DETAILED ACTION
Notice of Pre-AIA or AIA Status
1. The present application, filed on or after March 16, 2013, is being examined
under the first inventor to file provisions of the AIA .
Claim Objections
2. Claim 8 is objected to because of the following informalities: Claim 8 recites the words f wherein which is a typographical error. Appropriate correction is required.
Claim Rejections - 35 USC § 103
3. 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.
4. Claims 1, 4, 6, 11, 13-14, and 16-17 are rejected under 35 U.S.C. 103 as being unpatentable over Qiu et al. (“Void Filling of Digital Elevation Models with a Terrain Texture Learning Model Based on Generative Adversarial Networks”, published 28 November 2019) in view of Williams et al. (US Patent Application Publication No. 2020/0387739 A1) and further in view of Nemoto et al. (US Patent Application Publication No. 2020/0388043 A1).
5. Regarding Claim 1, Qiu discloses A computer-implemented method for using a trained generative adversarial network (Abstract reciting “… In this paper, we proposed a method based on a deep convolutional generative adversarial network (DCGAN) to address the problem of DEM void filling. …” DCGAN is a trained generative adversaria network (GAN).) to improve construction site evaluation, comprising: obtaining, by one or more processors, image data associated with a terrain of a site; (see FIG. 5; see Methodology on page 3 reciting “In this work, the voids in DEM data were filled using the proposed terrain texture generation model (TTGM) using the framework of a deep convolutional generative adversarial network. …” DEM corresponds to obtained image data of a terrain of any site on surface of Earth.)
determining, by the one or more processors, one or more gaps in elevation information of the terrain within the image data; (see 3.2.2. on page 11 reciting “
… First, data voids were detected, and the number of void pixels for each segmented void was determined. …”;
See Conclusions on page 19 reciting “In this paper, we present a void filling method that uses a TTGM for the reconstruction of areas in which data are missing in DEM. The TTGM was constructed to capture deep geospatial features and generate good-quality DEM patches. In addition to the elevation information, the slope and RDLS were incorporated as inputs to be fed into the neural network during the training process. …” The voids reconstructed in the DEM comprises voids of elevation data.)
generating, by the one or more processors, a gap-filled representation of the terrain of the site by probabilistically filling the one or more gaps using the trained generative adversarial network; (see Figure 4; see 2.2.1 on page 6 reciting “… A void filling result with more detailed textures can be obtained by paying more attention to the data close to the void region. If the weight of a pixel in an uncorrupted region is positively correlated with the number of void pixels around it, then the farther from the void region, the smaller the pixel weight. A 9 × 9 window is used to traverse the DEM data. …”;
see Methodology on page 3 reciting “In this work, the voids in DEM data were filled using the proposed terrain texture generation model (TTGM) using the framework of a deep convolutional generative adversarial network. …” Therefore a generative adversarial network (GAN) is used to fill the voids in DEM data.) and determining, by the one or more processors, one or more attributes of the site based upon the gap-filled representation of the terrain. (see 2.1.2. on pages 3-4 reciting “Our primary goal was to obtain a model that can simulate the natural terrain textures of DEM data and retain the intrinsic three-dimensional topographic information. Therefore, we propose a terrain texture generation model (TTGM) based on the DCGAN to address the problem of terrain texture generation. The TTGM was trained with the WGAN adversarial loss because the WGAN value function improves the stability of training and almost solves the mode collapse problem. …” The texture generated to fill in the void of the DEM corresponds to the one or more attributes of the site.)
While not explicitly disclosed by Qiu, Williams discloses by the one or more processors (paragraph [0154] reciting “The apparatus may then comprise a rendering processor (for example CPU 1100 operating under suitable software instruction) operable to render an image responsive to the point cloud output by the generative network of the trained GAN.”)
by probabilistically (paragraph [0101] reciting “Turning now to FIG. 12, a generative adversarial network comprises two neural networks; a generative network 1210, which can model the probability of features, and a discriminative network 1220 that can act as a classifier of features. The generative network thus learns to output data with target features, whilst the discriminative network learns to distinguish candidates produced by the generative network from true target data, based on such features.” So the GAN has 2 components one of which is a generative network which models the probability of features. Therefore, probability is used in the GAN to generate the void filling of DEMs in Qiu.)
