Detailed Action
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendment
Applicant’s submission filed 07/22/2025 includes changes to the claims, remarks and arguments related to the previous rejection. The above have been entered and considered. Claims 1-18 are currently pending.
Response to Arguments
With regard to the drawing objection:
Applicant has amended Figures 6-8 to list the units on the depicted graphs. Applicant has submitted a legible Figure 12. The drawings are accepted and the drawing objection is withdrawn.
With regard to the 112(b) rejection:
Applicant has not amended or argued the rejections to resolve the clarity of the claims. The 112(b) rejection of the claims is maintained.
With regard to the claim interpretations:
Applicant has not traversed the Examiner’s structural interpretations of the nonce terms plus functional language. The interpretations stand as a record.
With regard to the 103 rejection:
Applicant has amended Claims 1 & 10 to add new limitations that require search and consideration.
at least one artificial intelligence based signal processing unit comprising at least one processor at least one artificial intelligence and at least one software platform configured to process the at least one audible sound signature and select at least one auditory threshold that corresponds to at least one physical attribute of the at least one railroad track including a type of spike used, a substrate material for the at least one railroad track, a track age or combinations of the above.
Applicant’s arguments and/or amendments with regard to Claims 1-18 have been considered in light of the previous references. The arguments and amended claims do not overcome the prior art at the time of the filing of the invention. Upon further consideration, a new ground(s) of rejection is made in view of a new combination of the prior references of Taek and Poudel and in view of the new reference of Takuro.
Information Disclosure Statement
An information disclosure statement has not been received. If the applicant is aware of any prior art or any other co-pending applications not already of record, he/she is reminded of his/her duty under 37 CFR 1.56 to disclose the same.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
Rejection under 35 U.S.C. § 112(a) – Lack of Enablement
Claims 1-18 are rejected under 35 U.S.C. 112(a), as failing to comply with the enablement requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to enable one skilled in the art to which it pertains, or with which it is most nearly connected, to make and/or use the invention.
Claims 1 &10 are rejected under 35 U.S.C. § 112(a) because the specification, while being enabling for at least one artificial intelligence based signal processing unit comprising at least one processor at least one artificial intelligence and at least one software platform configured to process the at least one audible sound to determine a condition of a broken rail spike, does not reasonably enable one of ordinary skill in the art to make and use the claimed invention commensurate in scope with the claims.
Specifically, the claims recite “using an artificial intelligence system to determine at least one physical attribute of the at least one railroad track including a type of spike used, a substrate material for the at least one railroad track, a track age or combinations of the above.” However, the specification fails to disclose any details regarding the structure, architecture, training methodology, input preprocessing, feature extraction, or decision-making process of the artificial intelligence system. The disclosure does not identify the type of model (e.g., neural network, decision tree, support vector machine), the nature of the training data, or the algorithmic steps necessary to achieve the claimed determination.
In the absence of such information, undue experimentation would be required for one of ordinary skill in the art to implement the claimed AI-based determination. See MPEP § 2164 and In re Wands, 858 F.2d 731 (Fed. Cir. 1988) (listing factors relevant to undue experimentation). The mere statement that “an AI model determines the condition” is a result-oriented limitation that does not teach how to achieve the claimed result without extensive trial and error.
Therefore, the claims are not enabled commensurate with their scope.
Rejection under 35 U.S.C. § 112(a) – Lack of Written Description
Claims 1-18 are rejected under 35 U.S.C. § 112(a) because the specification does not reasonably convey to one of ordinary skill in the art that the inventor had possession of the claimed invention at the time of filing.
Claims 1 & 10 recite “at least one artificial intelligence based signal processing unit comprising at least one processor at least one artificial intelligence and at least one software platform configured to process the at least one audible sound signature and select at least one auditory threshold that corresponds to at least one physical attribute of the at least one railroad track including a type of spike used, a substrate material for the at least one railroad track, a track age or combinations of the above“. The specification, however, provides only a generic statement that “an AI model processes the data to determine the condition,” without describing the model’s architecture, training methodology, input features, or decision-making process.
The disclosure lacks any identification of the type of AI model, the nature of the training data, or the algorithmic steps necessary to achieve the claimed determination. As such, the specification does not demonstrate that the inventor had possession of the specific AI-based determination recited in the claims, but instead merely states a desired result. See MPEP § 2163 and Ariad Pharm., Inc. v. Eli Lilly & Co., 598 F.3d 1336 (Fed. Cir. 2010) (en banc) (“A description that merely renders the invention obvious does not satisfy the written description requirement.”).
Wands Factors
There are many factors to be considered when determining whether there is sufficient evidence to support a determination that a disclosure does not satisfy the enablement requirement and whether any necessary experimentation is “undue.” In this case, the relevant Wand factors Examiner has considered are :
2164.01(a) Undue Experimentation Factors [R-01.2024]
(A) The breadth of the claims;
(B) The nature of the invention;
(C) The state of the prior art;
(D) The level of one of ordinary skill;
(E) The level of predictability in the art;
(F) The amount of direction provided by the inventor;
(G) The existence of working examples; and
(H) The quantity of experimentation needed to make or use the invention based on the content of the disclosure.
