DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
This communication is responsive to the amendment to the original application. This action is Final. Claims 1-20 are pending and have been examined.
Response to Amendments
In the reply filed 8/21/24, claims 1, 13 and 20 were amended. Accordingly, claims 1 – 20 are pending.
Response to Arguments
Applicant's arguments with respect to claims 1 – 20 have been carefully considered but are moot and not deemed persuasive in view of rejections below.
Examiner has withdrawn the 101 rejections. However, examiner maintains the 103 rejections and respectfully disagrees with applicant’s arguments on pages 13 – 15, that prior art fails to teach, obtaining, via inputting the coarse-grained feature and the fine-grained feature into a second neural network (Chen [0006]: In some embodiments, the determining a plurality of candidate classifications of the structure by inputting the image data into a classification model may include obtaining the backbone feature, the segmentation feature, and the density feature by inputting the image data into the backbone network, the segmentation network, and the density classification network, respectively, and determining a probability of each of the plurality of candidate classifications of the structure based on the backbone feature, the segmentation feature, and the density feature.). Therefore, examiner is not persuaded.
All claims have been updated below with clarifying prior art citations. Kindly let me know if you have any questions. Thanks.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1 – 20 are rejected under 35 U.S.C. 103 as being unpatentable over Liu et al., U.S. Patent Application Publication No.: 2021/0383231 (Hereinafter “Liu”), and further in view of Chen et al., U.S. Patent Application Publication No.: 2020/0210761 (Hereinafter “Chen”).
Regarding claim 1, Liu teaches, an information pushing method, performed by at least one processor (Liu [0037]: processor) and comprising:
extracting an information feature of candidate information from an image, the information feature comprising a coarse-grained feature and a fine-grained feature, a number of tail value samples of the coarse-grained feature being greater than a number of tail value samples of the fine-grained feature (Liu [0040]: The present invention discloses the target cross-domain detection and understanding method. … The method of detecting and understanding a notable target according to a specific intention is efficient, objective and comprehensive, and can effectively enhance the environmental visual perception ability and active safety. Meanwhile, in conjunction with a coupling relationship between a coding position probability and image features, diagonal vertexes of a target candidate frame are efficiently located, network complexity is simplified, the difficulty and redundancy of fusion are avoided, the expenses of system calculation are saved, and the application needs for actual detection can be met.);
Liu does not clearly teach, obtaining, via inputting the coarse-grained feature into a first neural network, a first feature of the candidate information based on an intermediate feature, the intermediate feature being obtained in a process of extracting the coarse-grained feature; However, Chen [0013] teaches, “The method may include obtaining, by the computing device, image data representing a structure of a subject, determining, by the computing device, a plurality of candidate classifications of the structure and their respective probabilities by inputting the image data into a classification model, wherein the classification model includes a backbone network for determining a backbone feature of the structure, a segmentation network for determining a segmentation feature of the structure, and a density classification network for determining a density feature of the structure, and determining, by the computing device, a target classification of the structure based on the probabilities of the plurality of candidate classifications.”
obtaining, via inputting the coarse-grained feature and the fine-grained feature into a second neural network (Chen [0006]: In some embodiments, the determining a plurality of candidate classifications of the structure by inputting the image data into a classification model may include obtaining the backbone feature, the segmentation feature, and the density feature by inputting the image data into the backbone network, the segmentation network, and the density classification network, respectively, and determining a probability of each of the plurality of candidate classifications of the structure based on the backbone feature, the segmentation feature, and the density feature.), a second feature of the candidate information based on the information feature and the intermediate feature (Chen [0091]: In 640, a plurality of candidate classifications and their respective probabilities may be determined based on the backbone feature, the segmentation feature, and the density feature.);
obtaining, based on the first feature and the second feature, target information from a plurality of pieces of candidate information (Chen [0074]: In 530, a target classification of the structure may be determined based on at least a part of the probabilities of the plurality of candidate classifications. In some embodiments, the target classification of the structure may be determined by the classification determination module 430.); and
pushing the target information to one or more devices (Chen [0056]: The classification determination module 430 may determine a target classification of the structure of the subject. In some embodiments, the target classification of the structure may be determined based on probabilities of a plurality of candidate classifications. The plurality of candidate classifications of the structure and their respective probabilities may be determined by inputting the image data into a classification model. The classification model may trained according to a focal loss function, at least one weight of the focal loss function each of which corresponds to one of the plurality of candidate classifications being different from weights of the focal loss function corresponding to the remainder of the plurality of candidate classifications. In some embodiments, the target classification of the pulmonary nodule may be determined by identifying a candidate classification of the structure corresponding to a largest probability among the probabilities of the plurality of candidate classifications. The identified candidate classification may be designated as the target classification of the structure.).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to incorporate the teaching of Liu et al. to the Chen’s system by adding the feature of candidate information data. The references (Liu and Chen) teach features that are analogous art and they are directed to the same field of endeavor, such as databases. Ordinary skilled artisan would have been motivated to do so to provide Liu’s system with enhanced data. (See Chen [Abstract], [0227], [0240], [0276]). One of the biggest advantages of network machine learning database algorithms is their ability to improve over time. Machine learning technology typically improves efficiency and accuracy thanks to the ever-increasing amounts of data that are processed.