It would have been obvious to a person or ordinary skills in the art before the effective filing date of the claimed invention to modify Qiu with Williams so that the method in Qiu can be processed on a computer with a processor. Furthermore, Qiu discloses a generator G and a discriminator D similar to Williams disclose. Therefore, the GAN in Qiu also models probability of features which is then used in the overall GAN to fill in the voids of the DEMs. Therefore, Qiu can be modified with Williams to disclose a GAN having two components, the generative component and the discriminative component since both references disclose the two components of the GAN.
While the combination of Qiu and Williams does not explicitly disclose, Nemoto discloses to improve construction site evaluation, (paragraph [0022] reciting “The construction management device 200 receives an image pair in which the construction site is captured from the hydraulic excavator 100, and manages the image pair in association with an ID of the hydraulic excavator 100 and an imaging time. The construction management device 200 performs stereo measurement on the image pair, so as to generate point group data representing a landform of the construction site. The construction management device 200 calculates an amount of earthwork (an amount of work done) in the construction site by using the point group data. The point group data is an example of three-dimensional data. In other embodiments, as three-dimensional data, triangulated irregular network (TIN) data, digital elevation model (DEM) data, polygon data, or voxel data may be generated.” Therefore, an improved DEM without voids/gaps/holes is conducive to construction site evaluation for earthwork.)
It would have been obvious to modify to a person of ordinary skills in the art before the effective filing date of the claimed invention to modify Qiu and Williams with Nemoto so that the DEM without voids can be used for construction site evaluation. This is a beneficial modification since DEM sites are often evaluated for purposes one of which may be construction site evaluation.
6. Regarding Claim 4, Qiu further discloses The computer-implemented method of claim 1, wherein the one or more gaps comprise one or more regions in the image data associated with imaging artifacts. (see Introduction on pages 2-3 reciting “To this end, this study aimed to establish a learning mechanism of the terrain textural features to fill voids in DEM. To obtain a void filling result that has similar spatial structural patterns to those of the real DEM data, the terrain texture generation model (TTGM) is constructed considering the homogeneity and heterogeneity of terrain textures. Trained by a large number of DEM patches with complex terrain textures in mountainous areas, the TTGM was able to capture deep features of the DEM data and generate data with terrain structural patterns. The terrain texture details of the generated data were enhanced by training the model involving topographic information, such as the slope and relief degree of land surface (RDLS), together with DEM data. Additionally, a loss function composed of pixel-wise, contextual, and perceptual loss was designed to constrain data generation during the reconstruction process. The proposed method was then employed to generate a satisfactory fill surface to fill in the void, followed by a postprocessing procedure to remove boundary artifacts. In addition, two models trained on different input sizes (64 × 64 and 128 × 128 pixels) were combined to efficiently fill the voids of arbitrary shapes.” Boundary artifacts correspond to imaging artifacts as they are part of the imaged DEM.)
7. Regarding Claim 6, Nemoto further discloses The computer-implemented method of claim 1, wherein the one or more attributes of the site comprise a status of construction of a building at the site. (paragraph [0078] reciting “… In a case of the construction type, an operator of the hydraulic excavator 100 periodically (for example, every day) images a current status landform of the construction site with the stereo camera 122, and thus an image pair representing the latest current status landform is stored in the image storage unit 233 every predetermined period. …” DEM imaging includes data on current status landform of the construction site which corresponds to a status of construction of a building at the site. It is obvious Nemoto can have DEM data including status of construction of a terrain or landscape which can be further based on gap filled DEM.)
