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 .
Status of the Claims
Claims 1, 3, 5, 7-14, 16, and 18-20 stand rejected. Claims 2, 4, 6, 15, and 17 were cancelled by the applicant.
Response to Amendment
Applicant’s amendments to the claims, specification, and drawings, received 03/18/2026, have been entered into the record.
Applicant’s amendments to the drawings overcome the objection set out in the previous Office Action.
Applicant’s amendments to the specification do not overcome the objection set out in the previous Office Action. As stated in the previous Office Action, the equation of para. 0088 was exemplary of the unreadable characters in the equations but is not the only equation with the same issue. The equations of para. 0087, 0127, 0128, 0131, 0132, 0158, 0162, and 0164-0169 contain similar legibility issues that must be resolved by ethe applicant.
Applicant’s amendments to claim 3 do not overcome the objection set out in the previous Office Action because the applicant has not made the necessary correction, as set out in the previous Office Action. That objection is reproduced below.
Applicant’s amendments to claims 1, 12, and 14 do not overcome the rejection under 35 U.S.C. 112(b) set out in the previous Office Action. For example, the second line of claim 1 still reads, “each of classes,” a non-standard English construction that makes it unclear whether the same reference reflection is stored for every class or a separate reference reflection is stored for each class.
Response to Arguments
Applicant’s arguments, filed 03/18/2026, have been fully considered but are not persuasive.
The examiner understands the applicant to be making the following three arguments regarding the rejections under 35 U.S.C. 101, 102, and 103 of claim 1:
Claim 1 recites patent eligible subject matter because it does not recite an abstract idea, and, even if an abstract idea is recited, it is integrated into a practical solution and presents significantly more than the abstract idea (p. 8, para. 4-–p. 10, para. 1).
The examiner fails to set out a prima facie case of obviousness for the limitations of amended claim 1, which appeared in claim 2 of the previous claim set, because Bilik teaches a method of comparing the head part of the object whereas the claimed invention compares the overall shape of the object (p. 10, para. 7).
The examiner further fails to set out a prima facie case of obviousness for the limitations of amended claim 1, which appeared in claim 2 of the previous claim set, because Bilik is silent as to “modeling intensity of radar reflection signal data generated by the radar simulation signal generator as a mixed normal distribution.” (p. 10, para. 7).
Each of these arguments is carefully considered and addressed below, but only the argument regarding subject matter eligibility is persuasive.
Subject Matter Eligibility
The applicant’s arguments, see p. 9, have been carefully considered and are persuasive. The judicial exception of claim 1 is sufficiently integrated into a particular machine, because the reference reflections are required to be generated by a radar simulation signal generator as a mixed normal distribution for each object class,
Head Part vs. Overall Shape
In response to applicant's argument that the references fail to show certain features of the invention, it is noted that the features upon which applicant relies (i.e., modeling and comparing the overall shape of the object, see p. 9 of the applicant’s remarks) are not recited in the rejected claim(s). Although the claims are interpreted in light of the specification, limitations from the specification are not read into the claims. See In re Van Geuns, 988 F.2d 1181, 26 USPQ2d 1057 (Fed. Cir. 1993).
Modeling Intensity as a Mixed Normal Distribution
The applicant further argues that the prior art used in the previous rejection does not teach, “modeling intensity of radar reflection signal data generated by the radar simulation signal generator as a mixed normal distribution.” Therefore, the applicant argues, the combination of Bilik and Zhu does not teach the claimed invention. The examiner agrees that Zhu in view of Bilik does not teach said limitation, but disagrees that Zhu in view of Bilik and further in view of Matlab does not render the invention obvious. As can be seen in fig. 3 of Bilik, reproduced below, the sampled target echoes used to train the classification model are formed as spectrograms.
