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 Amendments
The amendments to claims 1-3 are accepted and entered.
Claims 1-5 are pending regarding this application.
Response to Arguments
Applicant’s arguments, see Remarks, filed 08/28/2026, with respect to the 112(f) interpretation and 112(b) rejection/112(a) rejections regarding the pre-processing module in claim 1 have been fully considered and are persuasive. Applicant has amended claim 1 such that all references of or to the pre-processing module have been deleted, as shown in the Claims filed 08/28/2026. As such, the 112(f) interpretation and 112(b) rejection/112(a) rejections regarding the pre-processing module in claim 1 have been withdrawn.
Applicant’s arguments, see Remarks, filed 08/28/2026, with respect to the 112(f) interpretation and 112(b) rejection/112(a) rejection regarding the “host computer detection system” in claim 1 have been fully considered and are persuasive. Applicant has amended claim 1 such that the host computer detection system (previously “host computer detection module”), as shown in the Claims filed 08/28/2026 now recites sufficient structure (i.e. microcontroller/sensor control board/system operation interface), which overcomes the previous 112(f) interpretation, thus rendering the 112(a)/112(b) rejections moot. As such, the 112(f) interpretation and 112(b) rejection/112(a) rejections regarding the host computer detection system in claim 1 have been withdrawn.
Applicant's arguments filed 08/28/2026 have been fully considered but they are not persuasive. Applicant has argued that the amendments to claim 1, as shown in the Claims filed 08/28/2026, sufficiently adds structure to the “detection model” such that it is no longer a generic placeholder coupled to functional language. However, examiner disagrees with this assertion. Neither a “Transformer model”, nor “a patch embedding layer”, nor “a positional embedding layer”, nor “a Transformer encoder”, nor “a normalization layer”, nor “a classification head” act as either a physical structure or a processor. As such, the fact that the detection model is a nonce term coupled to functional language is unchanged by the addition of the above elements to claim 1. Therefore, the 112(f) interpretation and 112(b)/112(a) rejection of claim 1 in regard to the detection model has been maintained.
Regarding the acknowledgement of the allowable subject matter, applicant’s amendment has necessitated the new ground(s) of rejection presented in this Office action. Please see the 103 rejection of claims 1-3 and 5 below regarding this matter.
Claim Objections
Claims 3 and 5 are objected to because of the following informalities:
Claim 3 recites “a plurality of cameras” in line 3, please amend to recite “the plurality of industrial cameras”, as claim 1, upon which claim 3 depends, already recites “a plurality of industrial camera”.
Claim 3 recites “the plurality of cameras” in line 4, please amend to recite “the plurality of industrial camera”.
Claim 3 recites “a crochet needle” in line 5, please amend to recite “the crochet needle”, as claim 1, upon which claim 3 depends, already recites “a crochet needle”.
Claim 3 recites “detection results” in line 8, please amend to recite “the detection results”, as claim 1, upon which claim 3 depends, already recites “detection results”
Claim 3 recites “a defect” in line 12, please amend to recite “the defect”, as claim 1, upon which claim 3 depends, already recites “a defect”
Claim 3 recites “a binary signal to a switching circuit” in line 14, please amend to recite “the binary signal to the switching circuit”, as claim 1, upon which claim 3 depends, already recites “a binary signal to a switching circuit”
Claim 3 recites “a type and location” in line 15. Please amend to recite “the type and location”, as claim 1, upon which claim 3 depends, already recites “a type and location”
Claim 3 recites “a broken filament or change a needle”. Please amend to recite “the broken filament or change the needle”, as claim 1, upon which claim 3 depends, already recites “a broken filament or change a needle”
Claim 5 recites “a Transformer model” in line 7. Please amend to recite “the Transformer model”, as claim 1, upon which claim 5 depends, already recites “a Transformer model”
Claim 5 recites “a plurality of patches” in lines 8-9. Please amend to recite “the plurality of patches”, as claim 1, upon which claim 5 depends, already recites “a plurality of patches”
Appropriate correction is required.