8. Regarding Claim 11, Qiu discloses A computing system for using a trained generative adversarial network (Abstract reciting “… In this paper, we proposed a method based on a deep convolutional generative adversarial network (DCGAN) to address the problem of DEM void filling. …” DCGAN is a trained generative adversaria network (GAN).) to improve construction site evaluation, comprising: obtain image data associated with a terrain of a site; (see FIG. 5; see Methodology on page 3 reciting “In this work, the voids in DEM data were filled using the proposed terrain texture generation model (TTGM) using the framework of a deep convolutional generative adversarial network. …” DEM corresponds to obtained image data of a terrain of any site on surface of Earth.)
determine one or more gaps in elevation information of the terrain within the image data; (see 3.2.2. on page 11 reciting “
… First, data voids were detected, and the number of void pixels for each segmented void was determined. …”;
See Conclusions on page 19 reciting “In this paper, we present a void filling method that uses a TTGM for the reconstruction of areas in which data are missing in DEM. The TTGM was constructed to capture deep geospatial features and generate good-quality DEM patches. In addition to the elevation information, the slope and RDLS were incorporated as inputs to be fed into the neural network during the training process. …” The voids reconstructed in the DEM comprises voids of elevation data.)
generate a gap-filled representation of the terrain of the site by probabilistically filling the one or more gaps using the trained generative adversarial network; (see Figure 4; see 2.2.1 on page 6 reciting “… A void filling result with more detailed textures can be obtained by paying more attention to the data close to the void region. If the weight of a pixel in an uncorrupted region is positively correlated with the number of void pixels around it, then the farther from the void region, the smaller the pixel weight. A 9 × 9 window is used to traverse the DEM data. …”;
see Methodology on page 3 reciting “In this work, the voids in DEM data were filled using the proposed terrain texture generation model (TTGM) using the framework of a deep convolutional generative adversarial network. …” Therefore a generative adversarial network (GAN) is used to fill the voids in DEM data.) and determine one or more attributes of the site based upon the gap-filled representation of the terrain. (see 2.1.2. on pages 3-4 reciting “Our primary goal was to obtain a model that can simulate the natural terrain textures of DEM data and retain the intrinsic three-dimensional topographic information. Therefore, we propose a terrain texture generation model (TTGM) based on the DCGAN to address the problem of terrain texture generation. The TTGM was trained with the WGAN adversarial loss because the WGAN value function improves the stability of training and almost solves the mode collapse problem. …” The texture generated to fill in the void of the DEM corresponds to the one or more attributes of the site.)
While not explicitly disclosed by Qiu, Williams discloses system (paragraph [0001] reciting “This disclosure relates to a digital model repair system and method system and method.”)
one or more processors, and one or more memories having stored thereon computer-executable instructions that, when executed, cause the computing system to: (paragraph [0091] reciting “… The CPU 1100 can perform any of the above methods under the control of program instructions stored in the RAM 1110 and/or the NTMRM 1120. The NTMRM 1120 therefore provides an example of a non-transitory machine-readable medium which stores computer software by which the CPU 1100
performs the method or methods discussed above.”)
by probabilistically (paragraph [0101] reciting “Turning now to FIG. 12, a generative adversarial network comprises two neural networks; a generative network 1210, which can model the probability of features, and a discriminative network 1220 that can act as a classifier of features. The generative network thus learns to output data with target features, whilst the discriminative network learns to distinguish candidates produced by the generative network from true target data, based on such features.” So the GAN has 2 components one of which is a generative network which models the probability of features. Therefore, probability is used in the GAN to generate the void filling of DEMs in Qiu.)
It would have been obvious to a person or ordinary skills in the art before the effective filing date of the claimed invention to modify Qiu with Williams so that the method in Qiu can be processed on a computer with a processor. Furthermore, Qiu discloses a generator G and a discriminator D similar to Williams disclose. Therefore, the GAN in Qiu also models probability of features which is then used in the overall GAN to fill in the voids of the DEMs. Therefore, Qiu can be modified with Williams to disclose a GAN having two components, the generative component and the discriminative component since both references disclose the two components of the GAN.