PNG
media_image1.png
548
971
media_image1.png
Greyscale
Spectrograms are, by definition, a measurement of intensity. These spectrograms were then used to form the database for the Gaussian mixture model (see Summary & Conclusions, “An automatic target recognition (ATR) algorithm, based on Greedy learning of Gaussian mixture model (GMM) is developed in this work. The GMMs were obtained for a wide range of ground surveillance radar targets such as: walking person(s), tracked or wheeled vehicles, animals and clutter.” See also section 2.1, para. 1, “In this work, each target class is represented by a GMM.” See also section 4, para. 1, “In this work, the target class pdfs were modeled by GMMs”). A Gaussian mixture model is a mixed normal distribution (see section 2.1, para. 1, “A Gaussian mixture density, defined as a weighted sum of K. Gaussian component densities, is a useful tool for pdf modeling.”). The examiner notes that “Gaussian distribution” is a synonym for “normal distribution .”
The sole difference between Zhu in view of Bilik and the claimed invention is that the spectrograms used to form the GMM in Bilik and to train the neural network in Zhu is that the spectrograms of Bilik and Zhu are formed using real data, while the claimed invention uses simulated signals to form the mixed normal distribution. However, the invention of Matlab shows that simulating radar reflections is a common technique in the art, thus rendering the claimed invention obvious.
For the reasons above, claims 1, 3, 5, 7-14, 16, and 18-20 stand rejected.
Specification
The disclosure is objected to because of the following informalities: The equations listed throughout the disclosure are blurry enough so as to render them unreadable, as are many of the special characters used in description of the equations. Appropriate correction is required.
Claim Objections
Claim 3 is objected to because it reads, “for each of the classes classifying the objects; determining, among the classes,” but should read, “for each of the classes classifying the objects, determining, among the classes”.
Claim 3 is objected to because of the following informalities: claim 3 recites "on relative distance and angle plane...based on Fast Fourier Transform" but should recite "on a relative distance and angle plane...based on a Fast Fourier Transform". Appropriate correction is required.
Claim 12 is objected to because it reads, “a number predetermined” but should read, “a predetermined number. Appropriate correction is required.
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 1-20 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 recites the limitations "each of classes" and “each of objects” in lines 2-3. There is insufficient antecedent basis for this limitation in the claim.
Claim 6 recites the limitation “to each of predetermined object sizes” in lines 2-3. There is insufficient antecedent basis for this limitation in the claim.
Claim 12 recites the limitation, “normalizing the similarity based on a number of the classes.” The metes and bounds of this limitation are unclear because the meaning of “a number of the classes” is not clear from the claims. A person of ordinary skill in the art would not know whether “a number of the classes” refers to a reference number, a measurement that is associated with each class, or simply the total number of classes being taken into account.
Claims 2-13 are rejected because they depend upon rejected claim 1.
Claim 14 is rejected for the same reasons as claim 1.
Claims 15-20 are rejected because they depend upon rejected claim 14.
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.
Claims 1, 5, 7-8, 10-14, 15, and 18-19 are rejected under 35 U.S.C. 103 as being unpatentable over Zhu in view of Bilik and further in view of Matlab, all cited in full in the previous office action.
Regarding claim 1, Zhu teaches (note: what Zhu does not teach is struck through),
A method for recognizing an object (para. 0002, “The present application relates to the field of computer vision technology, and in particular to an object recognition method and device, and a storage medium.”), the method comprising: storing a reference reflection characteristic for each of classes based (para. 0043, “Thus, it is possible to determine the category to which the object to be recognized belongs by comparing the similarity between the head shape of the object to be recognized and the head shape corresponding to each of the plurality of categories”), determining, among the classes, a class of a reference reflection characteristic of a high similarity with a reflection characteristic of received signal data transmitted from a radar; and identifying a target object of the received signal data based on the determined class and outputting information of the target object (para. 0042, “Therefore, in some embodiments, objects belonging to different categories have different head shapes; and determining, according to the target feature, the target category to which the object to be recognized belongs among the plurality of categories may include: determining, according to the target feature, a similarity between a head shape of the object to be recognized and a head shape corresponding to each of the plurality of categories to obtain a plurality of similarities; and determining a category corresponding to the maximum similarity among the plurality of similarities as the target category.” The examiner notes that para. 0029 specifies that the data is from a radar).
