaches wherein the processing device is configured for: inputting the image data to each of at least two different models, respectively arranged to detect marine mammal sounds or vocalizations in different frequency ranges corresponding respectively to different mammal sounds or vocalizations. (Section II.B.2, Section II.B.3, Table.1, Table.2)
Regarding claim 17, Kowarski does not explicitly teach wherein the processing device is configured to wherein a first model comprises a neural network iteratively trained to classify mid frequency acoustic data, wherein the neural network is trained on training set data comprising acoustic samples and label data indicating whether or not the sound or vocalization of a marine mammal is present in the sample.
Bergler teaches wherein the processing device is configured to wherein a first model comprises a neural network iteratively trained to classify mid frequency acoustic data, wherein the neural network is trained on training set data comprising acoustic samples and label data indicating whether or not the sound or vocalization of a marine mammal is present in the sample. (Abstract, Page.13, “Data preprocessing and training” and “Convolutional neural network (CNN)”)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate wherein the processing device is configured to wherein a first model comprises a neural network iteratively trained to classify mid frequency acoustic data, wherein the neural network is trained on training set data comprising acoustic samples and label data indicating whether or not the sound or vocalization of a marine mammal is present in the sample in order to retrieve sufficient vocalizations for further analysis and to enable an automated annotation procedure of large bioacoustics databases to marine mammal sounds.
Regarding claim 18, Kowarski does not explicitly teach wherein the processing device is configured to process the image data to include one or more of: resizing the image data to suit model input requirements; applying tonal noise reduction to the image data; and standardizing the image data to zero mean and unit variance.
Bergler teaches wherein the processing device is configured to process the image data to include one or more of: resizing the image data to suit model input requirements; applying tonal noise reduction to the image data; and standardizing the image data to zero mean and unit variance. (Page.13, “Data preprocessing and training”)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate wherein the processing device is configured to process the image data to include one or more of: resizing the image data to suit model input requirements; applying tonal noise reduction to the image data; and standardizing the image data to zero mean and unit variance in order to be able to compare the validation/test set to multiple models, shorter/longer validation and test signals.
Regarding claim 19, Kowarski does not explicitly teach wherein the processing device is configured to split the image data into at least training data and test data, comprising the steps of training the model on the training data and testing the data on the test data to determine acceptable performance of the model.
Bergler teaches wherein the processing device is configured to split the image data into at least training data and test data, comprising the steps of training the model on the training data and testing the data on the test data to determine acceptable performance of the model. (Page.6, “ORCA-SPOT – training/validation/test set metrics”, Page.14, lines 1-11)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate wherein the processing device is configured to split the image data into at least training data and test data, comprising the steps of training the model on the training data and testing the data on the test data to determine acceptable performance of the model in order to enable an automated annotation procedure of large bioacoustics databases to marine mammal sounds.
Regarding claim 26, Kowarski teaches wherein the processing device is configured to automatically or in response to accepting user input, ceasing at least one on board marine activity if the prediction indicates the presence of a marine mammal, wherein the acoustic data and/or image data and prediction are displayed to a user for validation, and optionally the method comprises receiving user input indicating validation of the prediction, wherein the user validation overrides the decision to cease the marine seismic activity. (Section I, Section II.B.4)
Claim(s) 20-25 are rejected under 35 U.S.C. 103 as being unpatentable over Kowarski in view of Bergler and Dugan.
Regarding claim 20, Kowarski does not explicitly teach wherein the processing device is configured to wherein the second model comprises a rule based approach operating on features extracted from the image data applied to low frequency acoustic data, wherein optionally the first model is arranged to detect at least dolphin sounds and the second model is arranged to detect at least whale sounds.
Bergler teaches wherein the processing device is configured to wherein the second model comprises an approach operating on features extracted from the image data applied to frequency acoustic data, wherein optionally the first model is arranged to detect at least dolphin sounds and the second model is arranged to detect at least whale sounds. (Abstract, Page.13, “Data preprocessing and training”)
Dugan teaches a rule based approach operating on features extracted from the image data applied to low frequency acoustic data. (Section IV.D, Section II.A)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate wherein the processing device is configured to wherein the second model comprises a approach operating on features extracted from the image data applied to frequency acoustic data, wherein optionally the first model is arranged to detect at least dolphin sounds and the second model is arranged to detect at least whale sounds as taught by Bergler in order to retrieve sufficient vocalizations for further analysis and to enable an automated annotation procedure of large bioacoustics databases to marine mammal sounds and further modify Kowarski to incorporate a rule based approach operating on features extracted from the image data applied to low frequency acoustic data as taught by Dugan in order to handle mixed feature sets which contain numeric and non-numeric data.
Regarding claim 21, Kowarski does not explicitly teach wherein the processing device is configured to process the image data to include one or more of: resizing the image data to suit model input requirements; applying tonal noise reduction to the image data; and filtering the spectrogram to expose acoustic artifacts.