While the combination of Qiu and Williams does not explicitly disclose, Nemoto discloses to improve construction site evaluation, (paragraph [0022] reciting “The construction management device 200 receives an image pair in which the construction site is captured from the hydraulic excavator 100, and manages the image pair in association with an ID of the hydraulic excavator 100 and an imaging time. The construction management device 200 performs stereo measurement on the image pair, so as to generate point group data representing a landform of the construction site. The construction management device 200 calculates an amount of earthwork (an amount of work done) in the construction site by using the point group data. The point group data is an example of three-dimensional data. In other embodiments, as three-dimensional data, triangulated irregular network (TIN) data, digital elevation model (DEM) data, polygon data, or voxel data may be generated.” Therefore, an improved DEM without voids/gaps/holes is conducive to construction site evaluation for earthwork.)
It would have been obvious to modify to a person of ordinary skills in the art before the effective filing date of the claimed invention to modify Qiu and Williams with Nemoto so that the DEM without voids can be used for construction site evaluation. This is a beneficial modification since DEM sites are often evaluated for purposes one of which may be construction site evaluation.
9. Regarding Claim 13, Qiu further discloses The computing system of claim 11, wherein the one or more gaps comprise one or more regions in the image data associated with imaging artifacts. (see Introduction on pages 2-3 reciting “To this end, this study aimed to establish a learning mechanism of the terrain textural features to fill voids in DEM. To obtain a void filling result that has similar spatial structural patterns to those of the real DEM data, the terrain texture generation model (TTGM) is constructed considering the homogeneity and heterogeneity of terrain textures. Trained by a large number of DEM patches with complex terrain textures in mountainous areas, the TTGM was able to capture deep features of the DEM data and generate data with terrain structural patterns. The terrain texture details of the generated data were enhanced by training the model involving topographic information, such as the slope and relief degree of land surface (RDLS), together with DEM data. Additionally, a loss function composed of pixel-wise, contextual, and perceptual loss was designed to constrain data generation during the reconstruction process. The proposed method was then employed to generate a satisfactory fill surface to fill in the void, followed by a postprocessing procedure to remove boundary artifacts. In addition, two models trained on different input sizes (64 × 64 and 128 × 128 pixels) were combined to efficiently fill the voids of arbitrary shapes.” Boundary artifacts correspond to imaging artifacts as they are part of the imaged DEM.)10. Regarding Claim 14, Qiu discloses for using a trained generative adversarial network (Abstract reciting “… In this paper, we proposed a method based on a deep convolutional generative adversarial network (DCGAN) to address the problem of DEM void filling. …” DCGAN is a trained generative adversaria network (GAN).)
obtain image data associated with a terrain of a site; (see FIG. 5; see Methodology on page 3 reciting “In this work, the voids in DEM data were filled using the proposed terrain texture generation model (TTGM) using the framework of a deep convolutional generative adversarial network. …” DEM corresponds to obtained image data of a terrain of any site on surface of Earth.)
determine one or more gaps in elevation information of the terrain within the image data; (see 3.2.2. on page 11 reciting “
… First, data voids were detected, and the number of void pixels for each segmented void was determined. …”;
See Conclusions on page 19 reciting “In this paper, we present a void filling method that uses a TTGM for the reconstruction of areas in which data are missing in DEM. The TTGM was constructed to capture deep geospatial features and generate good-quality DEM patches. In addition to the elevation information, the slope and RDLS were incorporated as inputs to be fed into the neural network during the training process. …” The voids reconstructed in the DEM comprises voids of elevation data.)