Bilik teaches (note: what Bilik does not teach is struck through),
…storing a reference reflection characteristic for each of classes based on modeling intensity of radar reflection signal data (section 2.3, para. 1, “The GMMs are estimated from the training database in an offline training stage. The detected radar target is classified into one of possible classes using models estimated in the training stage.” The examiner notes that GMM as used in this reference refers to Gaussian mixture modeling and further notes that a Gaussian distribution is a normal distribution).
Matlab teaches generating radar simulation signals using a radar simulation signal generator (p. 1, para. 3, “The vehicle ground truth can then be used to generate synthetic sensor detections.”)
Zhu and Bilik are analogous to the claimed invention because they teach methods of classifying objects using radar data. It would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Zhu with the Gaussian mixture modeling (i.e., mixed normal distribution) of Bilik because the GMM of Bilik has a maximum accurate classification rate of 96%, which outperforms human operators. It would further be obvious to replace the real signals used by Bilik and Zhu with the simulation signal generator of Matlab because it is faster and cheaper to simulate radar signals for use in modeling than it is to generate a database of real signals, as evidenced by the limited signal database used to train the GMM of Bilik. Using simulated signals increases the variety of objects that can be modeled by the database, thus increasing the precision with which objects can be classified.
Regarding claim 5, Zhu in view of Bilik and further in view of Matlab teaches the method of claim 1. Zhu further teaches
…wherein the classes include one or more classes selected from a class corresponding to a two-wheeled vehicle, a class corresponding to a passenger vehicle, and class corresponding to a commercial vehicle (paras. 0032 and 0091 both indicate “car” (a passenger vehicle) and “truck” (a commercial vehicle) as potential object categories).
Regarding claim 7, Zhu in view of Bilik and further in view of Matlab teaches the method of claim 1. Zhu further teaches,
…wherein the determining the class includes: applying the received signal data to a radar reflection characteristic model and obtaining the reflection characteristic with respect to a predetermined reference distance (para. 0040, “Assuming that the to-be-processed point cloud data includes M points, and coordinates of each point are expressed as (X, Y, Z), the points in the target point cloud data may be expressed in the form of W×H×N×(Xi, Yi, Zi), where N indicates there are N points in each target geometry and may generally be set or adjusted according to requirements such as accuracy requirements, and W×H indicates a preset range of the point cloud.”), and determining a similarity between the reference reflection characteristic for each of the classes and the reflection characteristic of the received signal data (para. 0042, “determining, according to the target feature, a similarity between a head shape of the object to be recognized and a head shape corresponding to each of the plurality of categories to obtain a plurality of similarities”).
Regarding claim 8, Zhu in view of Bilik and further in view of Matlab teaches the method of claim 7. Zhu further teaches,
…obtaining, from the radar, information indicating a relative distance and an observation angel between the radar and the target object, wherein the obtaining the reflection characteristic with respect to the predetermined reference distance comprises: when applying the received data to the radar reflection characteristic model, applying the information indicating the relative distance and the observation angle to the radar reflection characteristic model (para. 0038, “In some implementations, e.g., in driving environments where objects are moving substantially within a specific plane (e.g., parallel to ground), the radar intensity map and the Doppler map can be defined using two-dimensional coordinates, such as the radial distance and azimuthal angle: (R, ϕ), Δf (R, ϕ).”).
Regarding claim 10, Zhu in view of Bilik and further in view of Matlab teaches the method of claim 7. Zhu further teaches,
…obtaining detection information for determining location information of each of the objects from the radar (para. 0041, “In this way, compared with the independent scattered points in the to-be-processed point cloud data, the present application traverses the to-be-processed point cloud data by the target geometry at the target step length to obtain the target point cloud data containing location information, such that the to-be-processed point cloud data without structured information may be represented as the target point cloud data containing the structured information, which helps to obtain more accurate semantic features, thereby improving the accuracy of object recognition.”), wherein the similarity is based on the detection information (para. 0042, “…determining, according to the target feature, a similarity between a head shape of the object to be recognized and a head shape corresponding to each of the plurality of categories to obtain a plurality of similarities”).