Bergler teaches wherein the processing device is configured to process the image data to include one or more of: resizing the image data to suit model input requirements; applying tonal noise reduction to the image data; and filtering the spectrogram to expose acoustic artifacts. (Page.13, “Data preprocessing and training”)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate wherein the processing device is configured to process the image data to include one or more of: resizing the image data to suit model input requirements; applying tonal noise reduction to the image data; and filtering the spectrogram to expose acoustic artifacts as taught by Bergler in order to be able to compare the validation/test set to multiple models, shorter/longer validation and test signals.
Regarding claim 22, Kowarski does not explicitly teach wherein the processing device is configured to: extract at least one acoustic artifact in the image data; generate a plurality of features from the artifact; and use a rules-based classifier to infer whether a marine mammal is present.
Dugan teaches wherein the processing device is configured to: extract at least one acoustic artifact in the image data; generate a plurality of features from the artifact; and use a rules-based classifier to infer whether a marine mammal is present. (Section III, Section IV.D)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate wherein the processing device is configured to: extract at least one acoustic artifact in the image data; generate a plurality of features from the artifact; and use a rules-based classifier to infer whether a marine mammal is present as taught by Dugan in order to handle mixed feature sets which contain numeric and non-numeric data.
Regarding claim 23, Kowarski does not explicitly teach wherein the processing device is configured to draw a bounding box around the acoustic artifact, and wherein the plurality of features include one or more of: spatial position, including one or more of centroid, minimum x, minimum y, maximum x, maximum y positions, wherein x position in the time axis and y is the position in the frequency axis, and percentage coverage in relative to its bounding box and the whole image.
Dugan teaches wherein the processing device is configured to draw a bounding box around the acoustic artifact, and wherein the plurality of features include one or more of: spatial position, including one or more of centroid, minimum x, minimum y, maximum x, maximum y positions, wherein x position in the time axis and y is the position in the frequency axis, and percentage coverage in relative to its bounding box and the whole image. (Table 1, Section I)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate wherein the processing device is configured to draw a bounding box around the acoustic artifact, and wherein the plurality of features include one or more of: spatial position, including one or more of centroid, minimum x, minimum y, maximum x, maximum y positions, wherein x position in the time axis and y is the position in the frequency axis, and percentage coverage in relative to its bounding box and the whole image as taught by Dugan in order to handle mixed feature sets which contain numeric and non-numeric data.
Regarding claim 24, Kowarski teaches wherein the processing device is configured to standardize the features using a pre-trained scaler. (Section II.B.2, Table I)
Regarding claim 25, Kowarski teaches based on the extracted features, use of unsupervised learning machine learning techniques to cluster acoustic artifacts into labelled biological and non-biological categories. (Section II.A, Table I, Table II)
Kowarski does not explicitly teach wherein the processing device is configured to wherein at least one model comprises a high frequency model, the method comprising: filtering the audio data to obtain high frequency data; extract features from the filtered audio data to represent echolocation clicks present in the data; validating the clusters; using the labelled and validated data to train a machine learning model, such as a classification algorithm or a deep learning neural network, to learn the patterns and characteristics of echolocation clicks from the extracted features; and using the model to predict the presence of a marine mammal.
Dugan teaches wherein the processing device is configured to wherein at least one model comprises a frequency model, the method comprising: filtering the audio data to obtain high frequency data. (Section II.A, Section II.B)
Dugan also teaches extract features from the filtered audio data to represent echolocation clicks present in the data. (Section III, Section IV.D)
Dugan also teaches validating the clusters; using the labelled and validated data to train a machine learning model, such as a classification algorithm or a deep learning neural network, to learn the patterns and characteristics of echolocation clicks from the extracted features. (Section IV.C, Section IV.F, Section I, Table II)
Dugan also teaches using the model to predict the presence of a marine mammal. (Section II.B.3, Section IV.C)
Bergler teaches at least one model comprises a high frequency model. (Page.2)
It would have been obvious to one having ordinary skill in the art before the effective filling date to have modified Kowarski to incorporate wherein the processing device is configured to wherein at least one model comprises a frequency model, the method comprising: filtering the audio data to obtain high frequency data; extract features from the filtered audio data to represent echolocation clicks present in the data; validating the clusters; using the labelled and validated data to train a machine learning model, such as a classification algorithm or a deep learning neural network, to learn the patterns and characteristics of echolocation clicks from the extracted features; and using the model to predict the presence of a marine mammal as taught by Dugan in order to be able to compare the validation/test set to multiple models, shorter/longer validation and test signals.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ABDALLAH ABULABAN whose telephone number is (571)272-4755. The examiner can normally be reached Monday - Friday 7:00am-3:00pm EST.
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, Isam Alsomiri can be reached at 571-272-6970. 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.
/ABDALLAH ABULABAN/Examiner, Art Unit 3645