generate a gap-filled representation of the terrain of the site by probabilistically filling the one or more gaps using the trained generative adversarial network; (see Figure 4; see 2.2.1 on page 6 reciting “… A void filling result with more detailed textures can be obtained by paying more attention to the data close to the void region. If the weight of a pixel in an uncorrupted region is positively correlated with the number of void pixels around it, then the farther from the void region, the smaller the pixel weight. A 9 × 9 window is used to traverse the DEM data. …”;
see Methodology on page 3 reciting “In this work, the voids in DEM data were filled using the proposed terrain texture generation model (TTGM) using the framework of a deep convolutional generative adversarial network. …” Therefore a generative adversarial network (GAN) is used to fill the voids in DEM data.) and determine one or more attributes of the site based upon the gap-filled representation of the terrain. (see 2.1.2. on pages 3-4 reciting “Our primary goal was to obtain a model that can simulate the natural terrain textures of DEM data and retain the intrinsic three-dimensional topographic information. Therefore, we propose a terrain texture generation model (TTGM) based on the DCGAN to address the problem of terrain texture generation. The TTGM was trained with the WGAN adversarial loss because the WGAN value function improves the stability of training and almost solves the mode collapse problem. …” The texture generated to fill in the void of the DEM corresponds to the one or more attributes of the site.)
While not explicitly disclosed by Qiu, Williams discloses A non-transitory computer-readable medium having stored thereon computer-executable instructions that, when executed by one or more processors of a computing system, cause the computing system to: (paragraph [0091] reciting “… The CPU 1100 can perform any of the above methods under the control of program instructions stored in the RAM 1110 and/or the NTMRM 1120. The NTMRM 1120 therefore provides an example of a non-transitory machine-readable medium which stores computer software by which the CPU 1100
performs the method or methods discussed above.”)
by probabilistically (paragraph [0101] reciting “Turning now to FIG. 12, a generative adversarial network comprises two neural networks; a generative network 1210, which can model the probability of features, and a discriminative network 1220 that can act as a classifier of features. The generative network thus learns to output data with target features, whilst the discriminative network learns to distinguish candidates produced by the generative network from true target data, based on such features.” So the GAN has 2 components one of which is a generative network which models the probability of features. Therefore, probability is used in the GAN to generate the void filling of DEMs in Qiu.)
It would have been obvious to a person or ordinary skills in the art before the effective filing date of the claimed invention to modify Qiu with Williams so that the method in Qiu can be processed on a computer with a processor. Furthermore, Qiu discloses a generator G and a discriminator D similar to Williams disclose. Therefore, the GAN in Qiu also models probability of features which is then used in the overall GAN to fill in the voids of the DEMs. Therefore, Qiu can be modified with Williams to disclose a GAN having two components, the generative component and the discriminative component since both references disclose the two components of the GAN. While the combination of Qiu and Williams does not explicitly disclose, Nemoto discloses to improve construction site evaluation, (paragraph [0022] reciting “The construction management device 200 receives an image pair in which the construction site is captured from the hydraulic excavator 100, and manages the image pair in association with an ID of the hydraulic excavator 100 and an imaging time. The construction management device 200 performs stereo measurement on the image pair, so as to generate point group data representing a landform of the construction site. The construction management device 200 calculates an amount of earthwork (an amount of work done) in the construction site by using the point group data. The point group data is an example of three-dimensional data. In other embodiments, as three-dimensional data, triangulated irregular network (TIN) data, digital elevation model (DEM) data, polygon data, or voxel data may be generated.” Therefore, an improved DEM without voids/gaps/holes is conducive to construction site evaluation for earthwork.)
It would have been obvious to modify to a person of ordinary skills in the art before the effective filing date of the claimed invention to modify Qiu and Williams with Nemoto so that the DEM without voids can be used for construction site evaluation. This is a beneficial modification since DEM sites are often evaluated for purposes one of which may be construction site evaluation.