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 following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
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) or pre-AIA 35 U.S.C. 112, sixth paragraph, 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) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(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) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, 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) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitation(s) is/are:
In claim 1: “detection model is configured to…”
After a careful analysis, as disclosed above, and a careful review of the specification, the “detection model” in claim 1 lacks the required structure as outlined in MPEP 2181 II(B) in order to determine whether to interpret a 112(f) limitation as a computer-implemented means-plus-function limitation, and, therefore, it is not clear if the “detection model” in claim 1 is a structure or a computer implemented 112(f) limitation, which require a structure (usually a storage medium) and a specific algorithm. See the 112(a)/112(b) rejections regarding this matter. However, solely for prior art purposes, it is assumed that applicant intended for the “detection model” in claim 1 to be carried out by a processor (or equivalent structure). Below is the corresponding structure and algorithm which are being read into the above limitation:
“detection model” (Regarding the detection model, the specification does not include enough information to determine the structure for the detection model. More specifically, the specification does not clarify what the structure of the detection model is (i.e. processor, CPU, camera) in enough detail to determine whether or not to interpret the detection model and its corresponding functional language as a computer-implemented 112(f) limitation. See the 112(a)/112(b) rejections below regarding this matter. However, for the purposes of prior art, since the detection model is based on a Transformer model (see para. [0013] of applicant’s specification), it is assumed that applicant intends for the corresponding structure of the detection model to be a processor. As such, solely for the purposes of prior art searching, the “detection model” is being interpreted as computer-implemented 112(f), wherein the structure for the detection model is a processor and the corresponding algorithm which can be found in para. [0047]-[0055] and [0065]-[0081] of applicant’s specification.).
Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
Claim Rejections - 35 USC § 112(b)
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 1-5 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (preAIA ), 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.
Regarding claim 1, the “detection model” in claim 1 invokes 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. However, the written description fails to disclose the corresponding structure, material, or acts for performing the entire claimed function and to clearly link the structure, material, or acts to the function.
Regarding the “detection model”, applicant appears to describe the claimed detection model in para. [0047]-[0055] and [0065]-[0081]. However, nowhere in these sections does the applicant define any specific structure for the detection model. As such, the specification fails to describe a specific structure for the detection model in order to determine whether the detection model should be interpreted as a computer-implemented 112(f) limitation. However, as noted above in the 112(f) section, solely for the purposes of prior art searching and claim interpretation, since the detection model is based on a Transformer model (see para. [0013] of applicant’s specification), it is assumed that applicant intends for the corresponding structure of the detection model to be a processor. As such, while no structure is provided in the specification, the corresponding algorithm of the detection model (para. [0047]-[0055] and [0065]-[0081]) is being read into the recited “detection model” in claim 1 for the purposes of prior art searching.
Therefore, the claim is indefinite and is rejected under 35 U.S.C. 112(b) or pre-AIA 35
U.S.C. 112, second paragraph.
Applicant may:
(a) Amend the claim so that the claim limitation will no longer be interpreted as a limitation under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph;
(b) Amend the written description of the specification such that it expressly recites what structure, material, or acts perform the entire claimed function, without introducing any new matter (35 U.S.C. 132(a)); or
(c) Amend the written description of the specification such that it clearly links the structure,
material, or acts disclosed therein to the function recited in the claim, without
introducing any new matter (35 U.S.C. 132(a)).
If applicant is of the opinion that the written description of the specification already
implicitly or inherently discloses the corresponding structure, material, or acts and clearly links
them to the function so that one of ordinary skill in the art would recognize what structure,
material, or acts perform the claimed function, applicant should clarify the record by either:
(a) Amending the written description of the specification such that it expressly recites the corresponding structure, material, or acts for performing the claimed function and clearly links or associates the structure, material, or acts to the claimed function, without introducing any new matter (35 U.S.C. 132(a)); or
(b) Stating on the record what the corresponding structure, material, or acts, which are implicitly or inherently set forth in the written description of the specification, perform the claimed function. For more information, see 37 CFR 1.75(d) and MPEP §§ 608.01(o) and 2181.
Claims 2-5 are rejected as being dependent upon claim 1.
Claim Rejections - 35 USC § 112(a) (written description)
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.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
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 of carrying out his invention.