Regarding claim 2, the method according to claim 1, wherein the obtaining of the first feature comprises:
performing feature extraction on the coarse-grained feature to obtain at least one first intermediate feature of the candidate information; obtaining a first weight of each of the at least one first intermediate feature based on the coarse-grained feature; and obtaining the first feature of the candidate information based on the at least one first intermediate feature and the first weight of each of the at least one first intermediate feature (Liu [0040]: The present invention discloses the target cross-domain detection and understanding method. The regional weight can be partially reduced through spatial probability control and salient point pooling and by taking a spatial probability control layer as an input image channel, and an edge salient cross point pooling layer can help a network to better locate a target point; through cross-domain guiding semantic extraction and knowledge transfer, an inclusion relation between target depth visual features and guiding semantics for different domains is explored, network training is guided, and cross-domain invariant features are extracted to enhance the cross-domain perception of a model; and by analyzing a target notability degree, a semantic hierarchy cross-domain perception mapping effect and a back propagation mechanism are explored, and a problem of accuracy in notable target prediction and guiding semantic understanding under a specific intention is solved.).
Regarding claim 3, the method according to claim 2, wherein the obtaining of the second feature comprises:
performing feature extraction on the information feature to obtain at least one second intermediate feature of the candidate information; obtaining second weights based on the information feature, the second weights comprising a second weight of each of the at least one second intermediate feature and a second weight of each of the at least one first intermediate feature; and obtaining the second feature of the candidate information based on the second weight of each of the at least one second intermediate feature, the second weight of each of the at least one first intermediate feature, the at least one second intermediate feature, and the at least one first intermediate feature (Chen [0056]: The classification determination module 430 may determine a target classification of the structure of the subject. In some embodiments, the target classification of the structure may be determined based on probabilities of a plurality of candidate classifications. The plurality of candidate classifications of the structure and their respective probabilities may be determined by inputting the image data into a classification model. The classification model may trained according to a focal loss function, at least one weight of the focal loss function each of which corresponds to one of the plurality of candidate classifications being different from weights of the focal loss function corresponding to the remainder of the plurality of candidate classifications. In some embodiments, the target classification of the pulmonary nodule may be determined by identifying a candidate classification of the structure corresponding to a largest probability among the probabilities of the plurality of candidate classifications. The identified candidate classification may be designated as the target classification of the structure.).
Regarding claim 4, the method according to claim 3, wherein the obtaining of the second weights is further based on a popularity vector of the candidate information, the popularity vector indicating historical conversion times of the candidate information (Chen [0032]: At least one weight of the focal loss function each of which corresponds to one of the plurality of candidate classifications being different from weights of the focal loss function corresponding to the remainder of the plurality of candidate classifications. By adjusting the weights corresponding to the plurality of candidate classifications, negative effects induced by imbalance of samples corresponding to different candidate classifications on the classification model may be reduced, and a more robust classification model may be obtained.).
