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 .
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claims 1, 5, 8, 12, 15, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Liu et al. (CN 201310753844 A) in view of Gong et al. (US 20230121890 A1).
Regarding claim 1, Liu et al. teach the computer-implemented method comprising: sampling, by at least one computer processor, a contiguous-pixel portion of a frame of a digital video signal, thereby providing a sampled patch, the sampled patch being smaller than the full resolution of the frame (see Abstract; “wherein, frame video image is divided into N number of noise reduction region”, see also page 2, 5th para; “all pixels of a video image are equally divided into N noise reduction regions. One special case is that one pixel acts as a noise reduction region”, Note; noise reduction regions implies sampled patch and each divided noise reduction region is a portion of the video image smaller than the full frame); computing an estimate of noise in the digital video signal based on a comparison of the denoised patch and the sampled patch (see page 2, 3rd para; “determining a noise estimation mean value according to the difference value of each noise reduction area; and respectively comparing the difference value of each noise reduction area with the noise estimation mean value to determine the noise level of each noise reduction area” ); denoising the digital video signal by applying an amount of digital noise reduction (DNR) to the digital video signal that is based on the computed noise estimate (see page see page 2, 6th para; “And the noise level is a video noise estimation result required for carrying out noise reduction processing on a noise reduction area of the next frame of video image… the noise level of one noise reduction region determined based on the current frame video image is used for noise reduction processing of the noise reduction region in the next frame video image”). However, Liu et al. does not teach a computer-implemented method for automatically adjusting high-definition video noise reduction; denoising the sampled patch using artificial-intelligence-based denoising, thereby providing a denoised patch; and displaying the denoised video signal on a digital video display, wherein the denoising the digital video signal is performed in real time as the video signal is displayed.
In the same field of endeavor, Gong et al. teaches a computer-implemented method for automatically adjusting high-definition video noise reduction (see Abstract; “”A computer-implemented method is provided for improving live video quality, see also para [0074]; “the first stage denoising may be dynamically adjusted based on use applications ….an algorithm for motion estimation such as DeepFlow, Farneback algorithm or LiteFlowNet may be applied to the live video to estimate motion at a point in time and/or location of patches (e.g., x, y coordinates) within an image frame, then the degree of spatial denoising may be automatically adjusted (e.g., perform or skip the spatial denoising”), denoising the sampled patch using artificial-intelligence-based denoising (see para [0006]; “The provided methods and systems may improve live video/imaging quality by employing deep learning techniques so as to reduce noise”, see also para [0025]; “the provided systems and methods may employ a deep learning framework that can perform real-time video quality enhancement or video denoising during live video acquisition”), thereby providing a denoised patch (see para [0042]; “one family of existing solutions for video denoising are patch-based algorithms”, see also para [0074]; “spatial denoising may be skipped for selected frames or regions within a frame based on different surgical operations, different time points during surgical operations and/or for different portions of an image frame (e.g., subset of pixels, patches in an image frame)”); and displaying the denoised video signal on a digital video display, wherein the denoising the digital video signal is performed in real time as the video signal is displayed (see para [0046]; “Quality of the live video may be improved in real-time such that the physician or surgeon may view the improved video on the display 103 without time delay”, see also para [0052]; “The live video displayed on the display may be processed by the live imaging enhancement system 211 and have improved quality”, and para [0054]; “the live imaging enhancement system 211 may allow for different operation modes. In some cases, the different operation modes may comprise at least a live video denoising mode”). Accordingly, it would have been obvious to one of ordinary skill in the art before the effecting filling date of the invention to modify a kind of method and device of video noise estimation of Liu et al. in view of method an improved video denoising method allowing for spatiotemporal motion resilient video denoising of Gong et al. in order to improve operation safety to both patient and surgeon (see para [0074]).
Regarding claim 5, the rejection of claim 1 is incorporated herein.
Liu et al. in the combination further teach wherein the sampled patch is a first sampled patch, the denoised patch is a first denoised patch (see page 3, 10th para; “a noise value determining module 301, configured to determine a difference between pixel values of each noise reduction region of the current frame video image before noise reduction processing and pixel values of each noise reduction region after noise reduction processing, where the video image is divided into N noise reduction regions, and N is an integer not less than 2”, Note; a noise reduction region corresponds to a first patch, and the after noise reduction corresponds the first denoised patch), and the computer-implemented method further comprises: sampling a second contiguous-pixel portion of the frame, thereby providing a second sampled patch (see page 3, 5th para; “The video image is divided into N noise reduction areas, wherein N is an integer not less than 2. In general, all pixels of a video image are equally divided into N noise reduction regions; Note, N is at least 2 which implies first and second video-image region/patch); wherein the estimate of noise in the digital video signal is based on a statistical combination of (see page 5, 3rd para; “performing histogram statistics on the difference value of each noise reduction area; and determining the noise estimation mean value according to the histogram statistical result”): a first-patch noise metric based on the comparison of the first denoised patch and the first sampled patch, and a second-patch noise metric based on a comparison of the second denoised patch and the second sampled patch (see page 5, 6th para; “the noise value determining module is used for respectively determining the difference value of the pixel value of each noise reduction area of the current frame video image before noise reduction processing and the pixel value of each noise reduction area after noise reduction processing, wherein the video image is divided into N noise reduction areas, and N is an integer not less than 2”).