In Applicant’s case:
Nature of the Invention: AI-related models requires particular parameters or training data that +lead to varying determinations without disclosure.
Amount of Disclosure Provided: The specification merely cites the problem to be solved as aspirational using artificial Intelligence without detailing the training methodologies, hyperparameters, or optimization techniques used to reach that output.
Quantity of Experimentation: Applicant has place on the PHOSITA the entire design, coding and exhaustive trial and error experimentation.
Consistent with office policy, Examiner has weighed all the evidence for and against enablement of this invention and has concluded based on guidance provided by the MPEP and case law (including the Wands factors) that there is not enough evidence in favor of the enablement and written description.
Applicant may submit factual affidavits under 37 CFR 1.132 or cite references to show what one skilled in the art knew at the time of filing the application. A declaration or affidavit is, itself, evidence that must be considered. The weight to give a declaration or affidavit will depend upon the amount of factual evidence the declaration or affidavit contains to support the conclusion of enablement. In re Buchner, 929 F.2d 660, 661, 18 USPQ2d 1331, 1332 (Fed. Cir. 1991) (“expert' s opinion on the ultimate legal conclusion must be supported by something more than a conclusory statement”); cf. In re Alton, 76 F.3d 1168, 1174, 37 USPQ2d 1578, 1583 (Fed. Cir. 1996) (declarations relating to the written description requirement should have been considered)”.
All dependent claims are rejected for their dependence on a rejected base claim.
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims
particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
Claims 1-18 are rejected under 35 U.S.C. 112(b), as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor regards as the invention.
Claims 1, 5-6, 10 & 14-15 recite the limitation ”signal processing unit analyzes the at least one audible sound signature and at least one vibration characteristic length” where it is unclear as to what signal analysis is required to meet the “one vibration characteristic length”. Examiner looks to the specification [0066] standard signal analysis is applied without identifying standard wave analysis terms. Examiner interprets “one vibration characteristic length” as a wavelength of the vibration signal.
Claims 1 & 10 recite the musical terms of “pitch” and “tone“, which is unclear as to what signal analysis terms are performed on the received vibration signal. Examiner interprets pitch as frequency analysis performed in the frequency domain and tone as intensity in time domain expressed as decibels or spectral density performed in the frequency domain.
Claims 1 & 10 recite a limitation that corresponds to “at least one physical attribute of the at least one railroad track including a type of spike used, a substrate material for the at least one railroad track, a track age or combinations of the above” which is unclear as to what meets the limitation as “the at least one physical attribute of the at least one railroad track” is the genus and it seems to be the claimed requirement with the exemplary language listing species cited as “the including a type of spike used, a substrate material for the at least one railroad track, a track age or combinations of the above”. The species set in the exemplary language should be distinctly claimed in dependent claims.
Claims 4 & 13 cite the known acronym AI, however for clarity Artificial Intelligence (AI) should be introduced when first cited.
Claims 5-7 & 14-16 cite an AI processing unit and then a processing step. To distinctly claim a software processing step with computer implementation the claim should cite a processor configured to perform the software step or in this case an AI based signal processing unit configured to perform...
All dependent claims are rejected for their dependence on a rejected base claim.
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. - An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f), is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f):
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f). The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f), is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f). The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f), is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA , except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f), except as otherwise indicated in an Office action.
The cited limitations have been interpreted under 35 U.S.C. 112(f), because they use a generic placeholder “means” coupled with functional language without reciting sufficient structure to achieve the function. Furthermore, the generic placeholder is not preceded by a structural modifier.
Since the claim limitations invoke 35 U.S.C. 112(f), the claims have been interpreted to cover the corresponding structure described in the specification that achieves the claimed function, and equivalents thereof.
The following table provides the generic place holder, functional language and the review and citation of the specification that shows that the following appears to be the corresponding structure described in the specification for the 35 U.S.C. 112(f):
Claim(s)
Generic place holder language or “means”
Functional Language
Interpretation from written description
1-2 & 10-11
one hitting mechanism for
performing at least one strike to at least one railroad spike
[0006: one hitting mechanism may comprise at least one hammer]
If applicant wishes to provide further explanation or dispute the examiner’s interpretation of the corresponding structure, applicant must identify the corresponding structure with reference to the specification by page and line number, and to the drawing, if any, by reference characters in response to this Office action.
If applicant does not intend to have the claim limitation(s) treated under 35 U.S.C. 112(f) applicant may amend the claims so that they will clearly not invoke 35 U.S.C. 112(f), or present a sufficient showing that the claim recite sufficient structure, material, or acts for performing the claimed function to preclude application of 35 U.S.C. 112(f).
For more information, see MPEP § 2173 etseq. and Supplementary Examination Guidelines for Determining Compliance With 35 U.S.C. 112 and for Treatment of Related Issues in Patent Applications, 76 FR 7162, 7167 (Feb. 9, 2011).
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.
Claims 1-18 are rejected under 35 U.S.C. 103 as being unpatentable over Taek (KR 101027910; “Taek” translation provided for citations) in view of Poudel (US 20220365033; “Poudel”) and in further view of Takuro (JP 2018013348: “Takuro” translation provided for citations).
