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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 07/10/2026 has been entered.
Response to Amendment
The amendment presents claims 1, 14, and 17 as amended, claims 2, 7, 8, and 16 as cancelled, and claim 18 as added. Claims 1, 3-6, 9-15, and 17-18 are pending examination.
The cancellation of claims 2 and 16 addresses the previously indicated objections and rejections of the same under 35 USC 112 (b).
Response to Arguments
Applicant’s arguments have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Claim Interpretation
The claims use the term “neuromorphic image sensor.” Such term is understood, in view of the instant specification (e.g., para. 0014), to include “event-based image sensor” and “event-based camera.”
Claim Rejections - 35 USC § 112
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.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1, 3-6, and 9-13 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 1 is amended to recite that the input data is determined by a further transfer function and that the further transfer function between the image data and the input data is formed by a further trained neural network. However, claim 1 also requires that output data is based on the input data by a transfer function and that the transfer function between the input data and the output data is formed by a trained neural network. It is unclear in what way, if any, the transfer function and further transfer function differ. Paragraph 0022 of the specification appears to be the only section the details the further trained neural network and the further transfer function and states the following:
The computing unit may be configured to generate the input data via a further transfer function based on the image data. The additional transfer function may be formed by an additional trained neural network. The further transfer function may be used to reduce the amount of image data. Alternatively, the image data transmitted from the neuromorphic image sensor may be the input data or used as input data.
Paragraph 0023 further details that the trained neural network and/or the further trained neural network may be CNN, BNN, and/or RNN.
Based on the direction provided by the specification it is unclear in what way, if any, the claimed transfer functions differ. This creates confusion as to the intended metes and bounds being sought by the claimed transfer functions, as well as, the associated neural networks employing such transfer functions.
Dependent claims 3-6 and 9-13 inherit the above deficiency as a result of their respective dependency from claim 1.
Additionally, claim 6 recites “a further transfer function” which renders the claim indefinite as it is unclear if the intention is for this transfer function to refer to the “further transfer function” in claim 1 or to a second further transfer function.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 1, 4, 6, and 9-13 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kim (US20150001196) in view of Mitani (US20220349707; relying on effective filing date of 10/28/2019), Fukushima (US20200406392, relying of effective filing date of 06/28/2019), and in further view of Croxford (US2019/0355142).
Regarding claim 1, Kim teaches a system for monitoring a laser machining process on a workpiece (para. 0003; “a method and an apparatus for monitoring a laser welding bead, and more particularly, to a technology capable of determining in real time whether or not welding defects are generated during laser welding.”), said system (Fig. 1-3) comprising:
PNG
media_image1.png
394
426
media_image1.png
Greyscale
Fig. 1 of Kim
a
a computing unit (classification processor 1244 which is operatively coupled to camera 114 via interface 121 and processor 1242) configured to determine input data based on the image data (para. 0052; “determines whether or not welding defects are generated using the feature variable and classifies a defect form (pattern).”), and to determine output data (Fig. 3, processor 1244 operatively coupled to controller 1246 such that controller 1246 receives output data from processor 1244. Para. 0053; controller 1246 controls the operation of the laser welding machine 10 depending on data from processor 1244) based on the input data by means of a transfer function (processor 1244 includes quality inspection classification processor 710 and precise inspection classification processor 720-Figs. 7-9) (para. 0077-0079; processors 710 and 720 utilize a neural network) [Here, the neural network algorithm is understood to refer to a transfer function utilized by the computing unit to generate the output data], said output data containing information (defects from bead shape-para. 0010) about the laser machining process,
wherein the transfer function between the input data and the output data and/or the further transfer function between the image data and the input data is formed by a trained neural network (neural network of Kim).
Kim teaches substantially the claimed invention including using an image sensor to capture an image. Kim teaches using a high speed camera for real time monitoring of a resulting weld bead (see para. 0043) (para. 0010; measuring bead shape) where the camera senses light reflected from the surface of the weld bead. Kim is silent on the sensor (i.e., camera) being a neuromorphic sensor.
Along a related field, Mitani teaches a measurement device for measuring the shape of an object or information indicating the shape of an object (para. 0002), which is considered reasonably pertinent to quality control (See for instance, paras. 0003, 0016, and 0035 of the instant application that details monitoring for quality control purposes). In other words, Mitani is concerned with improving the measuring of an object, which would be reasonably pertinent to the quality control of the resulting weld beads formed by a laser processing machine.