Claims 1-5 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (preAIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
Regarding claim 1, as per MPEP § 2181(IV), “A means- (or step-) plus-function
limitation that is found to be indefinite under 35 U.S.C. 112(b) based on failure of the specification to disclose corresponding structure, material or act that performs the entire claimed function also lacks adequate written description” (emphasis added). Furthermore, as per MPEP 2163.03(VI), “(s)uch a limitation also lacks an adequate written description as required by 35 U.S.C. 112(a) or pre-AIA 35 U.S.C. 112, first paragraph, because an indefinite, unbounded functional limitation would cover all ways of performing a function and indicate that the inventor has not provided sufficient disclosure to show possession of the invention.” Therefore, since applicant has not defined any particular structure for the
“detection model” in claim 1, the inventor has not provided sufficient disclosure to show possession of the invention. Applicant has not provided any specific definition for the structure that carry out the functions disclosed in claim 1. Additionally, the claimed invention as a whole may not be adequately described if the claims require an essential or critical feature which is not adequately described in the specification and which is not conventional in the art or known to one of ordinary skill in the art. It appears that these components and/or features are essential and critical features of the applicants invention because without them applicant’s invention wouldn’t work. In particular, the structure of the pre-processing submodule or detection model are not described in any detail. Therefore, since applicant has not adequately described a particularly structure for performing each of the functions, a person skilled in the art at the time the invention was filed would not have recognized that the inventor was in possession of the invention as claimed.
Claims 2-5 are rejected based on their dependency to claim 1.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1 and 3 are rejected under 35 U.S.C. 103 as being unpatentable over Xiao et al. (CN 117074415 A, see attached English translation for citations), hereinafter Xiao in view of Ru et al. (CN 114820464 A, see attached English translation for citations), hereinafter Ru.
Regarding claim 1, Xiao teaches a system for detecting filament breakage and
an image acquisition module (Xiao teaches an image acquisition module in para. [0049]);
an image processing module (Xiao teaches “the intelligent wire breakage detection model” as shown in para. [0076]); and
a host computer detection system (Xiao teaches “the upper computer detecting system comprises: the warp knitting machine shutdown controller, the image data acquisition controller, the sensor control panel and the system operation interface” in para. [0095]);
wherein the image acquisition module comprises a light source, a plurality of industrial cameras and an industrial lens (Xiao teaches that “the image acquisition module includes: the industrial light source, the industrial camera and the industrial lens” in para. [0049]); the image processing module comprises a detection model based on a Transformer model (Xiao teaches “the intelligent wire breakage detection model” in para. [0076]), the detection model comprising a patch embedding layer (Xiao teaches dividing an image into patches and extracting features from said patches within layers as shown in para. [0078] and [0089]), a positional embedding layer (Xiao teaches the “embedding layer” in para. [0079]), a Transformer encoder (Xiao teaches a transformer encoder in para. [0080]), a normalization layer (Xiao teaches a normalization process within the intelligent wire breakage detection model as shown in para. [0076]-[0077]), and a classification head (Xiao teaches an output layer that generates predictions for each category which is interpreted as equivalent to the claimed classification head as shown in para. [0084]); and
the host computer detection system comprises a microcontroller (Xiao teaches the computer detecting system comprising a “warp knitting machine shutdown controller” in para. [0095]. This shutdown controller is interpreted as equivalent to the claimed microcontroller See also para. [0065] wherein the “warp knitting machine shutdown controller” operates through a single chip), a sensor control board (Xiao teaches the computer detecting system comprising a “sensor control panel” in para. [0095]) and a system operation interface (Xiao teaches the computer detecting system comprising a “system operation interface” in para. [0095]); and
the image acquisition module is configured to capture a blanket knitting image at a crochet needle of a knitting mechanism of a warp knitting machine in real time under irradiation of the light source (Xiao teaches “fixing a plurality of groups of industrial cameras, completing camera configuration, and setting ROI and exposure time; under the irradiation of an industrial light source, collecting blanket knitting images at the crochet hook of a looping mechanism of a warp knitting machine in real time” in para. [0062]);
the detection model is configured to divide the blanket knitting image into a plurality of patches, convert the plurality of patches into sequence data, process the sequence data by using the patch embedding layer, the positional embedding layer, the Transformer encoder, the normalization layer, and the classification head, and perform filament breakage and needle breakage detection in real time on the blanket knitting image to obtain detection results (Xiao teaches “performing defect detection by using a trained intelligent broken wire detection model, giving a new image, dividing the new image into a plurality of image blocks, and predicting the classification result or defect position information of each image block through the model so as to realize the target of defect detection” in para. [0089], wherein this process occurs using the layers/head/encoder as shown in para. [0076]-[0088]);
the detection results are sent to the host computer detection system via serial communication and displayed in the system operation interface (Xiao teaches “sending the detection result obtained in the step 2 to an upper computer through serial port communication, and displaying the detection result on an interface” as shown in para. [0064]); and
if there is a defect in a blanket according to the detection results, the microcontroller is configured to output a binary signal to a switching circuit of the warp knitting machine to stop the warp knitting machine, and the host computer detection system is configured to determine a type and location of the defect and warn an operator to repair a broken filament or change a needle according to the type and location of the defect (Xiao teaches “judging whether defects are generated according to the detection result, if yes, outputting a switching value to a switching circuit of the warp knitting machine through the single chip, stopping the machine, and informing a field operator to perform line supplementing or needle changing according to the defect types; otherwise, the step 2 is continuously executed, and the production condition of the warp knitting machine is monitored” in para. [0065]. Here, the switching value is interpreted as equivalent to the claimed binary signal).