Claim 1. Taek discloses a railroad inspection system (Figs. 4 & 9: bogie 210 with inspection elements)[0029: a method for automatically inspecting a railway sleeper using an acoustic signal is provided, which can simply and accurately inspect cracks in a railway sleeper while it is being transported on a bogie] comprising: a mobile platform (210 bogie/trolley) containing at least one hitting mechanism (140 hammer) for performing at least one strike to at least one railroad element (310)[0031: while being transported along the bogie (210), strikes a railway sleeper (310) by a hammer (140) as a striking part near a railway rail (300L, 300R) and measures sound resulting from vibration of the railway sleeper (310) to inspect for cracks]; wherein the at least one strike generates at least one audible sound signature (310) [0051: At this time, the comparative analysis unit (113-3) can determine whether cracks have occurred in the railroad sleeper based on the results of time domain analysis, frequency domain analysis, or wavelet analysis] and at least one vibration characteristic length for the at least one railroad element (310) [0070] & [0072: time domain analysis result in the railway sleeper crack inspection method according to the embodiment of the present invention shows that in the case of a defective product, the waveform length is relatively long and the amplitude is large] and at least one audio detection system (150 sound acquisition unit microphone)[0038: the striking unit (140) may be implemented as a hammer, and the sound acquisition unit (150) may be implemented as a microphone] captures the at least one audible sound signature [0035: built-in crack analysis unit that compares and analyzes the impact waveform and the acoustic waveform according to a preset signal processing program to determine whether a crack has occurred in the railway rail (300L, 300R)] and at least one vibration characteristic length [0070] & [0072: time domain analysis result in the railway sleeper crack inspection method according to the embodiment of the present invention shows that in the case of a defective product, the waveform length is relatively long and the amplitude is large]; wherein the at least one audible sound signature and at least one vibration characteristic [0070] & [0072: time domain analysis result in the railway sleeper crack inspection method according to the embodiment of the present invention shows that in the case of a defective product, the waveform length is relatively long and the amplitude is large] generate at least one pitch (e.g. frequency) for the at least one railroad element [0073: As shown in a) to c) of FIG. 13, in the frequency domain analysis result of the railway sleeper crack inspection method according to the embodiment of the present invention, it can be seen that in the case of a defective product, the level in the 1 kHz to 10 kHz band is relatively large. Here, the horizontal axis represents frequency and the vertical axis represents spectral density] and a tone (e.g. spectral density) of the at least one pitch indicates the element is broken [0073: As shown in a) to c) of FIG. 13, in the frequency domain analysis result of the railway sleeper crack inspection method according to the embodiment of the present invention, it can be seen that in the case of a defective product, the level in the 1 kHz to 10 kHz band is relatively large. Here, the horizontal axis represents frequency and the vertical axis represents spectral density]; wherein the system (Figs. 4 & 9: bogie 210 with inspection elements) is further configured to be non-destructive to the at least one element being tested (Fig. 9: track 300 with wheels 230 for moving bogie 210) [0047: Thereafter, the bogie (210) is moved to align the impact position according to the acquired image, and furthermore, if no cracks occur in the railroad sleeper (310), the bogie (210) is automatically moved to the next position by the bogie moving unit (220)]; and the system (Figs. 4 & 9 bogie 210 with inspection elements) operates on at least one railroad track (300) and provides results in real-time as the system traverses the at least one railroad track [Page 5 last ¶: According to the present invention, cracks in railway sleepers can be inspected at high speed while being mounted on a bogie and automatically transported]. Taek further discloses a processing unit (30) configured to process the at least one audible sound signature [0060] and select at least one auditory threshold that corresponds to at least one physical attribute of the at least one railroad track [0068: , whether cracks have occurred in the railway sleeper can be determined based on the results of time domain analysis, frequency domain analysis, or wavelet analysis]. Taek does not explicitly disclose:
1) The railroad element is a railroad spike
2) at least one artificial intelligence based signal processing unit comprising at least one processor at least one artificial intelligence and at least one software platform configured to process the at least one audible sound signature and select at least one auditory threshold.
With regard to 1) Poudel teaches a method for detecting breaks or defects in railroad spikes [Abstract]. Poudel further teaches an acoustic testing of a railroad element that is a railroad spike [0005: The operator taps each spike 10 with a hammer and attempts to distinguish different sounds (analogous to tap testing) to determine whether a spike is intact or broken].
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to use Poudel’s need for acoustic hammer tap testing of railroad spikes as a testing arrangement for Taek’s acoustic hammer tap testing of a railroad element because expanding inspection of railroad spikes and sleepers improves safe operation of train traffic along inspected and maintained stable rail lines [Poudel 0007].