Furthermore, Mitani’s imaging unit for measuring the shape of an object is considered analogous to Kim’s imaging unit for measuring the shape of a weld bead.
Mitani’s system uses an event-based sensor (even-based camera 30-para. 0060) that receives reflected light from the surface of an object (R; Fig. 1), which is considered analogous to the arrangement of the imaging camera (114) receiving reflected light from the surface of the workpiece (Fig. 2) of Kim.
Mitani teaches that the event-based camera generates “an image from the event data outputted from the imaging unit” and, specifically, “outputs event data (specifically, two-dimensional point data, time, and polarities of luminance changes) including two-dimensional point data that identifies the position of each pixel that undergoes luminance changes” (para. 0060). Mitani also teaches that the camera captures and outputs data regarding luminance changes determined by respective pixels of the camera (para. 0038 and 0061).
Therefore, it would have been obvious to someone with ordinary skill in the art at the time the invention was filed to modify Kim with Mitani, by substituting the pixel based high speed camera of Kim, with the pixel based event-based camera of Mitani, for in doing so would provide an imaging unit that reduces data communication and generates images of the object at a higher speed (see Mitani, para. 0007).
Furthermore, the proposed combination would amount to a simple substitution of art recognized cameras (high speed vs event-based) performing the same function of imaging light reflected from the surface of the workpiece in order to determine a shape and the results of the substitution would have been predictable. (See MPEP 2144.06-II).
The combination of Kim and Mitani teaches substantially the claimed invention except for wherein the trained neural network comprises a convolutional neural network, a binary neural network, and/or a recurrent neural network.
Fukushima relates to an evaluation system for monitoring a laser welding system (Abstract) and teaches a trained neural network being a convolutional neural network, a binary neural network, and/or a recurrent neural network (para. 0081; “evaluation model 52d may be generated using various techniques known in the machine learning art. Examples of the techniques include, but are not limited to, various deep learning techniques such as CNN (Convolutional Neural Network) and RNN (Recurrent Neural Network). Another non-limiting example technique is SVM (Support Vector Machine). It is to be noted that these techniques have been provided for exemplary purposes, and it is possible to select any other learning technique suitable for the information sought to be obtained in generating the evaluation model 52d.”).
Therefore, it would have been obvious to someone with ordinary skill in the art at the time the invention was filed to modify Kim, as modified by Mitani, with Fukushima by replacing the type of neural network of modified Kim, with the neural network of Fukushima for in doing so would provide an alternative machine learning known in the art.
The combination teaches the computing unit including a trained neural network comprising a CNN, or RNN. The combination, therefore, suggests the computing unit (processor 1244 of Kim) being configured to determine the input data by a further transfer function based on the image data, and wherein the further transfer function between the image data and the input data is formed by a trained neural network (For purposes of examination, the claim will be understood to reference the same type of neural network and for the transfer functions to refer to the different layers, between which data is passed, inherent in such a neural network. See Fig. 8 and paragraph 0083 of Kim).
Alternatively, Croxford relates to the field of image processing using trained neural networks to detect characteristics or objects in the image (para. 0002) and teaches using distinct trained neural networks (first and section neural networks 104 and 108) to process an image captured by an image sensor (para. 0013). Croxford details that using the multi-pass neural network system has the benefit of reducing the amount of image data to be processed (para. 0012, 0018, 0035, and 0039).
While Croxford does not specifically detail laser processing of a workpiece, Croxford remains analogous prior art as it is reasonably pertinent to the problem of reducing the volume of image data process identified by the applicant of the instant application (see instant specification at paragraph 0011 and 0022). In this case, both the instant specification and Croxford detail using distinct neural networks to reduce the amount of image data being processed.
Therefore, it would have been obvious to someone with ordinary skill in the art at the time the invention was filed to modify Kim, as modified above, by duplicating the neural network of modified Kim, to include at least two neural networks as taught by Croxford, for in doing so would reduce the amount of image data to be processed. Further, the using distinct neural networks amount to the mere duplication of neural networks that has no patentable significance unless a new and unexpected result is produced. See MPEP 2144.04-VI-B.