Regarding claim 5, the method according to claim 4, wherein the obtaining of the second weights comprises:
splicing the information feature and the popularity vector to obtain a first spliced feature of the candidate information; and obtaining the second weights based on the first spliced feature (Chen [0094]: For example, the combination of the backbone feature, the segmentation feature, and the density feature may be realized by converting the backbone feature, the segmentation feature, and the density feature into a one-dimensional backbone feature vector, a one-dimensional segmentation feature vector, and a one-dimensional density feature vector, respectively, and splicing the one-dimensional backbone feature vector, the one-dimensional segmentation feature vector, and the one-dimensional density feature vector to obtain a one-dimensional classification feature vector. The one-dimensional classification feature vector may be used to determine the plurality of candidate classifications and the probabilities corresponding to each candidate classification.).
Regarding claim 6, the method according to claim 1, wherein the obtaining of the target information comprises:
fusing the first feature and the second feature to obtain a fused feature of the candidate information; obtaining an estimated event probability of the candidate information based on the fused feature, the estimated event probability marking an estimated probability of a specified event after corresponding information is displayed; and obtaining the target information based on the estimated event probability (Liu [Abstract]: Through spatial probability control and salient point pooling and in conjunction with a coupling relationship between a coding position probability and image features, diagonal vertexes of a target candidate frame are efficiently located, and network complexity is simplified so as to meet application needs for actual detection; through cross-domain guiding semantic extraction and knowledge transfer, an inclusion relation between target depth visual features and guiding semantics for different domains is explored, network training is guided, and cross-domain invariant features are extracted to enhance the cross-domain perception of a model; and by analyzing a target notability degree, a semantic hierarchy cross-domain perception mapping effect and a back propagation mechanism are explored, and a problem of accuracy in notable target prediction and guiding semantic understanding under a specific intention is solved.).
Regarding claim 7, the method according to claim 6, wherein the fusing of the first feature and the second feature comprises:
obtaining a third weight of the second feature based on the information feature; and fusing the first feature and the second feature based on the third weight of the second feature to obtain the fused feature (Liu [0099]: “performing feature fusion on the intention feature and a target visual feature: ƒ.sub.fusion=ƒ(x)⊕ƒ.sub.int, where f(x) represents a visual feature extracted by a target through the NEGSS-Net backbone network; represents a channel splicing operator; f.sub.fusion represents the fused feature;”).
Regarding claim 8, the method according to claim 7, wherein the obtaining of the third weight is further based on a popularity vector of the candidate information, the popularity vector indicating historical conversion times of the candidate information (Chen [0102]: In some embodiments, the training sample set may be generated based on the historical image data, the structures labeled in the image data, and the target classifications corresponding to the structures in the historical image data. … Alternatively, the training sample set may be generated by the processing device 120. For example, the processing device 120 may pre-process historical image data representing structures of a plurality of subject with center points of the structures labeled. Merely by way of example, the historical image data may be resample to generate resampled image data having a target image resolution. Optionally, the processing device 120B may further normalize the resampled image data. As another example, the processing device 120 may extract one or more image crops from the resampled image data, and normalize each of the image crop(s).).
Regarding claim 9, the method according to claim 8, wherein the obtaining of the third weight comprises:
splicing the information feature and the popularity vector to obtain a second spliced feature of the candidate information; and obtaining the third weight of the second feature based on the second spliced feature (Chen [0094]: For example, the combination of the backbone feature, the segmentation feature, and the density feature may be realized by converting the backbone feature, the segmentation feature, and the density feature into a one-dimensional backbone feature vector, a one-dimensional segmentation feature vector, and a one-dimensional density feature vector, respectively, and splicing the one-dimensional backbone feature vector, the one-dimensional segmentation feature vector, and the one-dimensional density feature vector to obtain a one-dimensional classification feature vector. The one-dimensional classification feature vector may be used to determine the plurality of candidate classifications and the probabilities corresponding to each candidate classification.).