11. Regarding Claim 16, Qiu further discloses The non-transitory computer-readable medium of claim 14, wherein the one or more gaps comprise one or more regions in the image data associated with imaging artifacts. (see Introduction on pages 2-3 reciting “To this end, this study aimed to establish a learning mechanism of the terrain textural features to fill voids in DEM. To obtain a void filling result that has similar spatial structural patterns to those of the real DEM data, the terrain texture generation model (TTGM) is constructed considering the homogeneity and heterogeneity of terrain textures. Trained by a large number of DEM patches with complex terrain textures in mountainous areas, the TTGM was able to capture deep features of the DEM data and generate data with terrain structural patterns. The terrain texture details of the generated data were enhanced by training the model involving topographic information, such as the slope and relief degree of land surface (RDLS), together with DEM data. Additionally, a loss function composed of pixel-wise, contextual, and perceptual loss was designed to constrain data generation during the reconstruction process. The proposed method was then employed to generate a satisfactory fill surface to fill in the void, followed by a postprocessing procedure to remove boundary artifacts. In addition, two models trained on different input sizes (64 × 64 and 128 × 128 pixels) were combined to efficiently fill the voids of arbitrary shapes.” Boundary artifacts correspond to imaging artifacts as they are part of the imaged DEM.)12. Regarding Claim 17, Nemoto further discloses The non-transitory computer-readable medium of claim 14, wherein the one or more attributes of the site comprise a status of construction of a building at the site. (paragraph [0078] reciting “… In a case of the construction type, an operator of the hydraulic excavator 100 periodically (for example, every day) images a current status landform of the construction site with the stereo camera 122, and thus an image pair representing the latest current status landform is stored in the image storage unit 233 every predetermined period. …” DEM imaging includes data on current status landform of the construction site which corresponds to a status of construction of a building at the site. It is obvious Nemoto can have DEM data including status of construction of a terrain or landscape which can be further based on gap filled DEM.)
13. Claims 2-3, 12 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Qiu in view of Williams in view of Nemoto and further in view of Vernon R. Goodman (US Patent Application Publication No. 2017/0076456 A1).
14. Regarding Claim 2, while the combination of Qiu, Williams, and Nemoto does not explicitly disclose, Goodman discloses and The computer-implemented method of claim 1, wherein the one or more gaps comprise one or more regions in the image data associated with one or more objects obscuring corresponding portions of the terrain. (paragraph [0088] reciting “In Example 11 a method for filtering digital elevation map (DEM) data includes ingesting digital elevation map (DEM) data and intensity data from a panchromatic (PAN) or multi-spectral (MS) image, filling in voids in the ingested DEM data using local interpolation to create interpolated DEM data, creating a shadow map based on the received intensity data, modifying, using the created shadow map, a height of one or more pixels in the interpolated DEM data to create modified DEM data, and providing signals to a display that cause an image to be displayed based on the modified DEM data.” The shadows in the DEM corresponds to additional voids that must be corrected/interpolated as well.)
It would have been obvious to a person of ordinary skills in the art before the effective filing date of the claimed invention to modify Qiu, Williams, and Nemoto with Goodman so that the shadows are also interpolated using the GAN of Qiu modified by Williams. This is an obviously beneficial modification because shadows often obscure interpolation values and Goodman offers a solution to interpolate the points of the DEM obscured by the shadow.
15. Regarding Claim 3, Goodman further discloses The computer-implemented method of claim 2, wherein the one or more objects include a portion of at least one of the following: a tree, a structure, a vehicle, or a person. (see FIG. 8 and 9 wherein the shadow is casted by a structure (building).)