Xiao fails to specifically teach detecting needle breakage.
However, Ru teaches detecting needle breakage (Ru teaches detecting needle breakage in para. [0131]).
Xiao and Ru are both considered to be analogous to the claimed invention because they are in the same field of analyzing images of knitting machines to detect defects. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of Xiao to incorporate the teachings of Ru and include “detecting needle breakage”. The motivation for doing so is that it would have been obvious to one of ordinary skill in the art before the effective filing date to additionally analyze a needle for breakage in a knitting machine environment in order to further improve production efficiency and reduce raw material loss, which are among the intended improvements for the teachings of Xiao as shown in para. [0006]. Furthermore, improved quality inspection and “solving the problem that automatic broken needle detection of the glove machine cannot be realized” would further “improve the yield and reduce defective products” within the field of unmanned knitting machines, as suggested by Ru in para. [0005] and [0007]. Therefore, it would have been obvious to one of ordinary skill at the time the invention was filed to combine Xiao with Ru to obtain the invention specified in claim 1.
Regarding claim 3, Xiao and Ru teach a method for detecting filament breakage and needle breakage by using the system of claim 1, comprising:
(S1) fixing a plurality of cameras; setting Region of Interest (ROI) and exposure time for each of the plurality of cameras (Xiao teaches “fixing a plurality of groups of industrial cameras, completing camera configuration, and setting ROI and exposure time” in para. [0062]); and
capturing, by the plurality of cameras, the blanket knitting image at a crochet needle of the knitting mechanism in real time under irradiation of the light source (Xiao teaches “under the irradiation of an industrial light source, collecting blanket knitting images at the crochet hook of a looping mechanism of a warp knitting machine in real time to form a defect detection data set, and generating a rich data set through a sample” as shown in para. [0062]);
(S2) performing, by the detection model, the filament breakage and needle breakage detection in real time on the blanket knitting image to obtain detection results (Ru teaches detecting needle breakage as shown in claim 1. Xiao further teaches “detecting the broken yarn defects of the acquired blanket weaving image in real time by using an intelligent broken yarn detection model, and judging the types and positions of the blanket defects” in para. [0063]);
(S3) sending the detection results obtained in step (S2) to the host computer detection system via serial communication; and displaying the detection results in the system operation interface (Xiao teaches “sending the detection result obtained in the step 2 to an upper computer through serial port communication, and displaying the detection result on an interface” in para. [0064]); and
(S4) judging whether there is a defect in a blanket according to the detection results; if yes, outputting, by the microcontroller in the host computer detection system, a binary signal to a switching circuit of the warp knitting machine to stop the warp knitting machine, determining a type and location of the defect, and warning an operator to repair a broken filament or change a needle according to the type and location of the defect; otherwise, returning to step (S2) to continuously monitor operation condition of the warp knitting machine (Xiao teaches “judging whether defects are generated according to the detection result, if yes, outputting a switching value to a switching circuit of the warp knitting machine through the single chip, stopping the machine, and informing a field operator to perform line supplementing or needle changing according to the defect types; otherwise, the step 2 is continuously executed, and the production condition of the warp knitting machine is monitored” as shown in para. [0065]. Here, the switching value is interpreted as equivalent to the claimed binary signal).
Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Xiao et al. (CN 117074415 A, see attached English translation for citations), hereinafter Xiao in view of Ru et al. (CN 114820464 A, see attached English translation for citations), hereinafter Ru and Hou et al. (CN 108564611 A, see attached English translation for citations), hereinafter Hou.