Regarding claim 10, the method according to claim 7, wherein the fusing of the first feature and the second feature comprises:
weighting the second feature based on the second feature to obtain a weighted feature of the candidate information; and adding the weighted feature and the first feature to obtain the fused feature (Chen [0073]: In some embodiments, each of the plurality of candidate classifications may correspond to a weight in the focal loss function. At least one weight of the focal loss function each of which corresponds to one of the plurality of candidate classifications being different from weights of the focal loss function corresponding to the remainder of the plurality of candidate classifications. For example, a smaller weight may be assigned to a benign pulmonary nodule, and a larger weight may be assigned to a malignant pulmonary nodule. By adjusting the weights corresponding to the plurality of candidate classifications, negative effects induced by imbalance of samples corresponding to different candidate classifications on the classification model may be reduced, and a more robust classification model may be obtained.).
Regarding claim 11, the method according to claim 6, wherein:
the obtaining of the first feature comprises processing the coarse-grained feature through a first extraction branch in a probability estimation model; the obtaining of the second feature comprises processing the information feature and the intermediate feature through a second extraction branch in the probability estimation model (Liu [0040]: The present invention discloses the target cross-domain detection and understanding method. … The method of detecting and understanding a notable target according to a specific intention is efficient, objective and comprehensive, and can effectively enhance the environmental visual perception ability and active safety. Meanwhile, in conjunction with a coupling relationship between a coding position probability and image features, diagonal vertexes of a target candidate frame are efficiently located, network complexity is simplified, the difficulty and redundancy of fusion are avoided, the expenses of system calculation are saved, and the application needs for actual detection can be met.);
the obtaining of the fused feature comprises processing the first feature and the second feature through a fusion branch in the probability estimation model; and the obtaining of the estimated event probability comprises processing the fused feature through an estimation branch in the probability estimation model (Liu [0099]: “performing feature fusion on the intention feature and a target visual feature: ƒ.sub.fusion=ƒ(x)⊕ƒ.sub.int, where f(x) represents a visual feature extracted by a target through the NEGSS-Net backbone network; represents a channel splicing operator; f.sub.fusion represents the fused feature;”).
Regarding claim 12, the method according to claim 11, further comprising, before extracting the information feature of the candidate information:
extracting an information feature of sample information; processing the coarse-grained feature of the sample information through the first extraction branch to obtain a first feature of the sample information (Chen [0094]: For example, the combination of the backbone feature, the segmentation feature, and the density feature may be realized by converting the backbone feature, the segmentation feature, and the density feature into a one-dimensional backbone feature vector, a one-dimensional segmentation feature vector, and a one-dimensional density feature vector, respectively, and splicing the one-dimensional backbone feature vector, the one-dimensional segmentation feature vector, and the one-dimensional density feature vector to obtain a one-dimensional classification feature vector. The one-dimensional classification feature vector may be used to determine the plurality of candidate classifications and the probabilities corresponding to each candidate classification.);
processing the information feature of the sample information and the intermediate feature of the sample information through the second extraction branch to obtain a second feature of the sample information; processing the first feature of the sample information and the second feature of the sample information through the fusion branch to obtain a fused feature of the sample information; processing the fused feature of the candidate information through the estimation branch in the probability estimation model to obtain an estimated event probability of the sample information; (Chen [0073]: In some embodiments, each of the plurality of candidate classifications may correspond to a weight in the focal loss function. At least one weight of the focal loss function each of which corresponds to one of the plurality of candidate classifications being different from weights of the focal loss function corresponding to the remainder of the plurality of candidate classifications. For example, a smaller weight may be assigned to a benign pulmonary nodule, and a larger weight may be assigned to a malignant pulmonary nodule. By adjusting the weights corresponding to the plurality of candidate classifications, negative effects induced by imbalance of samples corresponding to different candidate classifications on the classification model may be reduced, and a more robust classification model may be obtained.);
obtaining a loss function value based on the estimated event probability of the sample information, an event probability label of the sample information, and a training weight of the sample information, the training weight being inversely related to a popularity of the sample information, the event probability label indicating a labeling probability of the specified event after the sample information is displayed; and updating a parameter of the probability estimation model based on the loss function value (Chen [0056]: The classification determination module 430 may determine a target classification of the structure of the subject. In some embodiments, the target classification of the structure may be determined based on probabilities of a plurality of candidate classifications. The plurality of candidate classifications of the structure and their respective probabilities may be determined by inputting the image data into a classification model. The classification model may trained according to a focal loss function, at least one weight of the focal loss function each of which corresponds to one of the plurality of candidate classifications being different from weights of the focal loss function corresponding to the remainder of the plurality of candidate classifications. In some embodiments, the target classification of the pulmonary nodule may be determined by identifying a candidate classification of the structure corresponding to a largest probability among the probabilities of the plurality of candidate classifications. The identified candidate classification may be designated as the target classification of the structure.).