With regard to 2) Takuro teaches [0005: a sound inspection system comprising: an input unit that receives sound data, which is sound data generated when a structure is struck; an extraction unit that extracts feature quantities from the sound data input to the input unit; a selection unit that selects predetermined feature quantities from the feature quantities extracted by the extraction unit as selected feature quantities; a learning unit that learns selected feature quantities based on target sound data from the sound data input to the input unit from the selected feature quantities selected by the selection unit and generates a learner] at least one AI based signal processing unit (523 & 524) [0027: the learning unit 523 learns selected features from among the selected features selected by the selection unit 522, based on the target impact sound data among the impact sound data input to the impact sound input unit 421, and generates a learner. Furthermore, the determination unit 524 uses the selected feature quantities selected by the selection unit 522, which are based on the target impact sound data from the impact sound data input to the impact sound input unit 421, and the learner generated by the learning unit 523, to determine whether or not there is an abnormality in the structure] & [0013: the analysis device 50 may learn the impact sound data and judgment results through machine learning, for example, using Jubatus, a machine learning framework e.g. known AI/MS platform where machine learning is a subset of AI] comprising at least one processor (50)[0026] at least one artificial intelligence [0013] and at least one software platform (523) configured to process the at least one audible sound signature [0040: the impact sound input unit 421 receives impact sound data along with a label indicating whether or not there is an abnormality in the structure] and select at least one auditory threshold [0041-0042: At this time, the learning unit 523 generates an anomaly detector that calculates the degree of discrepancy between the impact sound data to be judged and the impact sound data to be learned. The degree of deviation is, for example, the distance between data points. In this case, if the degree of discrepancy between the impact sound data to be judged and the impact sound data to be learned is greater than or equal to a predetermined threshold, the determination unit 524 determines that the impact sound data to be judged originated from a structure with an abnormality. An anomaly detector is an example of a learning device. Furthermore, the learning unit 523 may use the Local Outlier Factor as an unsupervised machine learning algorithm].
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to use Takuro’s Machine Learning of abnormalities in a structure from impact sound data with Taek’s, as modified, sound impact data because machine learning improves accuracy in determining abnormalities in a structure by learning to eliminate noise and identify abnormal signatures [Takuro 0052-0053].
Claim 10. Taek discloses a method for inspecting railroad element (Figs. 4 & 9: bogie 210 with inspection elements)[0029: a method for automatically inspecting a railway sleeper using an acoustic signal is provided, which can simply and accurately inspect cracks in a railway sleeper while it is being transported on a bogie] comprising: moving a non-destructive railroad element inspection system (210 bogie/trolley) [0029: a method for automatically inspecting a railway sleeper using an acoustic signal is provided, which can simply and accurately inspect cracks in a railway sleeper while it is being transported on a bogie along at least one railroad track (300); striking at least one railroad element (310)[0031: while being transported along the bogie (210), strikes a railway sleeper (310) by a hammer (140) as a striking part near a railway rail (300L, 300R) and measures sound resulting from vibration of the railway sleeper (310) to inspect for cracks] positioned along the railroad track (300) with at least one hitting mechanism (140 hammer); generating via the at least one strike at least one audible sound signature (310) [0051: At this time, the comparative analysis unit (113-3) can determine whether cracks have occurred in the railroad sleeper based on the results of time domain analysis, frequency domain analysis, or wavelet analysis] and at least one vibration characteristic length for the at least one railroad element (310) [0070] & [0072: time domain analysis result in the railway sleeper crack inspection method according to the embodiment of the present invention shows that in the case of a defective product, the waveform length is relatively long and the amplitude is large] and at least one audio detection system (150 sound acquisition unit microphone)[0038: the striking unit (140) may be implemented as a hammer, and the sound acquisition unit (150) may be implemented as a microphone] and at least one audio detection system (150) captures the at least one audible sound signature [0035: built-in crack analysis unit that compares and analyzes the impact waveform and the acoustic waveform according to a preset signal processing program to determine whether a crack has occurred in the railway rail (300L, 300R)] and at least one vibration characteristic length [0070] & [0072: time domain analysis result in the railway sleeper crack inspection method according to the embodiment of the present invention shows that in the case of a defective product, the waveform length is relatively long and the amplitude is large]; wherein the at least one audible sound signature [0035] and at least one vibration characteristic [0070] & [0072: time domain analysis result in the railway sleeper crack inspection method according to the embodiment of the present invention shows that in the case of a defective product, the waveform length is relatively long and the amplitude is large] generate at least one pitch (e.g. frequency) for the at least one railroad element [0073: FIG. 13, in the frequency domain analysis result of the railway sleeper crack inspection method according to the embodiment of the present invention, it can be seen that in the case of a defective product, the level in the 1 kHz to 10 kHz band is relatively large. Here, the horizontal axis represents frequency and the vertical axis represents spectral density] and a tone (e.g. spectral density) of the at least one pitch indicates the element is broken [0073: As shown in a) to c) of FIG. 13, in the frequency domain analysis result of the railway sleeper crack inspection method according to the embodiment of the present invention, it can be seen that in the case of a defective product, the level in the 1 kHz to 10 kHz band is relatively large. Here, the horizontal axis represents frequency and the vertical axis represents spectral density]; where configuring the system (Figs. 4 & 9: bogie 210 with inspection elements) to be non-destructive to the at least one element being tested (Fig. 9: track 300 with wheels 230 for moving bogie 210) [0047: Thereafter, the bogie (210) is moved to align the impact position according to the acquired image, and furthermore, if no cracks occur in the railroad sleeper (310), the bogie (210) is automatically moved to the next position by the bogie moving unit (220)]; and operating system (Figs. 4 & 9 bogie 210 with inspection elements)on at least one railroad track (300) and providing results in real-time as the system traverses the at least one railroad track [Page 5 last ¶: According to the present invention, cracks in railway sleepers can be inspected at high speed while being mounted on a bogie and automatically transported].