Regarding claim 11, Zhu in view of Bilik and further in view of Matlab teaches the method of claim 10. Zhu does not teach,
…wherein determining the similarity comprises: applying a weight, an average and a variance of the reflection characteristic of the received signal data and the detection information to a mixed normal distribution model to determine the similarity between the reference reflection characteristic for each of the classes and the reflection characteristic of the received signal data
Bilik teaches,
…wherein determining the similarity comprises: applying a weight, an average and a variance of the reflection characteristic of the received signal data and the detection information to a mixed normal distribution model to determine the similarity between the reference reflection characteristic for each of the classes and the reflection characteristic of the received signal data (eqs. 6, noting that the pdf of a classification feature is understood to be its probability density function, which takes into account a mixing weight and the Gaussian for a particular component, which includes its mean and variance. See, e.g., the documentation for the pdf function in MATLAB included with this office action. The examiner further notes that both equations labeled with the number 6 include the probability density function fK--).
It would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Zhu with the similarity determination of Bilik that uses weight, average, and variance to classify data because weighing data based on average and variance is a well-known technique in the art to measure the similarity between different datasets.
Regarding claim 12, Zhu in view of Bilik and further in view of Matlab teaches the method of claim 11. Zhu does not teach,
…further comprising: determining a reference similarity for each of the classes by normalizing the similarity based on a number of classes
Bilik teaches,
…further comprising: determining a reference similarity for each of the classes by normalizing the similarity based on a number of classes (section 2.3.2, para. 2, “Thus, the LRT is performed between each pair of hypotheses, and the corresponding threshold is optimized independently of all other tests. The optimal threshold, γm, is determined to minimize the classification error for the pair of target classes, (m,n). The pair-wise decisions are combined by voting, that is the class with the most pair-wise wins is selected.” The examiner notes that the majority voting concept is a method of normalizing the similarity based on the number of classes, since similarity is compared between each pair of classes to determine “winners,” with the class with the most votes being the identified class. That is, the number of classes affects the number of votes, thus affecting how much each similarity comparison matters).
It would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Zhu with the pairwise majority voting technique of Bilik because the technique of Bilik enables adjusting estimation thresholds for each pair, thus outperforming ML-based decision-making in the presence of modeling errors (see Bilik, section 2.3.2).
Regarding claim 13, Zhu in view of Bilik and further in view of Matlab teaches the method of claim 12. Zhu does not teach,
…wherein the determining the class comprises: identifying one or more classes having the reference similarity exceeding a threshold value among the classes, and identifying a class having a highest similarity among the identified one or more classes as the class of the reference reflection characteristic of the high similarity with the reflection characteristic of the received signal data transmitted from the radar
Bilik teaches,
…wherein the determining the class comprises: identifying one or more classes having the reference similarity exceeding a threshold value among the classes, and identifying a class having a highest similarity among the identified one or more classes as the class of the reference reflection characteristic of the high similarity with the reflection characteristic of the received signal data transmitted from the radar (section 2.3.3, para. 2, “The “majority voting” decision rule enables to adjust the thresholds for each pair in order to minimize the cost function evaluated using the training database. Thus, the LRT is performed between each pair of hypotheses, and the corresponding threshold is optimized independently of all other tests. The optimal threshold, γm, is determined to minimize the classification error for the pair of target classes, (m,n). The pair-wise decisions are combined by voting, that is the class with the most pair-wise wins is selected.” The examiner notes that any class that has a similarity that optimizes the pairwise threshold γm is identified as exceeding said threshold by its reception of a vote, and the class with the highest similarity is identified by receiving the majority of these votes).
It would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Zhu with the thresholding of Bilik because the thresholding of Bilik, as used in the AVA majority voting model, reduces the effect of modeling errors on the final classification decision (see Bilik, section 2.3.2).