Regarding claim 13, Liu teaches, an information pushing apparatus, comprising:
at least one memory configured to store program code; and at least one processor configured to read the program code and operate as instructed by the program code, the program code comprising (Liu [0037]: Computer equipment includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor implements the steps of the target cross-domain detection and understanding method based on attention estimation when executing the computer program.):
information feature extraction code, configured to cause the at least one processor to extract information feature of candidate information from an image, the information feature comprising a coarse-grained feature and a fine-grained feature, a number of tail value samples of the coarse- grained feature being greater than a number of tail value samples of the fine-grained feature (Liu [0040]: The present invention discloses the target cross-domain detection and understanding method. … The method of detecting and understanding a notable target according to a specific intention is efficient, objective and comprehensive, and can effectively enhance the environmental visual perception ability and active safety. Meanwhile, in conjunction with a coupling relationship between a coding position probability and image features, diagonal vertexes of a target candidate frame are efficiently located, network complexity is simplified, the difficulty and redundancy of fusion are avoided, the expenses of system calculation are saved, and the application needs for actual detection can be met.);
Liu does not clearly teach, first feature obtaining code, configured to cause the at least one processor to obtain, via inputting the coarse-grained feature into a first neural network, a first feature of the candidate information based on an intermediate feature, the intermediate feature being obtained in a process of extracting the coarse-grained feature; However, Chen [0013] teaches, “The method may include obtaining, by the computing device, image data representing a structure of a subject, determining, by the computing device, a plurality of candidate classifications of the structure and their respective probabilities by inputting the image data into a classification model, wherein the classification model includes a backbone network for determining a backbone feature of the structure, a segmentation network for determining a segmentation feature of the structure, and a density classification network for determining a density feature of the structure, and determining, by the computing device, a target classification of the structure based on the probabilities of the plurality of candidate classifications.”
second feature obtaining code, configured to cause the at least one processor to obtain, via inputting the coarse-grained feature and the fine-grained feature into a second neural network (Chen [0006]: In some embodiments, the determining a plurality of candidate classifications of the structure by inputting the image data into a classification model may include obtaining the backbone feature, the segmentation feature, and the density feature by inputting the image data into the backbone network, the segmentation network, and the density classification network, respectively, and determining a probability of each of the plurality of candidate classifications of the structure based on the backbone feature, the segmentation feature, and the density feature.), a second feature of the candidate information based on the information feature and the intermediate feature (Chen [0091]: In 640, a plurality of candidate classifications and their respective probabilities may be determined based on the backbone feature, the segmentation feature, and the density feature.);
information obtaining code, configured to cause the at least one processor to obtain, based on the first feature and the second feature, target information from a plurality of pieces of the candidate information based on the first feature and the second feature (Chen [0074]: In 530, a target classification of the structure may be determined based on at least a part of the probabilities of the plurality of candidate classifications. In some embodiments, the target classification of the structure may be determined by the classification determination module 430.); and
information pushing code, configured to cause the at least one processor to push the target information to one or more devices (Chen [0056]: The classification determination module 430 may determine a target classification of the structure of the subject. In some embodiments, the target classification of the structure may be determined based on probabilities of a plurality of candidate classifications. The plurality of candidate classifications of the structure and their respective probabilities may be determined by inputting the image data into a classification model. The classification model may trained according to a focal loss function, at least one weight of the focal loss function each of which corresponds to one of the plurality of candidate classifications being different from weights of the focal loss function corresponding to the remainder of the plurality of candidate classifications. In some embodiments, the target classification of the pulmonary nodule may be determined by identifying a candidate classification of the structure corresponding to a largest probability among the probabilities of the plurality of candidate classifications. The identified candidate classification may be designated as the target classification of the structure.).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to incorporate the teaching of Liu et al. to the Chen’s system by adding the feature of candidate information data. The references (Liu and Chen) teach features that are analogous art and they are directed to the same field of endeavor, such as databases. Ordinary skilled artisan would have been motivated to do so to provide Liu’s system with enhanced data. (See Chen [Abstract], [0227], [0240], [0276]). One of the biggest advantages of network machine learning database algorithms is their ability to improve over time. Machine learning technology typically improves efficiency and accuracy thanks to the ever-increasing amounts of data that are processed.