Regarding claim 4, the primary combination, as applied in claim 1, teaches each claimed limitation, including wherein said neuromorphic image sensor (Kim as modified to include the event-based camera of Mitani) comprises a plurality of pixels configured to generate image data independently of one another in response to changes in brightness sensed by the respective pixel (Mitani; Abstract; “The object is optically imaged by an imaging unit and an image based on event data is acquired. The event data, which are outputted from the image sensor, include two-dimensional point data that specifies the positions of pixels corresponding to the pixels that had luminance changes responsively to the stripe pattern projected. Based on the event data, an image of the object is obtained.” Para. 0016; “the imaging unit is provided with an image sensor, the image sensor outputting event data including data of two-dimensional points whose positions of pixels are specified corresponding to changes of luminescence when receiving the light, and is configured to generate the captured images from the event data outputted by the image sensor.” ).
Regarding claim 6, the primary combination, as applied in claim 1, teaches each claimed limitation, including wherein said computing unit (Kim, 1244) is configured to determine the input data by a further transfer function based on the image data, and/or [Note: the use of “and/or” allows, under broadest reasonable interpretation for only one limitation to be met by the prior art] wherein the image data transmitted from said neuromorphic image sensor (Kim as modified by Mitani) are the input data (as detailed above in claim 1. The image data from the sensor is the input data going into the computing unit).
Regarding claim 9, the primary combination, as applied in claim 1, teaches each claimed limitation, including wherein the information about the laser machining process includes information about a state of the laser machining process, about a machining result, about a machining error and/or about a machining area of said workpiece [Note: the use of “and/or” allows, under broadest reasonable interpretation for only one limitation to be met by the prior art] (Kim, as detailed in claim 1 above, teaches that the information includes a shape of the weld bead and determining a defect based on that information).
Regarding claim 10, the primary combination, as applied in claim 1, teaches each claimed limitation, including wherein the computing unit (Kim, 1244) is configured to output the output data as control data for a laser machining system carrying out the laser machining process (Fig. 3 of Kim shows processor 1244 operatively coupled to controller 1246 such that controller 1246 receives output data from processor 1244. Para. 0053 discloses that controller 1246 controls operation of the laser welding machine based on data from processor 1244).
Regarding claim 11, the primary combination, as applied in claim 1, teaches each claimed limitation, including wherein a laser machining system (Kim, machining system 10) for machining a workpiece (20A/B) using a laser beam (laser welding-para. 0003, said laser machining system comprising: a laser machining head (the machining head is taken as the head shown in Figure 1 that outputs the laser onto workpiece 20A/B and generates the weld bead 30) for radiating a laser beam onto said workpiece; and the system according to claim 1 (as detailed above).
Regarding claim 12, the primary combination, as applied in claim 11, teaches each claimed limitation, including wherein [Kim] said computing unit (1244) is arranged on or in said laser machining head (10), and/or wherein said neuromorphic image sensor (Kim as modified by Mitani; Kim’s camera is shown on and outside of the machining head in Figure 1. The proposed combination seeks to use Mitani’s camera in the same location as Kim’s camera) is arranged on an outside of said laser machining head and/or on said laser machining head.
Regarding claim 13, the primary combination, as applied in claim 11, teaches each claimed limitation, including wherein [Kim] a laser source configured to generate the laser beam (Kim teaches system 10 is for laser welding. A laser source in inherently required); and a control unit configured to control, based on the output data determined by said computing unit, said laser machining system and/or said laser machining head and/or said laser source and/or to control the laser machining process (Fig. 3 of Kim shows processor 1244 operatively coupled to controller 1246 such that controller 1246 receives output data from processor 1244. Para. 0053 discloses that controller 1246 controls operation of the laser welding machine based on data from processor 1244).
Claim(s) 3 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kim (US20150001196) in view of Mitani (US20220349707; relying on effective filing date of 10/28/2019), Fukushima (US20200406392, relying of effective filing date of 06/28/2019), and in view of Takigawa (US20200387131).
Regarding claim 3, the primary combination, as applied in claim 1, teaches each claimed limitation, except for wherein said neuromorphic image sensor is configured to transmit image data to said computing unit continuously and/or asynchronously.