Taek further discloses a processing unit (30) configured to process the at least one audible sound signature [0060] and select at least one auditory threshold that corresponds to at least one physical attribute of the at least one railroad track [0068: , whether cracks have occurred in the railway sleeper can be determined based on the results of time domain analysis, frequency domain analysis, or wavelet analysis]. Taek does not explicitly disclose:
1) The railroad element is a railroad spike.
2) at least one artificial intelligence based signal processing unit comprising at least one processor at least one artificial intelligence and at least one software platform configured to process the at least one audible sound signature and select at least one auditory threshold.
With regard to 1) Poudel teaches a method for detecting breaks or defects in railroad spikes [Abstract]. Poudel further teaches an acoustic testing of a railroad element that is a railroad spike [0005: The operator taps each spike 10 with a hammer and attempts to distinguish different sounds (analogous to tap testing) to determine whether a spike is intact or broken].
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to use Poudel’s need for acoustic hammer tap testing of railroad spikes as a testing arrangement for Taek’s acoustic hammer tap testing of a railroad element because expanding inspection of railroad spikes and sleepers improves safe operation of train traffic along inspected and maintained stable rail lines [Poudel 0007].
With regard to 2) Takuro teaches [0005: a sound inspection system comprising: an input unit that receives sound data, which is sound data generated when a structure is struck; an extraction unit that extracts feature quantities from the sound data input to the input unit; a selection unit that selects predetermined feature quantities from the feature quantities extracted by the extraction unit as selected feature quantities; a learning unit that learns selected feature quantities based on target sound data from the sound data input to the input unit from the selected feature quantities selected by the selection unit and generates a learner] at least one AI based signal processing unit (523 & 524) [0027: the learning unit 523 learns selected features from among the selected features selected by the selection unit 522, based on the target impact sound data among the impact sound data input to the impact sound input unit 421, and generates a learner. Furthermore, the determination unit 524 uses the selected feature quantities selected by the selection unit 522, which are based on the target impact sound data from the impact sound data input to the impact sound input unit 421, and the learner generated by the learning unit 523, to determine whether or not there is an abnormality in the structure] & [0013: the analysis device 50 may learn the impact sound data and judgment results through machine learning, for example, using Jubatus, a machine learning framework e.g. known AI/MS platform where machine learning is a subset of AI] comprising at least one processor (50)[0026] at least one artificial intelligence [0013] and at least one software platform (523) configured to process the at least one audible sound signature [0040: the impact sound input unit 421 receives impact sound data along with a label indicating whether or not there is an abnormality in the structure] and select at least one auditory threshold [0041-0042: At this time, the learning unit 523 generates an anomaly detector that calculates the degree of discrepancy between the impact sound data to be judged and the impact sound data to be learned. The degree of deviation is, for example, the distance between data points. In this case, if the degree of discrepancy between the impact sound data to be judged and the impact sound data to be learned is greater than or equal to a predetermined threshold, the determination unit 524 determines that the impact sound data to be judged originated from a structure with an abnormality. An anomaly detector is an example of a learning device. Furthermore, the learning unit 523 may use the Local Outlier Factor as an unsupervised machine learning algorithm].
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to use Takuro’s Machine Learning of abnormalities in a structure from impact sound data with Taek’s, as modified, sound impact data because machine learning improves accuracy in determining abnormalities in a structure by learning to eliminate noise and identify abnormal signatures [Takuro 0052-0053].
Claims 2 & 11. Dependent on the respective system of claim 1 and method of claim 10. Taek further discloses the at least one hitting mechanism (140) comprises at least one hammer [0077: According to Fig. 15b examined above, it can be seen that the impact part (140) of the railway sleeper crack inspection device (100) is implemented as an impact hammer
Claims 3 & 12. Dependent on the respective system of claim 1 and method of claim 10. Taek further discloses the audio detection system (150) comprises at least one microphone [0080: In addition, it can be seen that the above-mentioned response acquisition unit (150) is implemented with a microphone].
Claims 4 & 13. Dependent on the respective system of claim 1 and method of claim 10. Taek further discloses a processor (110) [0076: includes a PC (110), an image acquisition unit (120), a driving unit (130), a striking unit (140), and an audio acquisition unit (150). As described above, the PC (110) may include a control unit (111), a display (112), and a crack analysis unit (113). Taek, as modified, does not explicitly disclose:
at least one AI based signal processing unit.