Regarding claim 14, the apparatus according to claim 13, wherein the first feature obtaining code causes the
at least one processor to obtain the first feature by (Liu [0037]: processor):
performing feature extraction on the coarse-grained feature to obtain at least one first intermediate feature of the candidate information, obtaining a first weight of each of the at least one first intermediate feature based on the coarse-grained feature, and obtaining the first feature of the candidate information based on the at least one first intermediate feature and the first weight of each of the at least one first intermediate feature (Liu [0040]: The present invention discloses the target cross-domain detection and understanding method. The regional weight can be partially reduced through spatial probability control and salient point pooling and by taking a spatial probability control layer as an input image channel, and an edge salient cross point pooling layer can help a network to better locate a target point; through cross-domain guiding semantic extraction and knowledge transfer, an inclusion relation between target depth visual features and guiding semantics for different domains is explored, network training is guided, and cross-domain invariant features are extracted to enhance the cross-domain perception of a model; and by analyzing a target notability degree, a semantic hierarchy cross-domain perception mapping effect and a back propagation mechanism are explored, and a problem of accuracy in notable target prediction and guiding semantic understanding under a specific intention is solved.); and
wherein the second feature obtaining code causes the at least one processor to obtain the second feature by: performing feature extraction on the information feature to obtain at least one second intermediate feature of the candidate information, obtaining second weights based on the information feature, the second weights comprising a second weight of each of the at least one second intermediate feature and a second weight of each of the at least one first intermediate feature (Chen [0032]: At least one weight of the focal loss function each of which corresponds to one of the plurality of candidate classifications being different from weights of the focal loss function corresponding to the remainder of the plurality of candidate classifications. By adjusting the weights corresponding to the plurality of candidate classifications, negative effects induced by imbalance of samples corresponding to different candidate classifications on the classification model may be reduced, and a more robust classification model may be obtained.), and
obtaining the second feature of the candidate information based on the second weight of each of the at least one second intermediate feature, the second weight of each of the at least one first intermediate feature, the at least one second intermediate feature, and the at least one first intermediate feature (Chen [0056]: The classification determination module 430 may determine a target classification of the structure of the subject. In some embodiments, the target classification of the structure may be determined based on probabilities of a plurality of candidate classifications. The plurality of candidate classifications of the structure and their respective probabilities may be determined by inputting the image data into a classification model. The classification model may trained according to a focal loss function, at least one weight of the focal loss function each of which corresponds to one of the plurality of candidate classifications being different from weights of the focal loss function corresponding to the remainder of the plurality of candidate classifications. In some embodiments, the target classification of the pulmonary nodule may be determined by identifying a candidate classification of the structure corresponding to a largest probability among the probabilities of the plurality of candidate classifications. The identified candidate classification may be designated as the target classification of the structure.).
Regarding claim 15, the apparatus according to claim 14, wherein the second feature obtaining code causes the at least one processor to obtain the second weights by:
splicing the information feature and a popularity vector to obtain a first spliced feature of the candidate information, the popularity vector indicating historical conversion times of the candidate information; and obtaining the second weights based on the first spliced feature (Chen [0094]: For example, the combination of the backbone feature, the segmentation feature, and the density feature may be realized by converting the backbone feature, the segmentation feature, and the density feature into a one-dimensional backbone feature vector, a one-dimensional segmentation feature vector, and a one-dimensional density feature vector, respectively, and splicing the one-dimensional backbone feature vector, the one-dimensional segmentation feature vector, and the one-dimensional density feature vector to obtain a one-dimensional classification feature vector. The one-dimensional classification feature vector may be used to determine the plurality of candidate classifications and the probabilities corresponding to each candidate classification.).