Takigawa relates to a laser machining system in which image data from an imaging sensor [(Fig. 2; 13) (Fig. 5; 15) (Fig. 7; 17)] is used as input data to a computing unit (machine learning device 6; Fig. 8). Takigawa teaches the laser machining system including a controller (33; Fig. 8) that receives an output from the computing unit (6) and controls at least one parameter in response (para. 0138).
Takigawa also teaches the computing unit (6) acquiring imaging data continuously, as well as, at a point in time when an abnormality occurs (para. 0137).
Therefore, it would have been obvious to someone with ordinary skill in the art at the time the invention was filed to modify Kim, as modified by Mitani and Fukushima, with Takigawa by adding to the temporal operation of the imaging unit of modified Kim, as well as, adding to the functionality of the computing unit of Kim, with the continuous supplying of imaging data taught by Takigawa for in doing so would provide additional data points that would improve the functionality of the computing unit by predicting the occurrence of a defect (see para. 0137 of Takigawa).
Claim(s) 5 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kim (US20150001196) in view of Mitani (US20220349707; relying on effective filing date of 10/28/2019), Fukushima (US20200406392, relying of effective filing date of 06/28/2019), and in further view of Lee (US20150030204).
Regarding claim 5, the primary combination, as applied in claim 4, teaches each claimed limitation, including wherein the image data of a pixel comprise at least a pixel address corresponding to the pixel (Mitani; output data includes two-dimensional point data that identifies the location of a pixel-para. 0038).
The combination is silent on wherein the image data of a pixel comprises a time stamp corresponding to the sensed change in brightness.
Lee relates to a system for analyzing at least one of an appearance of an object and a motion of an object (Abstract).
Lee teaches using an event-based vision sensor that asynchronously provides an output signal in response to detection of a predetermined event, which includes a change in brightness of light incident on the event-based vision sensor (para. 0030).
Lee further teaches that the “event signal may include a time stamp at which a time of a predetermined event is detected, an indicator for indicating a type of an event, and an index of a pixel in which the predetermined event is detected” and the “time stamps corresponding to the pixels of a resolution may be stored in a table in memory, thereby time signals of event times for pixels may be utilized, as discussed below” (para. 0040). Lee, therefore, teaches the image data of a pixel comprising a time stamp corresponding to the sensed change in brightness.
Therefore, it would have been obvious to someone with ordinary skill in the art at the time the invention was filed to modify Kim, as modified by Mitani and Fukushima, with Lee by adding to the image data from the imaging unit of modified Kim, as well as, adding to the functionality of the computing unit of Kim, with the time stamp data taught by Lee, for in doing so would allow for the classifying of pixel patterns (Lee, para. 0041), thereby further improving imaging of the workpiece (Lee teaches using the determined pattern to analyze the shape, outline, or location of the object. In this case, providing time stamp data to classify a pixel pattern would improve the monitoring system’s ability to accurately measure the shape of the weld bead and, as a result, a defect.).
Claim(s) 14-15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kim (US20150001196) in view of Mitani (US20220349707; relying on effective filing date of 10/28/2019), Wersborg (US2013/0178953) and in further view of Lee (US20150030204).
Regarding claim 14, Kim teaches a method for monitoring a laser machining process on a workpiece (para. 0003; “a method and an apparatus for monitoring a laser welding bead, and more particularly, to a technology capable of determining in real time whether or not welding defects are generated during laser welding.”), said method (Fig. 1-3) comprising:
generating image data from a machining area of said workpiece using a
determining input data based on the image data (classification processor 1244 which is operatively coupled to camera 114 via interface 121 and processor 1242) (para. 0052; “determines whether or not welding defects are generated using the feature variable and classifies a defect form (pattern).”),
determining output data based on the input data by a transfer function (Fig. 3, processor 1244 operatively coupled to controller 1246 such that controller 1246 receives output data from processor 1244. Para. 0053; controller 1246 controls the operation of the laser welding machine 10 depending on data from processor 1244) (processor 1244 includes quality inspection classification processor 710 and precise inspection classification processor 720-Figs. 7-9) (para. 0077-0079; processors 710 and 720 utilize a neural network) [Here, the neural network algorithm is understood to refer to a transfer function utilized by the computing unit to generate the output data], said output data containing information about the laser machining process (defects from bead shape-para. 0010).
wherein the transfer function between the input data and the output data and/or the further transfer function between the image data and the input data is formed by a trained neural network (neural network of Kim).