Claim 14 is rejected for the same reasons and using the same references as claim 1. The examiner notes that Zhu further teaches an apparatus, a memory, and a processor (para. 0015, “A third aspect of the present application features an object recognition device. The device includes at least one processor; and one or more memories coupled to the at least one processor and storing programming instructions for execution by the at least one processor to perform operations”).
Claim 18 is rejected for the same reasons and using the same citations as claim 5.
Claim 19 is rejected for the same reasons and using the same citations as claim 7.
Claims 3 and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Zhu in view of Bilik and further in view of Matlab as applied to claims 1 and 14, respectively, above, and further in view of Xia et al. (U.S. Pub. No. 2023/0351243 A1), hereinafter Xia.
Regarding claim 3, Zhu in view of Bilik and further in view of Matlab teaches the method of claim 1. Zhu does not teach,
…wherein the intensity of the radar reflection signal data includes and intensity of the radar reflection signal on relative distance and angle plane extracted from a radar data cube generated based on Fast Fourier Transform of the radar reflection signal data
Xia teaches,
…wherein the intensity of the radar reflection signal data includes and intensity of the radar reflection signal on relative distance and angle plane extracted from a radar data cube generated based on Fast Fourier Transform of the radar reflection signal data (para. 0038, “the radar intensity map and the Doppler map can be defined using two-dimensional coordinates, such as the radial distance and azimuthal angle: (R, ϕ), Δf (R, ϕ).” The examiner notes that FFTs are used to generate radar intensity maps).
Xia is analogous to the claimed invention because it teaches radar-based object identification. It would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Zhu in view of Bilik with the radial distance-azimuthal angle plane extraction of Xia because objects that are moving along the ground have relatively consistent elevation angles. Therefore, extracting a 2D coordinate graph for the purposes of classification reduces processing needs without significantly affecting data accuracy.
Claim 16 is rejected for the same reasons and using the same citations as claim 3.
Claims 9 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Zhu in view of Bilik and further in view of Matlab as applied to claims 8 and 19 above, respectively, and further in view of Wodrich et al. (U.S. Pub. No. 2018/0067194 A1), hereinafter Wodrich.
Regarding claim 9, Zhu in view of Bilik and further in view of Matlab teaches the method of claim 8. Zhu further teaches (note: what Zhu does not teach is struck through),
…wherein obtaining the reflection characteristic with respect to the predetermined reference distance further comprises: when applying the received data to the radar reflection characteristic model, applying a predetermined radar distance (para. 0040, “Assuming that the to-be-processed point cloud data includes M points, and coordinates of each point are expressed as (X, Y, Z), the points in the target point cloud data may be expressed in the form of W×H×N×(Xi, Yi, Zi), where N indicates there are N points in each target geometry and may generally be set or adjusted according to requirements such as accuracy requirements, and W×H indicates a preset range of the point cloud.”).
Wodrich teaches
…wherein obtaining the reflection characteristic with respect to the predetermined reference distance further comprises: when applying the received data to the radar reflection characteristic model, applying a predetermined radar distance and a predetermined angular resolution of the radar to the radar reflection characteristic model (para. 0016, “For both systems, the FOV of the sensor controls what can be seen, and at what location relative to the source or equipped vehicle. For radar systems, this is further effected by the effective range and angular resolution of the radar, controlled respectively by the available signal bandwidth and the beam shape defined by the antenna design.”).
Wodrich is analogous to the claimed invention because it teaches object classification using radar sensors. It would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Zhu with the known angular resolution of Wodrich because taking angular resolution into account increases the accuracy of classification.
Claim 20 is rejected for the same reasons and using the same citations as claim 9.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Anna K Gosling whose telephone number is (571)272-0401. The examiner can normally be reached Monday - Thursday, 7:30-4:30 Eastern, Friday, 10:00-2:00 Eastern.
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, Vladimir Magloire can be reached at (571) 270-5144. 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.
/Anna K. Gosling/Examiner, Art Unit 3648
/VLADIMIR MAGLOIRE/Supervisory Patent Examiner, Art Unit 3648
Prosecution Insights