Regarding claim 16, the apparatus according to claim 13, wherein the information obtaining code causes the at least one processor to obtain the target information by:
fusing the first feature and the second feature to obtain a fused feature of the candidate information; obtaining an estimated event probability of the candidate information based on the fused feature, the estimated event probability marking an estimated probability of a specified event after corresponding information is displayed; and obtaining the target information based on the estimated event probability (Liu [Abstract]: Through spatial probability control and salient point pooling and in conjunction with a coupling relationship between a coding position probability and image features, diagonal vertexes of a target candidate frame are efficiently located, and network complexity is simplified so as to meet application needs for actual detection; through cross-domain guiding semantic extraction and knowledge transfer, an inclusion relation between target depth visual features and guiding semantics for different domains is explored, network training is guided, and cross-domain invariant features are extracted to enhance the cross-domain perception of a model; and by analyzing a target notability degree, a semantic hierarchy cross-domain perception mapping effect and a back propagation mechanism are explored, and a problem of accuracy in notable target prediction and guiding semantic understanding under a specific intention is solved.).
Regarding claim 17, the apparatus according to claim 16, wherein the information obtaining code causes the at least one processor to fuse the first feature and the second feature by:
obtaining a third weight of the second feature based on the information feature; and fusing the first feature and the second feature based on the third weight of the second feature to obtain the fused feature (Liu [0099]: “performing feature fusion on the intention feature and a target visual feature: ƒ.sub.fusion=ƒ(x)⊕ƒ.sub.int, where f(x) represents a visual feature extracted by a target through the NEGSS-Net backbone network; represents a channel splicing operator; f.sub.fusion represents the fused feature;”).
Regarding claim 18, the apparatus according to claim 17, wherein the information obtaining code causes the at least one processor to obtain the third weight by:
splicing the information feature and a popularity vector to obtain a second spliced feature of the candidate information, the popularity vector indicating historical conversion times of the candidate information; and obtaining the third weight of the second feature based on the second spliced feature (Chen [0094]: For example, the combination of the backbone feature, the segmentation feature, and the density feature may be realized by converting the backbone feature, the segmentation feature, and the density feature into a one-dimensional backbone feature vector, a one-dimensional segmentation feature vector, and a one-dimensional density feature vector, respectively, and splicing the one-dimensional backbone feature vector, the one-dimensional segmentation feature vector, and the one-dimensional density feature vector to obtain a one-dimensional classification feature vector. The one-dimensional classification feature vector may be used to determine the plurality of candidate classifications and the probabilities corresponding to each candidate classification.).
Regarding claim 19, the apparatus according to claim 17, wherein the information obtaining code causes the at least one processor to fuse the first feature and the second feature by:
weighting the second feature based on the second feature to obtain a weighted feature of the candidate information; and adding the weighted feature and the first feature to obtain the fused feature (Chen [0073]: In some embodiments, each of the plurality of candidate classifications may correspond to a weight in the focal loss function. At least one weight of the focal loss function each of which corresponds to one of the plurality of candidate classifications being different from weights of the focal loss function corresponding to the remainder of the plurality of candidate classifications. For example, a smaller weight may be assigned to a benign pulmonary nodule, and a larger weight may be assigned to a malignant pulmonary nodule. By adjusting the weights corresponding to the plurality of candidate classifications, negative effects induced by imbalance of samples corresponding to different candidate classifications on the classification model may be reduced, and a more robust classification model may be obtained.).