16. Regarding Claim 12, while the combination of Qiu, Williams, and Nemoto does not explicitly disclose, Goodman discloses The computing system of claim 11, wherein the one or more gaps comprise one or more regions in the image data associated with one or more objects obscuring corresponding portions of the terrain. (paragraph [0088] reciting “In Example 11 a method for filtering digital elevation map (DEM) data includes ingesting digital elevation map (DEM) data and intensity data from a panchromatic (PAN) or multi-spectral (MS) image, filling in voids in the ingested DEM data using local interpolation to create interpolated DEM data, creating a shadow map based on the received intensity data, modifying, using the created shadow map, a height of one or more pixels in the interpolated DEM data to create modified DEM data, and providing signals to a display that cause an image to be displayed based on the modified DEM data.” The shadows in the DEM corresponds to additional voids that must be corrected/interpolated as well.)
It would have been obvious to a person of ordinary skills in the art before the effective filing date of the claimed invention to modify Qiu, Williams, and Nemoto with Goodman so that the shadows are also interpolated using the GAN of Qiu modified by Williams. This is an obviously beneficial modification because shadows often obscure interpolation values and Goodman offers a solution to interpolate the points of the DEM obscured by the shadow.
17. Regarding Claim 15, while the combination of Qiu, Williams, and Nemoto does not explicitly disclose, Goodman discloses The non-transitory computer-readable medium of claim 14, wherein the one or more gaps comprise one or more regions in the image data associated with one or more objects obscuring corresponding portions of the terrain. (paragraph [0088] reciting “In Example 11 a method for filtering digital elevation map (DEM) data includes ingesting digital elevation map (DEM) data and intensity data from a panchromatic (PAN) or multi-spectral (MS) image, filling in voids in the ingested DEM data using local interpolation to create interpolated DEM data, creating a shadow map based on the received intensity data, modifying, using the created shadow map, a height of one or more pixels in the interpolated DEM data to create modified DEM data, and providing signals to a display that cause an image to be displayed based on the modified DEM data.” The shadows in the DEM corresponds to additional voids that must be corrected/interpolated as well.)
It would have been obvious to a person of ordinary skills in the art before the effective filing date of the claimed invention to modify Qiu, Williams, and Nemoto with Goodman so that the shadows are also interpolated using the GAN of Qiu modified by Williams. This is an obviously beneficial modification because shadows often obscure interpolation values and Goodman offers a solution to interpolate the points of the DEM obscured by the shadow.
18. Claims 9 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Qiu in view of Williams in view of Nemoto and further in view of Yun Zhang (US Patent Application Publication No. 2016/0173850 A1).
19. Regarding Claim 9, while the combination of Qiu, Williams, and Nemoto does not explicitly disclose, Zhang discloses The computer-implemented method of claim 1, wherein the image data comprises a three-dimensional point cloud. (paragraph [0016] reciting “A three-dimensional (“3D”) model as used herein may be a digital 3D model with continuously distributed 3D points, such as a digital elevation model (“DEM”) (sometimes also referred to in the art as Digital Terrain Model (“DTM”)), digital surface model (DSM) or one or more discrete 3D points (also referred to as a 3D point cloud).“)
It would have been obvious to a person of ordinary skills in the art before the effective filing date of the claimed invention to modify Qiu, Williams, and Nemoto with Zhang so that the DEM collected is a set of continuously distributed 3D points which correspond to a 3D point cloud. This allows the DEM in Qiu to be acquired in a familiar manner.
20. Regarding Claim 19, while the combination of Qiu, Williams, and Nemoto does not explicitly disclose, Zhang discloses The non-transitory computer-readable medium of claim 14, wherein the image data comprises a three-dimensional point cloud. (paragraph [0016] reciting “A three-dimensional (“3D”) model as used herein may be a digital 3D model with continuously distributed 3D points, such as a digital elevation model (“DEM”) (sometimes also referred to in the art as Digital Terrain Model (“DTM”)), digital surface model (DSM) or one or more discrete 3D points (also referred to as a 3D point cloud).“)
It would have been obvious to a person of ordinary skills in the art before the effective filing date of the claimed invention to modify Qiu, Williams, and Nemoto with Zhang so that the DEM collected is a set of continuously distributed 3D points which correspond to a 3D point cloud. This allows the DEM in Qiu to be acquired in a familiar manner.