Regarding claim 2, Xiao and Ru teach the system of claim 1,
wherein the patch embedding layer is configured to divide the blanket knitting image into the plurality of patches, and convert each of the plurality of patches into a vector representation with a fixed length by a linear projection operation (Xiao teaches “dividing the new image into a plurality of image blocks, and predicting the classification result or defect position information of each image block through the model so as to realize the target of defect detection” in para. [0089] wherein the image blocks are then used for “obtaining the low-dimensional feature representation of the blanket knitting image by transmitting an image sample through a convolutional layer and a pooling layer” as shown in para. [0078], wherein the patches are further converted into a vector representation with a fixed length as shown in para. [0079]. Here, since the fixed length is matched with the dimension of the input graphic feature, it is inherent that this is a linear projection operation);
the positional embedding layer is configured to add positional information to the vector representation so that the detection model is capable of capturing relative positional relationships between the plurality of patches (Xiao teaches “position coding: adding position codes to the extracted graphic features to capture relative position information between the features; the position code is a matrix, the dimension of the position code is matched with the dimension of the input graphic feature, and relative position information is embedded into the low-dimensional feature representation” in para. [0079]);
the Transformer encoder comprises a plurality of encoder layers (Xiao teaches “forming a deep layer transducer model by stacking a plurality of transducer encoder layers” in para. [0080]);
each of the plurality of encoder layers comprises a multi-head self-attention layer and a feed- forward neural network (Xiao teaches a multi-head self-attention layer and a feed- forward neural network within each encoder layer within para. [0080]-[0083]);
the multi-head self-attention layer is configured to establish an attention relationship between elements in the sequence data (Xiao teaches “by calculating the attention weight among elements in the low-dimensional feature representation, capturing the relation among the features, the global context sensing can be carried out on the features, and the richer feature representation can be extracted” in para. [0082]); and
the feed-forward neural network is configured to perform non-linear transformation on each of the elements in the sequence data by using a multi-layer perceptron (Xiao teaches “the output of the self-attention mechanism is further processed and mapped by a multi-layer perceptron (MLP) to generate a richer feature representation” in para. [0083]);
the classification head is a final layer of the Transformer model, and is configured to map an output of a final encoder layer among the plurality of encoder layers to a probability distribution of defect categories (Xiao teaches an “output layer: adding a full-join layer to the output of the last Transformer encoder layer for mapping the high-dimensional representation to the probability distribution of the target class; generating probabilistic predictions for each category using a softmax activation function” as shown in para. [0084]).
Xiao and Ru fail to teach the normalization layer is configured to normalize an output of each of the plurality of encoder layers to control a gradient flow and accelerate a training process.
However, Hou teaches the normalization layer is configured to normalize an output of each of the plurality of encoder layers to control a gradient flow and accelerate a training process (Hou teaches “every layer of convolutional layer is followed by batch and normalizes layer (Batch Normalization), it is convenient for the passback of the flowing and error of gradient, utilizes trained Fast Convergent” as shown in para. [0033] and [0039]. Here, Hou’s teaching of using the normalization layer can be combined with Xiao’s teaching of analyzing/utilizing the output of the plurality of the encoder layers as shown in para. [0080]-[0084]).
Xiao, Ru, and Hou are all considered to be analogous to the claimed invention because they are in the same field of analyzing images using models with a plurality of layers. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of Xiao (as modified by Ru) to incorporate the teachings of Hou and include “the normalization layer is configured to normalize an output of each of the plurality of encoder layers to control a gradient flow and accelerate a training process”. The motivation for doing so would have been that “it is convenient for the passback of the flowing and error of gradient” and training, as suggested by Hou in para. [0033] and [0039]. Therefore, it would have been obvious to one of ordinary skill at the time the invention was filed to combine Xiao and Ru with Hou to obtain the invention specified in claim 2.
Claim 5 is rejected under 35 U.S.C. 103 as being unpatentable over Xiao et al. (CN 117074415 A, see attached English translation for citations), hereinafter Xiao in view of Ru et al. (CN 114820464 A, see attached English translation for citations), hereinafter Ru and Wang et al. (CN 118429636 A, see attached English translation for citations), hereinafter Wang.