Regarding claim 20, Liu teaches, a non-transitory computer-readable storage medium, storing at least one computer instruction, wherein the at least one computer instruction is executable by at least one processor to cause the at least one processor to (Liu [0037]: Computer equipment includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor implements the steps of the target cross-domain detection and understanding method based on attention estimation when executing the computer program.):
extract an information feature of candidate information from an image, the information feature comprising a coarse-grained feature and a fine-grained feature, a number of tail value samples of the coarse- grained feature being greater than a number of tail value samples of the fine-grained feature (Liu [0040]: The present invention discloses the target cross-domain detection and understanding method. … The method of detecting and understanding a notable target according to a specific intention is efficient, objective and comprehensive, and can effectively enhance the environmental visual perception ability and active safety. Meanwhile, in conjunction with a coupling relationship between a coding position probability and image features, diagonal vertexes of a target candidate frame are efficiently located, network complexity is simplified, the difficulty and redundancy of fusion are avoided, the expenses of system calculation are saved, and the application needs for actual detection can be met.);
Liu does not clearly teach, obtain, via inputting the coarse-grained feature into a first neural network, a first feature of the candidate information based on an intermediate feature, the intermediate feature being obtained in a process of extracting the coarse-grained feature; However, Chen [0013] teaches, “The method may include obtaining, by the computing device, image data representing a structure of a subject, determining, by the computing device, a plurality of candidate classifications of the structure and their respective probabilities by inputting the image data into a classification model, wherein the classification model includes a backbone network for determining a backbone feature of the structure, a segmentation network for determining a segmentation feature of the structure, and a density classification network for determining a density feature of the structure, and determining, by the computing device, a target classification of the structure based on the probabilities of the plurality of candidate classifications.”
obtain, via inputting the coarse-grained feature and the fine-grained feature into a second neural network (Chen [0006]: In some embodiments, the determining a plurality of candidate classifications of the structure by inputting the image data into a classification model may include obtaining the backbone feature, the segmentation feature, and the density feature by inputting the image data into the backbone network, the segmentation network, and the density classification network, respectively, and determining a probability of each of the plurality of candidate classifications of the structure based on the backbone feature, the segmentation feature, and the density feature.), a second feature of the candidate information based on the information feature and the intermediate feature (Chen [0091]: In 640, a plurality of candidate classifications and their respective probabilities may be determined based on the backbone feature, the segmentation feature, and the density feature.);
obtain, based on the first feature and the second feature, target information from a plurality of pieces of candidate information (Chen [0074]: In 530, a target classification of the structure may be determined based on at least a part of the probabilities of the plurality of candidate classifications. In some embodiments, the target classification of the structure may be determined by the classification determination module 430.); and
push the target information to one or more devices (Chen [0056]: The classification determination module 430 may determine a target classification of the structure of the subject. In some embodiments, the target classification of the structure may be determined based on probabilities of a plurality of candidate classifications. The plurality of candidate classifications of the structure and their respective probabilities may be determined by inputting the image data into a classification model. The classification model may trained according to a focal loss function, at least one weight of the focal loss function each of which corresponds to one of the plurality of candidate classifications being different from weights of the focal loss function corresponding to the remainder of the plurality of candidate classifications. In some embodiments, the target classification of the pulmonary nodule may be determined by identifying a candidate classification of the structure corresponding to a largest probability among the probabilities of the plurality of candidate classifications. The identified candidate classification may be designated as the target classification of the structure.).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to incorporate the teaching of Liu et al. to the Chen’s system by adding the feature of candidate information data. The references (Liu and Chen) teach features that are analogous art and they are directed to the same field of endeavor, such as databases. Ordinary skilled artisan would have been motivated to do so to provide Liu’s system with enhanced data. (See Chen [Abstract], [0227], [0240], [0276]). One of the biggest advantages of network machine learning database algorithms is their ability to improve over time. Machine learning technology typically improves efficiency and accuracy thanks to the ever-increasing amounts of data that are processed.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure.
Tu, US 2021/0019479, Text Translation method and apparatus, storage medium, and computer device
Shin, US 2021/0365470, Apparatus for recommending feature and method for recommending feature using the same
Wang, US 2021/0216561, Information search method and apparatus, device and storage medium
Zhang, US 2019/0300870, Target recognition motifs and uses thereof
Inaba, US 2018/0247705, Information Processing Apparatus, method and program
Haws, US 2014/0207800, Hill-climbing feature selection with max-relevancy and minimum redundancy criteria
THIS ACTION IS MADE FINAL. 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 mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SABA AHMED whose telephone number is (571) 270-0236. The examiner can normally be reached on MON – FRI: 8AM – 5PM EST.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Hosain Alam can be reached on (571) 272-3978. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/SABA AHMED/
Examiner, Art Unit 2154
/BORIS GORNEY/Supervisory Patent Examiner, Art Unit 2154
Prosecution Insights