Takuro teaches [0005: a sound inspection system comprising: an input unit that receives sound data, which is sound data generated when a structure is struck; an extraction unit that extracts feature quantities from the sound data input to the input unit; a selection unit that selects predetermined feature quantities from the feature quantities extracted by the extraction unit as selected feature quantities; a learning unit that learns selected feature quantities based on target sound data from the sound data input to the input unit from the selected feature quantities selected by the selection unit and generates a learner] at least one AI based signal processing unit (523 & 524) [0027: the learning unit 523 learns selected features from among the selected features selected by the selection unit 522, based on the target impact sound data among the impact sound data input to the impact sound input unit 421, and generates a learner. Furthermore, the determination unit 524 uses the selected feature quantities selected by the selection unit 522, which are based on the target impact sound data from the impact sound data input to the impact sound input unit 421, and the learner generated by the learning unit 523, to determine whether or not there is an abnormality in the structure] & [0013: the analysis device 50 may learn the impact sound data and judgment results through machine learning, for example, using Jubatus, a machine learning framework e.g. known AI/MS platform where machine learning is a subset of AI] comprising at least one processor (50)[0026] at least one artificial intelligence [0013] and at least one software platform (523) configured to process the at least one audible sound signature [0040: the impact sound input unit 421 receives impact sound data along with a label indicating whether or not there is an abnormality in the structure] and select at least one auditory threshold [0041-0042: At this time, the learning unit 523 generates an anomaly detector that calculates the degree of discrepancy between the impact sound data to be judged and the impact sound data to be learned. The degree of deviation is, for example, the distance between data points. In this case, if the degree of discrepancy between the impact sound data to be judged and the impact sound data to be learned is greater than or equal to a predetermined threshold, the determination unit 524 determines that the impact sound data to be judged originated from a structure with an abnormality. An anomaly detector is an example of a learning device. Furthermore, the learning unit 523 may use the Local Outlier Factor as an unsupervised machine learning algorithm].
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to use Takuro’s Machine Learning of abnormalities in a structure from impact sound data with Taek’s, as modified, sound impact data because machine learning improves accuracy in determining abnormalities in a structure by learning to eliminate noise and identify abnormal signatures [Takuro 0052-0053].
Claim 5. Dependent on the respective system of claim 4. Taek further discloses the at least one audible sound signature (310) [0051: At this time, the comparative analysis unit (113-3) can determine whether cracks have occurred in the railroad sleeper based on the results of time domain analysis, frequency domain analysis, or wavelet analysis] and at least one vibration characteristic length for the at least one railroad element [0070] & [0072: time domain analysis result in the railway sleeper crack inspection method according to the embodiment of the present invention shows that in the case of a defective product, the waveform length is relatively long and the amplitude is large] and at least one audio detection system (150 sound acquisition unit microphone). Taek, as modified, does not explicitly disclose:
the at least one AI based signal processing unit analyzes the at least one audible sound signature
Takuro teaches [0005: a sound inspection system comprising: an input unit that receives sound data, which is sound data generated when a structure is struck; an extraction unit that extracts feature quantities from the sound data input to the input unit; a selection unit that selects predetermined feature quantities from the feature quantities extracted by the extraction unit as selected feature quantities; a learning unit that learns selected feature quantities based on target sound data from the sound data input to the input unit from the selected feature quantities selected by the selection unit and generates a learner] at least one AI based signal processing unit (523 & 524) [0027: the learning unit 523 learns selected features from among the selected features selected by the selection unit 522, based on the target impact sound data among the impact sound data input to the impact sound input unit 421, and generates a learner. Furthermore, the determination unit 524 uses the selected feature quantities selected by the selection unit 522, which are based on the target impact sound data from the impact sound data input to the impact sound input unit 421, and the learner generated by the learning unit 523, to determine whether or not there is an abnormality in the structure] & [0013: the analysis device 50 may learn the impact sound data and judgment results through machine learning, for example, using Jubatus, a machine learning framework e.g. known AI/MS platform where machine learning is a subset of AI] comprising at least one processor (50)[0026] at least one artificial intelligence [0013] and at least one software platform (523) configured to process the at least one audible sound signature [0040: the impact sound input unit 421 receives impact sound data along with a label indicating whether or not there is an abnormality in the structure] and select at least one auditory threshold [0041-0042: At this time, the learning unit 523 generates an anomaly detector that calculates the degree of discrepancy between the impact sound data to be judged and the impact sound data to be learned. The degree of deviation is, for example, the distance between data points. In this case, if the degree of discrepancy between the impact sound data to be judged and the impact sound data to be learned is greater than or equal to a predetermined threshold, the determination unit 524 determines that the impact sound data to be judged originated from a structure with an abnormality. An anomaly detector is an example of a learning device. Furthermore, the learning unit 523 may use the Local Outlier Factor as an unsupervised machine learning algorithm].
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to use Takuro’s Machine Learning of abnormalities in a structure from impact sound data with Taek’s, as modified, sound impact data because machine learning improves accuracy in determining abnormalities in a structure by learning to eliminate noise and identify abnormal signatures [Takuro 0052-0053].
Claims 6 & 15. Dependent on the respective system of claim 5 and method of claim 14. Taek further discloses the unit filters noise [0053: This Fourier transform method is particularly useful for analyzing electrical signals that are a mixture of multiple sine waves with different frequencies, and is used to reduce noise by removing only signals with unwanted frequencies from the signal] from the at least one audible (310) [0051: At this time, the comparative analysis unit (113-3) can determine whether cracks have occurred in the railroad sleeper based on the results of time domain analysis, frequency domain analysis, or wavelet analysis] and unit filters noise [0056: These wavelets are mainly used for noise removal or signal compression through signal analysis.For example, it is used in the analysis of signals that are completely different from sine waves[,] at least one vibration characteristic length for the at least one railroad element (310) [0070] & [0072: time domain analysis result in the railway sleeper crack inspection method according to the embodiment of the present invention shows that in the case of a defective product, the waveform length is relatively long and the amplitude is large]. Taek, as disclosed, does not explicitly disclose:
at least one AI based signal processing unit.
