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 . This action is in response to the application filed on July 18, 2024. Claims 1-7 are pending in the application and have been examined.
Priority
Acknowledgment is made of applicant’s claim for foreign priority under 35 U.S.C. 119 (a)-(d). The certified copy has been filed in parent Application No. KR10-2023-0099076, filed on July 28, 2023.
Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55.
Information Disclosure Statement
The information disclosure statements (IDS) submitted on July 18, 2024, and May 12, 2026, are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statements are being considered by the examiner.
Claim Rejections - 35 USC § 102
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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 1,2,4,5,7 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by König et al., “Optimized deep encoder-decoder methods for crack segmentation”.
Regarding Claim 1, König et al. teaches a method of training a deep learning model, performed by a computing device comprising at least a processor (König, pg. 6, sec. 4.3, “training and evaluation were performed on a NVIDIA Titan XP GPU”, where the training of König’s system requires a GPU, denoting a computing device comprising at least a processor), the method comprising: generating a basic model; and training the basic model (König, pg. 2, sec. 1, “We therefore propose to report results after performing a statistical analysis and do so by showing the performance of our model … over ten training runs”, where to implement König’s system, they must train the model denoting training the basic model, further generating a basic model is implicit given that training a model is already being performed, and thus one would have to had to generate one to do so), wherein the basic model includes a plurality of layer blocks each including at least one layer (König, pg. 4, sec. 3.2, “Each decoder block in those levels contains a standard Conv-BN-ReLU sequence followed by two residual blocks.”, where the recitation of a plurality of blocks denotes a plurality of layer blocks and the blocks consisting of a Conv-ReLU sequence denotes including at least one layer), and the layer blocks include first layer blocks configured to receive input data or output of a previous layer block to compress or encode data, or to output a latent vector corresponding to the input data and second layer blocks configured to receive the latent vector and the output of the previous layer block to restore or decode the data (König, pg. 4, sec. 3.2, “Our proposed decoder can be added to a variety of encoder backbones. However, encoders which aim to be used with our proposed decoder, need to have six levels, each distinguishable by their different spatial size. These are addressed with level numbers l=1 to l=6, each having half the spatial size of its previous one. Hence, an encoder needs to have five down sampling layers. … The input into the decoder sequence is the up-sampled output of the sixth level of the encoder backbone”, where the variety of encoder backbones present at multiple levels of the architecture and consisting of a plurality of layers denotes encoding layer blocks, the input into the decoder being the output of the encoder backbone denotes to output a latent representation corresponding to the input to the decoder layer blocks), or to derive inference results corresponding to the input data (König, pg. 9, sec. 4.9, “During inference, … the model architecture can create predictions with a speed of 11.7 frames per second on images”, where the model output deriving inference on input images denotes inference results corresponding to the input data).
Regarding Claim 2, König et al. teaches the method of claim 1 (and thus the rejection of claim 1 is incorporated), wherein: the first layer blocks are sequentially assigned levels from a first level to an N-th level, N denoting an arbitrary natural number, the second layer blocks are assigned levels in reverse order from the first level to the N-th level (König, pg. 4, sec. 3.2, “Our proposed decoder can be added to a variety of encoder backbones. However, encoders which aim to be used with our proposed decoder, need to have six levels …. These are addressed with level numbers l=1 to l=6, …. The decoder blocks are thus situated in levels l=1 to l=5 and are connected to the encoder through U-Net style skip connections”, where the assigned levels for the decoder/encoder blocks denote sequentially assigned levels from a first level to a N-th level), the training of the basic model comprises training the layer blocks in ascending order of levels (König, pg. 3, Fig. 2)
PNG
media_image1.png
352
712
media_image1.png
Greyscale
Annotated Fig. 2, König et al.
Regarding Claim 4, König et al. teaches the method of claim 1 (and thus the rejection of claim 1 is incorporated), wherein the basic model further includes at least one additional layer block that follows the plurality of layer blocks (König, pg. 4, sec. 3.2, “A final segmentation output is then generated through running the resulting feature maps of the decoder block at level l=1through a 3 ×3 convolutional layer with 1 filter and a Sigmoid activation function”, where a final decoder block that consists of additional convolution and activation function denotes at least one additional layer block following a plurality of other layer blocks).
Regarding Claim 5, König et al. teaches the method of claim 4 (and thus the rejection of claim 4 is incorporated), further comprising: performing training of the at least one additional layer block (König, pg. 4, sec. 3.2, “A final segmentation output is then generated through running the resulting feature maps of the decoder block at level l=1 through a 3 ×3 convolutional layer with 1 filter and a Sigmoid activation function”, where a block that consists of additional convolution layers denotes at least one additional trainable layer block following a plurality of other layer blocks).
Regarding Claim 7, König et al. teaches the method of claim 1 (and thus the rejection of claim 1 is incorporated), wherein the deep learning model of which training is completed has a plurality of inference paths (König, pg. 4, sec. 3.4, “Recall that this architecture contains a total of five outputs, four interim deeply supervised outputs at level l = 2 to l =5 and the final segmentation output”, where given the architecture has 5 outputs, it would mean that the model has a plurality of inference paths in order to reach/obtain those 5 outputs, thus would correspond to the plurality of inference paths recited).
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 3,6 is/are rejected under 35 U.S.C. 103 as being unpatentable over König et al., “Optimized deep encoder-decoder methods for crack segmentation” in view of Lau et al., “Automated Pavement Crack Segmentation Using U-Net-Based Convolutional Neural Network”.
Regarding Claim 3, König et al. teaches the method of claim 2 (and thus the rejection of claim 2 is incorporated).
König does not teach, but Lau teaches wherein the training of the basic model comprises performing training of layer blocks of a current level while freezing a parameter for layer blocks of a previous level (Lau, pg. 114895, sec. III(D)(1), “In the first stage, we “freeze” the first layer-group, i.e., temporarily set the learning rate to zero so that all their parameters are not updated. We train the second and third layer-groups normally for 15 epochs.”, where freezing a first layer-group while training further layer-groups denotes freezing a previous layer block parameter(s) while performing training of the current level blocks). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to implement parameter freezing into training a deep learning model. The motivation to do so being Lau’s approach “requires minimal feature engineering compared to other machine learning techniques; hence, it is useful when there is a lack of domain expertise” (Lau, pg. 114898, sec. VI).
Regarding Claim 6, König et al. teaches the method of claim 5 (and thus the rejection of claim 5 is incorporated).
König does not teach, but Lau teaches wherein the performing training of the at least one additional layer block is performed while freezing a parameter of each of the layer blocks (Lau, pg. 114895, sec. III(D)(1), “In the first stage, we “freeze” the first layer-group, i.e., temporarily set the learning rate to zero so that all their parameters are not updated. We train the second and third layer-groups normally for 15 epochs”, where freezing a previous first layer-group while training the current layer-group denotes freezing previous layer block parameters while performing training of an additional layer block). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to implement parameter freezing into training a deep learning model. The motivation to do so being Lau’s approach “requires minimal feature engineering compared to other machine learning techniques; hence, it is useful when there is a lack of domain expertise” (Lau, pg. 114898, sec. VI).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HAYDEN Y ARDALE whose telephone number is (571)270-5128. The examiner can normally be reached Monday-Thursday (7:30-5).
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, Kakali Chaki can be reached at (571) 272-3719. 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.
/H.Y.A./Examiner, Art Unit 2122
/MICHAEL H HOANG/PRIMARY EXAMINER, Art Unit 2122
Prosecution Insights