Kim teaches substantially the claimed invention including using an image sensor to capture an image. Kim teaches using a high speed camera for real time monitoring of a resulting weld bead (see para. 0043) (para. 0010; measuring bead shape) where the camera senses light reflected from the surface of the weld bead. Kim is silent on the sensor (i.e., camera) being a neuromorphic sensor.
Along a related field, Mitani teaches a measurement device for measuring the shape of an object or information indicating the shape of an object (para. 0002), which is considered reasonably pertinent to quality control (See for instance, paras. 0003, 0016, and 0035 of the instant application that details monitoring for quality control purposes). In other words, Mitani is concerned with improving the measuring of an object, which would be reasonably pertinent to the quality control of the resulting weld beads formed by a laser processing machine.
Furthermore, Mitani’s imaging unit for measuring the shape of an object is considered analogous to Kim’s imaging unit for measuring the shape of a weld bead.
Mitani’s system uses an event-based sensor (even-based camera 30-para. 0060) that receives reflected light from the surface of an object (R; Fig. 1), which is considered analogous to the arrangement of the imaging camera (114) receiving reflected light from the surface of the workpiece (Fig. 2) of Kim.
Mitani teaches that the event-based camera generates “an image from the event data outputted from the imaging unit” and, specifically, “outputs event data (specifically, two-dimensional point data, time, and polarities of luminance changes) including two-dimensional point data that identifies the position of each pixel that undergoes luminance changes” (para. 0060). Mitani also teaches that the camera captures and outputs data regarding luminance changes determined by respective pixels of the camera (para. 0038 and 0061).
Therefore, it would have been obvious to someone with ordinary skill in the art at the time the invention was filed to modify Kim with Mitani, by substituting the pixel based high speed camera of Kim, with the pixel based event-based camera of Mitani, for in doing so would provide an imaging unit that reduces data communication and generates images of the object at a higher speed (see Mitani, para. 0007).
Furthermore, the proposed combination would amount to a simple substitution of art recognized cameras (high speed vs event-based) performing the same function of imaging light reflected from the surface of the workpiece in order to determine a shape and the results of the substitution would have been predictable. (See MPEP 2144.06-II).
The combination of Kim and Mitani teaches substantially the claimed invention except for wherein the machining area includes a cutting kerf or a vapor capillary and/or a melt pool surrounding the vapor capillary.
Wersborg relates to a method for controlling a laser processing operation by means of a reinforcement learning agent (para. 0001) and teaches the benefit of using machine learning to improve laser cutting and laser welding processes (see paragraph 0003; “Laser cutting and laser welding may thus benefit from the cognitive capabilities of artificial agents. If these agents can learn how to weld or cut, it would not only reduce the system configuration effort, but also increase its flexibility. Moreover, if an agent could improve itself over time, it could gain the capability to develop its everyday tasks, increase output, and assure quality. Many manufacturers wish to have a prompt cutting or welding technique, a system that does not need to be reconfigured when it takes over a new production task. This kind of system would significantly increase welding and cutting efficiency and assure quality. Quality assurance is especially important when processing parts are associated with safety, for instance within cars or airplanes.”).
Wersborg further teaches using image data of the cutting or welding region for feedback control (para. 0007 and 0128; high speed camera) which is used, as an input, by the machine learning system (para. 0120 and 0128).
Accordingly, Wersborg teaches using image data as an input parameter to a machine learning system in which the image data is representative of a machining area of a laser cutting or welding process. Wersborg teaches the formation of kerf during cutting (para. 0006, kerf and cutting edge quality, including dross, roughness or parallelism of edges, with para. 0077 indicating using the camera to detect kerf) (para. 0120 details using machine learning to achieve predefined melt pool size, kerf width, or cutting quality) (para. 0081 similarly discusses formation of melt pools and keyholes).
Therefore, it would have been obvious to someone with ordinary skill in the art at the time the invention was filed to modify Kim with Mitani, by substituting the laser welding process of Kim, with the laser cutting process of Wersborg, for in doing so would provide an alternative laser machining process that would benefit from the monitoring and control methodology using image data and a trained neural network of Kim. In other words, those of ordinary skill in the art would recognize the benefit, based on the teachings of Wersborg, of using the methodology outlined in Kim in a laser cutting process in order to improve the quality of automating laser cutting processes.