Takuro teaches [0005: a sound inspection system comprising: an input unit that receives sound data, which is sound data generated when a structure is struck; an extraction unit that extracts feature quantities from the sound data input to the input unit; a selection unit that selects predetermined feature quantities from the feature quantities extracted by the extraction unit as selected feature quantities; a learning unit that learns selected feature quantities based on target sound data from the sound data input to the input unit from the selected feature quantities selected by the selection unit and generates a learner] at least one AI based signal processing unit (523 & 524) [0027: the learning unit 523 learns selected features from among the selected features selected by the selection unit 522, based on the target impact sound data among the impact sound data input to the impact sound input unit 421, and generates a learner. Furthermore, the determination unit 524 uses the selected feature quantities selected by the selection unit 522, which are based on the target impact sound data from the impact sound data input to the impact sound input unit 421, and the learner generated by the learning unit 523, to determine whether or not there is an abnormality in the structure] & [0013: the analysis device 50 may learn the impact sound data and judgment results through machine learning, for example, using Jubatus, a machine learning framework e.g. known AI/MS platform where machine learning is a subset of AI] comprising at least one processor (50)[0026] at least one artificial intelligence [0013] and at least one software platform (523) configured to process the at least one audible sound signature [0040: the impact sound input unit 421 receives impact sound data along with a label indicating whether or not there is an abnormality in the structure] and filters noise [0051-0052].
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to use Takuro’s Machine Learning of abnormalities in a structure from impact sound data with Taek’s, as modified, sound impact data because machine learning improves accuracy in determining abnormalities in a structure by learning to eliminate noise and identify abnormal signatures [Takuro 0052-0053].
Claims 7 & 16. Dependent on the respective system of claim 4 and method of claim 13. Taek, as modified, further discloses the signal processing unit establishes multiple auditory thresholds for determining whether a rail element is broken [0073: it can be seen that in the case of a defective product, the level in the 1 kHz to 10 kHz band is relatively large (e.g. above levels at each frequency)] and applies at least one auditory threshold to a track segment (300) being inspected by selecting an auditory threshold established for physical conditions that most closely resemble physical conditions present for the track segment [0015: railroad track, and if it determines that the degree of cracks and damage exceeds the allowable level, it outputs a control signal to the image detection unit (20) to notify that a defect has occurred]. Taek, as modified, does not explicitly disclose:
An AI based signal processing unit
Takuro teaches [0005: a sound inspection system comprising: an input unit that receives sound data, which is sound data generated when a structure is struck; an extraction unit that extracts feature quantities from the sound data input to the input unit; a selection unit that selects predetermined feature quantities from the feature quantities extracted by the extraction unit as selected feature quantities; a learning unit that learns selected feature quantities based on target sound data from the sound data input to the input unit from the selected feature quantities selected by the selection unit and generates a learner] at least one AI based signal processing unit (523 & 524) [0027: the learning unit 523 learns selected features from among the selected features selected by the selection unit 522, based on the target impact sound data among the impact sound data input to the impact sound input unit 421, and generates a learner. Furthermore, the determination unit 524 uses the selected feature quantities selected by the selection unit 522, which are based on the target impact sound data from the impact sound data input to the impact sound input unit 421, and the learner generated by the learning unit 523, to determine whether or not there is an abnormality in the structure] & [0013: the analysis device 50 may learn the impact sound data and judgment results through machine learning, for example, using Jubatus, a machine learning framework e.g. known AI/MS platform where machine learning is a subset of AI] comprising at least one processor (50)[0026] at least one artificial intelligence [0013] and at least one software platform (523) configured to process the at least one audible sound signature [0040: the impact sound input unit 421 receives impact sound data along with a label indicating whether or not there is an abnormality in the structure] and select at least one auditory threshold [0041-0042: At this time, the learning unit 523 generates an anomaly detector that calculates the degree of discrepancy between the impact sound data to be judged and the impact sound data to be learned. The degree of deviation is, for example, the distance between data points. In this case, if the degree of discrepancy between the impact sound data to be judged and the impact sound data to be learned is greater than or equal to a predetermined threshold, the determination unit 524 determines that the impact sound data to be judged originated from a structure with an abnormality. An anomaly detector is an example of a learning device. Furthermore, the learning unit 523 may use the Local Outlier Factor as an unsupervised machine learning algorithm].
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to use Takuro’s Machine Learning of abnormalities in a structure from impact sound data with Taek’s, as modified, sound impact data because machine learning improves accuracy in determining abnormalities in a structure by learning to eliminate noise and identify abnormal signatures [Takuro 0052-0053].
Claims 8 & 17. Dependent on the respective system of claim 1 and method of claim 10. Taek does not explicitly disclose:
analyzing the at least one railroad spike is at least partially embedded in at least one railroad tie.