Gong et al. in the combination further teach and denoising the second sampled patch using the artificial-intelligence-based denoising thereby providing a second denoised patch (see para [0025]; “The deep learning framework for video denoising may dynamically tune the degree of temporal denoising and/or spatial denoising depending on the sequence of input frames and/or the specific application”).
Regarding claim 8, the scope of claim 8 is fully incorporated in claim 1, and the
rejection of claim 1 is equally applicable here.
Regarding claim 12, the rejection of claim 8 is incorporated herein.
Liu et al. in the combination further teach wherein the sampled patch is a first sampled patch, the denoised patch is a first denoised patch (see page 3, 10th para; “a noise value determining module 301, configured to determine a difference between pixel values of each noise reduction region of the current frame video image before noise reduction processing and pixel values of each noise reduction region after noise reduction processing, where the video image is divided into N noise reduction regions, and N is an integer not less than 2”, Note; a noise reduction region corresponds to a first patch, and the after noise reduction corresponds the first denoised patch), and the operations further comprises: sampling a second contiguous-pixel portion of the frame, thereby providing a second sampled patch (see page 3, 5th para; “The video image is divided into N noise reduction areas, wherein N is an integer not less than 2. In general, all pixels of a video image are equally divided into N noise reduction regions; Note, N is at least 2 which implies first and second video-image region/patch); wherein the estimate of noise in the digital video signal is based on a statistical combination of (see page 5, 3rd para; “performing histogram statistics on the difference value of each noise reduction area; and determining the noise estimation mean value according to the histogram statistical result”): a first-patch noise metric based on the comparison of the first denoised patch and the first sampled patch, and a second-patch noise metric based on a comparison of the second denoised patch and the second sampled patch (see page 5, 6th para; “the noise value determining module is used for respectively determining the difference value of the pixel value of each noise reduction area of the current frame video image before noise reduction processing and the pixel value of each noise reduction area after noise reduction processing, wherein the video image is divided into N noise reduction areas, and N is an integer not less than 2”).
Gong et al. in the combination further teach and denoising the second sampled patch using the artificial-intelligence-based denoising thereby providing a second denoised patch (see para [0025]; “The deep learning framework for video denoising may dynamically tune the degree of temporal denoising and/or spatial denoising depending on the sequence of input frames and/or the specific application”).
Regarding claim 15, the scope of claim 15 is fully incorporated in claim 1, and the
rejection of claim 1 is equally applicable here (see also para [0009]; “a memory for storing a set of software instructions, and (iii) one or more processors configured to execute the set of software instructions” of Gong et al.).
Regarding claim 19, the rejection of claim 15 is incorporated herein.
Liu et al. in the combination further teach wherein the sampled patch is a first sampled patch, the denoised patch is a first denoised patch (see page 3, 10th para; “a noise value determining module 301, configured to determine a difference between pixel values of each noise reduction region of the current frame video image before noise reduction processing and pixel values of each noise reduction region after noise reduction processing, where the video image is divided into N noise reduction regions, and N is an integer not less than 2”, Note; a noise reduction region corresponds to a first patch, and the after noise reduction corresponds the first denoised patch), and the operations further comprises: sampling a second contiguous-pixel portion of the frame, thereby providing a second sampled patch (see page 3, 5th para; “The video image is divided into N noise reduction areas, wherein N is an integer not less than 2. In general, all pixels of a video image are equally divided into N noise reduction regions; Note, N is at least 2 which implies first and second video-image region/patch); wherein the estimate of noise in the digital video signal is based on a statistical combination of (see page 5, 3rd para; “performing histogram statistics on the difference value of each noise reduction area; and determining the noise estimation mean value according to the histogram statistical result”): a first-patch noise metric based on the comparison of the first denoised patch and the first sampled patch, and a second-patch noise metric based on a comparison of the second denoised patch and the second sampled patch (see page 5, 6th para; “the noise value determining module is used for respectively determining the difference value of the pixel value of each noise reduction area of the current frame video image before noise reduction processing and the pixel value of each noise reduction area after noise reduction processing, wherein the video image is divided into N noise reduction areas, and N is an integer not less than 2”).
Gong et al. in the combination further teach and denoising the second sampled patch using the artificial-intelligence-based denoising thereby providing a second denoised patch (see para [0025]; “The deep learning framework for video denoising may dynamically tune the degree of temporal denoising and/or spatial denoising depending on the sequence of input frames and/or the specific application”).