Regarding claim 5, Xiao and Ru teach the method of claim 3, wherein in step (S2), the detection model is trained through steps of:
(b1) preparing a dataset required for a blanket defect detection task, wherein the dataset comprises labelled image samples comprising a defective image and a non- defective image, labels respectively corresponding to the defective image and the non- defective image (Xiao teaches “collecting data sample sets of various different blanket yarn breakage conditions, and marking the characteristics of the inner and outer yarn breakage, so as to ensure that the data sample sets contain the characteristics of real samples and corresponding labels” in para. [0068]);
(b2) constructing the detection model based on a Transformer model (Xiao teaches “based on example 2, the intelligent wire breakage detection model in step 2 is based on a transducer model and is built by the following steps, as shown in fig. 3” in para. [0076]);
(b3) dividing, by the detection model, the blanket knitting image into a plurality of patches (Xiao teaches dividing the image into a plurality of patches in para. [0089]) and performing data augmentation on the dataset to increase diversity of the labelled image samples (Xiao teaches “generating a new sample: and (3) using a trained generator, giving a condition label and a noise vector, generating a new sample through the generator, ensuring that the generated new sample accords with expected characteristics and distribution, enriching a defect detection data set, and facilitating subsequent model training” in para. [0074]);
(b4) selecting a binary cross-entropy loss function as a loss function for the blanket defect detection task (Xiao teaches “calculating the difference between the prediction result of the intelligent broken wire detection model and the real label by using a binary cross entropy loss function, and minimizing the difference; guiding the model to optimize parameters in the training process so as to improve classification performance” in para. [0085]);
(b5) based on the dataset and the loss function, training the detection model by using a backpropagation algorithm through steps of: inputting the plurality of patches into the detection model to obtain an output; comparing the output with the labels to calculate a loss; and updating parameters of the detection model based on the loss to optimize performance of the detection model (Xiao teaches “training an intelligent broken wire detection model through a BP back propagation algorithm by using the blanket weaving image subjected to data pretreatment in the step 2.1 and the loss function defined in the step 2.6; in the training process, an input blanket weaving image is sent into a model to obtain the output of the model, the output is compared with a real label, the loss is calculated, and model parameters are updated according to the loss so as to optimize the performance of the model” in para. [0086]);
(b6) evaluating a trained detection model using an independent test set; and calculating an accurate rate, a precision ratio, and a missing rate to evaluate performance of the trained detection model in terms of the blanket defect detection task (Xiao teaches “evaluating the trained intelligent broken wire detection model by using an independent test set, and calculating indexes of accuracy, precision and omission factor to evaluate the performance of the model on a defect detection task” in para. [0087]);
(b7) adjusting hyperparameters of the trained detection model according to actual requirements and performance evaluation results to improve the performance of the trained detection model; wherein the hyperparameters comprise a learning rate, a batch size, and the number of training iterations (Xiao teaches “the super parameters in the step 2.9 include a learning rate, a batch size, and a training iteration number” in para. [0046] wherein “according to actual requirements and performance evaluation results, super parameters of the intelligent broken wire detection model are adjusted to further improve the performance of the model” as shown in para. [0041]); and
(b8) performing defect detection by using the trained detection model through steps of dividing a new image into a plurality of new patches; and predicting classification results or defect location information of each of the plurality of new patches by the trained detection model, so as to complete the defect detection (Xiao teaches “performing defect detection by using a trained intelligent broken wire detection model, giving a new image, dividing the new image into a plurality of image blocks, and predicting the classification result or defect position information of each image block through the model so as to realize the target of defect detection” in para. [0089]).
Xiao fails to teach bounding box information of the defective image and dividing an image into patches with a fixed size; normalizing a pixel value of each of the plurality of patches to a preset range.
Wang teaches bounding box information of the defective image (Wang teaches “the bounding box represents the smallest bounding rectangle of the target object, i.e., the target object is close to the four edges of the bounding box, and the bounding box does not overlap other regions except for the target object itself. These properties are critical to building positive and negative sample sets that are used in the MILs framework to train the model” in para. [0057]. Here, these bounding boxes are applied to defective images (see also para. [0066]-[0071])) and dividing an image into patches with a fixed size (Wang teaches “the input image is first divided into a series of fixed-size blocks, typically squares. Each block is considered to be an input unit of the feature encoding network” in para. [0048]);
normalizing a pixel value of each of the plurality of patches to a preset range (Wang teaches “normalization of pixel values to ensure data quality and stability of model training” as shown in para. [0040]. Here, the normalization process inherently involves normalizing the pixel values to a preset range, as the process of normalization in the context of pixel values involves adjusting the range of values in an image so that they fall within a standard scale).