Poudel teaches a method for detecting breaks or defects in railroad spikes [Abstract]. Poudel further teaches an acoustic testing of a railroad element that is a railroad spike [0005: The operator taps each spike 10 with a hammer and attempts to distinguish different sounds (analogous to tap testing) to determine whether a spike is intact or broken] analyzing the railroad spike is at least partially embedded in at least one railroad tie [0005: Currently, railroads rely on a walking inspection approach to identify broken spikes. The operator taps each spike 10 with a hammer and attempts to distinguish different sounds (analogous to tap testing) to determine whether a spike is intact or broken] & [0007: The present invention aims to improve the efficiency of current track maintenance practice and ensure track safety].
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to use Poudel’s need for acoustic hammer tap testing of embedded railroad spikes as a testing arrangement for Taek’s mobile acoustic hammer tap testing of an installed railroad element because expanding inspection of installed railroad spikes and sleepers improves safe operation of train traffic along inspected and maintained stable rail lines [Poudel 0007].
Claims 9 & 18. Dependent on the respective system of claim 1 and method of claim 10.
Taek further discloses the at least one audio detection system (150) does not contact the at least one railroad element (Fig. 4: 150 microphone shown not in contact with inspected sleeper 310). Taek, as modified, does not explicitly disclose:
the at least one audio detection system does not contact the at least one railroad spike.
Poudel teaches a method for detecting breaks or defects in railroad spikes [Abstract]. Poudel further teaches an acoustic testing of a railroad element that is a railroad spike [0005: The operator taps each spike 10 with a hammer and attempts to distinguish different sounds (analogous to tap testing) to determine whether a spike is intact or broken] audio detection system (Fig. 5: 20) does not contact the at least one railroad spike [0023 The ultrasonic transducer 20 also receives these reflected ultrasonic signals].
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to use Poudel’s need for acoustic hammer tap testing with non-contact audio detection as a testing arrangement for Taek’s, as modified, mobile acoustic hammer tap testing of an installed railroad element because keeping audio reception as a non-contact inspection of installed railroad spikes and sleepers improves the efficiency of inspection while providing safe operation of train traffic along inspected and maintained stable rail lines [Poudel 0007].
Claim 11. Dependent on the method of claim 10. Taek further discloses the at least one hitting mechanism (140) comprises at least one hammer [0058].
Claim 14. Dependent on the method of claim 13. Taek further discloses via the signal processing unit, at least one audible sound signature (310) [0051: At this time, the comparative analysis unit (113-3) can determine whether cracks have occurred in the railroad sleeper based on the results of time domain analysis, frequency domain analysis, or wavelet analysis] and at least one vibration characteristic length for the at least one railroad element (310) [0070] & [0072: time domain analysis result in the railway sleeper crack inspection method according to the embodiment of the present invention shows that in the case of a defective product, the waveform length is relatively long and the amplitude is large]. Taek, as modified, does not explicitly disclose:
the at least one AI based signal processing unit analyzes the at least one audible sound signature
Takuro teaches [0005: a sound inspection system comprising: an input unit that receives sound data, which is sound data generated when a structure is struck; an extraction unit that extracts feature quantities from the sound data input to the input unit; a selection unit that selects predetermined feature quantities from the feature quantities extracted by the extraction unit as selected feature quantities; a learning unit that learns selected feature quantities based on target sound data from the sound data input to the input unit from the selected feature quantities selected by the selection unit and generates a learner] at least one AI based signal processing unit (523 & 524) [0027: the learning unit 523 learns selected features from among the selected features selected by the selection unit 522, based on the target impact sound data among the impact sound data input to the impact sound input unit 421, and generates a learner. Furthermore, the determination unit 524 uses the selected feature quantities selected by the selection unit 522, which are based on the target impact sound data from the impact sound data input to the impact sound input unit 421, and the learner generated by the learning unit 523, to determine whether or not there is an abnormality in the structure] & [0013: the analysis device 50 may learn the impact sound data and judgment results through machine learning, for example, using Jubatus, a machine learning framework e.g. known AI/MS platform where machine learning is a subset of AI] comprising at least one processor (50)[0026] at least one artificial intelligence [0013] and at least one software platform (523) configured to process the at least one audible sound signature [0040: the impact sound input unit 421 receives impact sound data along with a label indicating whether or not there is an abnormality in the structure] and select at least one auditory threshold [0041-0042: At this time, the learning unit 523 generates an anomaly detector that calculates the degree of discrepancy between the impact sound data to be judged and the impact sound data to be learned. The degree of deviation is, for example, the distance between data points. In this case, if the degree of discrepancy between the impact sound data to be judged and the impact sound data to be learned is greater than or equal to a predetermined threshold, the determination unit 524 determines that the impact sound data to be judged originated from a structure with an abnormality. An anomaly detector is an example of a learning device. Furthermore, the learning unit 523 may use the Local Outlier Factor as an unsupervised machine learning algorithm].
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to use Takuro’s Machine Learning of abnormalities in a structure from impact sound data with Taek’s, as modified, sound impact data because machine learning improves accuracy in determining abnormalities in a structure by learning to eliminate noise and identify abnormal signatures [Takuro 0052-0053].
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 Monica S Young whose telephone number is (303)297-4785. The examiner can normally be reached M-F 08:30-05:30 MST.
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, Peter Macchiarolo can be reached on 571-273-2375. 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.
/MONICA S YOUNG/Examiner, Art Unit 2855 /PETER J MACCHIAROLO/Supervisory Patent Examiner, Art Unit 2855