The combination of Kim, Mitani, and Wersborg, teaches each claimed limitation, including wherein the image data of a pixel comprise at least a pixel address corresponding to the pixel (Mitani; output data includes two-dimensional point data that identifies the location of a pixel-para. 0038).
The combination is silent on wherein the image data of a pixel comprises a time stamp corresponding to the sensed change in brightness.
Lee relates to a system for analyzing at least one of an appearance of an object and a motion of an object (Abstract).
Lee teaches using an event-based vision sensor that asynchronously provides an output signal in response to detection of a predetermined event, which includes a change in brightness of light incident on the event-based vision sensor (para. 0030).
Lee further teaches that the “event signal may include a time stamp at which a time of a predetermined event is detected, an indicator for indicating a type of an event, and an index of a pixel in which the predetermined event is detected” and the “time stamps corresponding to the pixels of a resolution may be stored in a table in memory, thereby time signals of event times for pixels may be utilized, as discussed below” (para. 0040). Lee, therefore, teaches the image data of a pixel comprising a time stamp corresponding to the sensed change in brightness.
Therefore, it would have been obvious to someone with ordinary skill in the art at the time the invention was filed to modify Kim, as modified by Mitani and Wersborg, with Lee by adding to the image data from the imaging unit of modified Kim, as well as, adding to the functionality of the computing unit of Kim, with the time stamp data taught by Lee, for in doing so would allow for the classifying of pixel patterns (Lee, para. 0041), thereby further improving imaging of the workpiece (Lee teaches using the determined pattern to analyze the shape, outline, or location of the object. In this case, providing time stamp data to classify a pixel pattern would improve the monitoring system’s ability to accurately measure the shape of the weld bead and, as a result, a defect.).
Regarding claim 15, the primary combination, as applied in claim 14, teaches each claimed limitation, including controlling, in real time, at least one parameter of the laser machining process based on the determined output data (Fig. 3 of Kim shows processor 1244 operatively coupled to controller 1246 such that controller 1246 receives output data from processor 1244. Para. 0053 discloses that controller 1246 controls operation of the laser welding machine based on data from processor 1244) (para. 0009; “an apparatus and a method for monitoring a laser welding bead capable of easily managing a welding production process by performing welding bead quality monitoring in real time and immediately notifying a user of defects upon sensing the defects.”) (see also Fig. 4; steps S412-S422).
Regarding claim 17, the primary combination, as applied in claim 14, teaches each claimed limitation, including wherein the cutting kerf includes a cutting front (See Wersborg as applied above in claim 14, where a cutting front is necessarily present in the formation of a cut) and/or puncture hole.
Claim(s) 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kim (US20150001196) in view of Mitani (US20220349707; relying on effective filing date of 10/28/2019), Wersborg (US2013/0178953), Lee (US20150030204), and in further view of Fukushima (US20200406392, relying of effective filing date of 06/28/2019).
Regarding claim 18, the combination teaches substantially the claimed invention, as applied in claim 14, except for wherein the trained neural network comprises a convolutional neural network, a binary neural network, and/or a recurrent neural network.
Fukushima relates to an evaluation system for monitoring a laser welding system (Abstract) and teaches a trained neural network being a convolutional neural network, a binary neural network, and/or a recurrent neural network (para. 0081; “evaluation model 52d may be generated using various techniques known in the machine learning art. Examples of the techniques include, but are not limited to, various deep learning techniques such as CNN (Convolutional Neural Network) and RNN (Recurrent Neural Network). Another non-limiting example technique is SVM (Support Vector Machine). It is to be noted that these techniques have been provided for exemplary purposes, and it is possible to select any other learning technique suitable for the information sought to be obtained in generating the evaluation model 52d.”).
Therefore, it would have been obvious to someone with ordinary skill in the art at the time the invention was filed to modify Kim, as modified above, with Fukushima by replacing the type of neural network of modified Kim, with the neural network of Fukushima for in doing so would provide an alternative machine learning known in the art.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JUSTIN C DODSON whose telephone number is (571)270-0529. The examiner can normally be reached Mon.-Fri. 12:00-8:00 PM (ET).
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, Steven Crabb can be reached at (571)270-5095. 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.
/JUSTIN C DODSON/ Primary Examiner, Art Unit 3761
Prosecution Insights