Xiao, Ru, and Wang are all considered to be analogous to the claimed invention because they are in the same field of analyzing images to determine defects. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of Xiao (as modified by Ru) to incorporate the teachings of Wang and include “bounding box information of the defective image and dividing an image into patches with a fixed size; normalizing a pixel value of each of the plurality of patches to a preset range”. The motivation for doing so would have been “to build[] positive and negative sample sets that are used in the MILs framework to train the model” and “help[] the feature extraction network to further encode the raw data”, as suggested by Wang in para. [0057] and [0048]. Therefore, it would have been obvious to one of ordinary skill at the time the invention was filed to combine Xiao and Ru with Wang to obtain the invention specified in claim 5.
Allowable Subject Matter
Claim 4 would be allowable if rewritten or amended to overcome the rejection(s) under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), 2nd paragraph and U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), 1st paragraph, set forth in this Office action.
The following is a statement of reasons for the indication of allowable subject matter.
The best prior art of record is Xiao, Ru, Hou, Wang, Zhang et al. (CN 109632817 A, see attached English translation), hereinafter Zhang, Chen et al. (CN107144570B, see attached English translation), hereinafter Chen ‘570, and Chen et al. (CN-212611220-U), hereinafter Chen ‘220 . Prior art applied alone or in combination with fails to anticipate or render obvious claim 4.
Claim 4
Regarding claim 4, Xiao and Ru teach the method of claim 3.
Xiao additionally teaches wherein step (S1) comprises:
(a1) collecting a data sample set involving various types of filament breakages in the blanket for labelling inner and outer broken filament characteristics;
(a2) generating, by a generative adversarial network, a plurality of virtual images with a boundary defect feature to expand the data sample set and provide more training data.
However, neither Xiao, nor Hou, nor Wang, nor Zhang, nor Ru, nor Chen ‘570, nor Chen ‘220, nor Xiao, nor the combination, teaches (a3) extracting edge information from an image set comprising the plurality of virtual images and the blanket knitting image by using a Laplacian operator; calculating an edge change rate based on the edge information to evaluate a blur degree of the image set; and removing an image with the blur degree exceeding a pre-set threshold from the image set, in combination with the other elements of the claim, which includes the corresponding algorithm of the detection model as outlined in para. [0047]-[0055] and [0065]-[0081].
***Please note that the detection model is being interpreted under 112(f) as a computer-implemented means-plus-function limitation, wherein the corresponding algorithm of the detection model is being read into the limitation: “the detection model is configured to divide the blanket knitting image into a plurality of patches, convert the plurality of patches into sequence data, process the sequence data”. While the specification does not include enough information to determine the structure for the detection model to determine whether or not to interpret the detection model and its corresponding functional language as a computer-implemented 112(f) limitation, for the purposes of prior art, since the detection model is based on a Transformer model (see para. [0013] of applicant’s specification), it is assumed that applicant intends for the corresponding structure of the detection model to be a processor. See also the 112(a)/112(b) rejections of claim 1. As such, solely for the purposes of prior art searching, the “detection model” is being interpreted as computer-implemented 112(f), wherein the structure for the detection model is a processor and the corresponding algorithm which can be found in para. [0047]-[0055] and [0065]-[0081] of applicant’s specification. In this instance, the structure corresponding to a 35 U.S.C. 112(f) claim limitation for a computer-implemented function must include the algorithm needed to transform the structure disclosed in the specification (see reasoning above regarding computer-implemented 112(f) interpretation despite lack of structure for the detection model in the specification). See MPEP 2181(II)(B). The specific information in the specification regarding the algorithm associated with the detection model that makes this limitation allowable when analyzed in conjunction with the rest of the claim elements includes, but is not limited to, para. [0047]-[0055] and [0065]-[0081].
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 extension fee 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 date of this final action.
Contact
Any inquiry concerning this communication or earlier communications from the examiner
should be directed to KYLA G ALLEN whose telephone number is (703)756-5315. The examiner can
normally be reached M-F 7:30am - 4:30pm EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a
USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use
the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor,
John Villecco can be reached on (571) 272-7319. 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.
/Kyla Guan-Ping Tiao Allen/
Examiner, Art Unit 2661
/JOHN VILLECCO/Supervisory Patent Examiner, Art Unit 2661