21. Claims 10 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Qiu in view of Williams in view of Nemoto and further in view of Zhou et al. (US Patent Application Publication No. 2017/0039765 A1).
22. Regarding Claim 10, while the combination of Qiu, Williams, and Nemoto does not explicitly disclose, Zhou discloses The computer-implemented method of claim 1, wherein the image data comprises a plurality of two-dimensional images. (paragraph [0153] reciting “… The world (earth) coordinates of the video frame can be determined at Multi-View Triangulation library 1145. 3D points are computed from the DEM data store 1115 and the 2D images from the video frame are projected onto the DEM data at 2D projections 1160. …” The DEM of Qiu can be modified to include 2D image projections.)
It would have been obvious to a person of ordinary skills in the art before the effective filing date of the claimed invention to modify Qiu, Williams, and Nemoto with Zhou so that the DEM incorporates a projection of 2D images. This is a beneficial modification as the 2D image projections taught in Zhou facilitates the generation of a more accurate surface texture for the DEM disclosed in Qiu.
23. Regarding Claim 20, while the combination of Qiu, Williams, and Nemoto does not explicitly disclose, Zhou discloses The non-transitory computer-readable medium of claim 14, wherein the image data comprises a plurality of two-dimensional images. (paragraph [0153] reciting “… The world (earth) coordinates of the video frame can be determined at Multi-View Triangulation library 1145. 3D points are computed from the DEM data store 1115 and the 2D images from the video frame are projected onto the DEM data at 2D projections 1160. …” The DEM of Qiu can be modified to include 2D image projections.)
It would have been obvious to a person of ordinary skills in the art before the effective filing date of the claimed invention to modify Qiu, Williams, and Nemoto with Zhou so that the DEM incorporates a projection of 2D images. This is a beneficial modification as the 2D image projections taught in Zhou facilitates the generation of a more accurate surface texture for the DEM disclosed in Qiu.
Allowable Subject Matter
24. Claims 5, 7, 8, 18 objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
25. The following is a statement of reasons for the indication of allowable subject matter: Claim 5 recites the limitation wherein the one or more attributes of the site comprise a volumetric soil measurement of at least a part of the site which is neither disclosed nor suggested by the cited references, either singly or in combination. 26. While Qiu discloses void filling of DEM using GAN, Qiu is silent with regards to volumetric soil measurement for the site. All the other cited references in this office action are also silent with respect to volumetric soil measurement as an attribute of the filled DEM.
27. Claim 7 recites the limitation wherein the one or more attributes of the site comprise water drainage associated with at least a part of the site which is neither disclosed nor suggested by the cited references, either singly or in combination.
28. Similarly, claim 18 recites the limitation wherein the one or more attributes of the site comprise water drainage associated with at least a part of the site which is neither disclosed nor suggested by the cited references, either singly or in combination.
29. While Qiu discloses void filling the DEM, Qiu is silent with regards to attributes being water drainage associated with a site of the DEM. All the other cited references are also silent with regards to attributes being water drainage associated with a site of the DEM. 30. Claim 8 recites the limitation wherein the one or more attributes of the site comprise one or more locations for utility infrastructure elements at the site which is neither disclosed nor suggested by the cited references, either singly or in combination. 31. While Qiu discloses void filling of DEM using GAN, Qiu is silent with regards to locations for utility infrastructure elements at the site. All the other cited references in this office action are also silent with respect to locations for utility infrastructure elements at the site.
CONTACT
Any inquiry concerning this communication or earlier communications from the examiner should be directed to FRANK S CHEN whose telephone number is (571)270-7993. The examiner can normally be reached Mon - Fri 8-11:30 and 1:30-6.
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, Kee Tung can be reached at 5712727794. 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.
/FRANK S CHEN/Primary Examiner, Art Unit 2611
Prosecution Insights