Claims 2, 9, 16 are rejected under 35 U.S.C. 103 as being unpatentable over Liu et al. in view of Gong et al. as applied in claim 1 above, and further in view of Tico et al. (US 11151702 B1).
Regarding claim 2, the rejection of claim 1 is incorporated herein. The combination of Liu et al. and Gong et al. as a whole does not teach wherein the artificial-intelligence-based denoising is performed using a neural processing unit (NPU) or a graphics processing unit (GPU) of a system-on-a-chip (SoC).
In the same field of endeavor, Tico et al. teaches wherein the artificial-intelligence-based denoising is performed using a neural processing unit (NPU) or a graphics processing unit (GPU) of a system-on-a-chip (SoC) (see col. 23, lines 12-14; “Processor 605 may be a system-on-chip (SOC) such as those found in mobile devices and include one or more dedicated graphics processing units (GPUs)…. graphics hardware 620 may include one or more programmable graphics processing units (GPUs) and/or one or more specialized SOCs, e.g., an SOC specially designed to implement neural network and machine learning operations”). Accordingly, it would have been obvious to one of ordinary skill in the art before the effecting filling date of the invention to modify a kind of method and device of video noise estimation of Liu et al. in view of method an improved video denoising method allowing for spatiotemporal motion resilient video denoising of Gong et al. and machine learning to perform improved image fusion and/or noise reduction of Tico et al. in order to increase efficiency and memory utilization (see col. 23, lines 12-14).
Regarding claim 9, the rejection of claim 8 is incorporated herein.
Tico et al. in the combination further teach wherein the system further comprises a neural processing unit (NPU) of a system-on-a-chip (SoC) or a graphics processing unit (GPU) of the SoC, and wherein the artificial-intelligence-based denoising is performed using the NPU or the GPU of the SoC (see col. 23, lines 12-14; “Processor 605 may be a system-on-chip (SOC) such as those found in mobile devices and include one or more dedicated graphics processing units (GPUs)…. graphics hardware 620 may include one or more programmable graphics processing units (GPUs) and/or one or more specialized SOCs, e.g., an SOC specially designed to implement neural network and machine learning operations”). Accordingly, it would have been obvious to one of ordinary skill in the art before the effecting filling date of the invention to modify a kind of method and device of video noise estimation of Liu et al. in view of method an improved video denoising method allowing for spatiotemporal motion resilient video denoising of Gong et al. and machine learning to perform improved image fusion and/or noise reduction of Tico et al. in order to increase efficiency and memory utilization (see col. 23, lines 12-14).
Regarding claim 16, the rejection of claim 15 is incorporated herein.
Tico et al. in the combination further teach wherein the artificial-intelligence-based denoising is performed using a neural processing unit (NPU) or a graphics processing unit (GPU) of a system-on-a-chip (SoC) (see col. 23, lines 12-14; “Processor 605 may be a system-on-chip (SOC) such as those found in mobile devices and include one or more dedicated graphics processing units (GPUs)…. graphics hardware 620 may include one or more programmable graphics processing units (GPUs) and/or one or more specialized SOCs, e.g., an SOC specially designed to implement neural network and machine learning operations”).
Claims 3, 10, and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Liu et al. Gong et al. in view of Tico et al. as applied in claims 1-2 above, and further in view of Zhang et al. NPL “Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising”.
Regarding claim 3, the rejection of claim 2 is incorporated herein. The combination of Liu et al., Gong et al. and Tico et al. as a whole does not teach wherein the artificial-intelligence-based denoising is performed using a diffusion model or a WDnCNN, FFDNet, DnCNN, BM3D, or C-BM3D method.
In the same field of endeavor, Zhang et al. teaches wherein the artificial-intelligence-based denoising is performed using a diffusion model or a WDnCNN, FFDNet, DnCNN, BM3D, or C-BM3D method (see page 3150, Fig. 9; “Fig. 9. Average PSNR improvement over BM3D/CBM3D with respect to different noise levels by our DnCNN-B/CDnCNN-B model. The results are evaluated on the gray/color BSD68 dataset”). Accordingly, it would have been obvious to one of ordinary skill in the art before the effecting filling date of the invention to modify a kind of method and device of video noise estimation of Liu et al. in view of method an improved video denoising method allowing for spatiotemporal motion resilient video denoising of Gong et al. and machine learning to perform improved image fusion and/or noise reduction of Tico et al. and further in view of residual learning of Deep CNN for image denoising of Zhang et al. in order to speed up the training process as well as boost the denoising performance (see page 3150, Fig. 9).
Regarding claim 10, the rejection of claim 9 is incorporated herein.
Zhang et al. in the combination further teach wherein the artificial-intelligence-based denoising is performed using a diffusion model or a WDnCNN, FFDNet, DnCNN, BM3D, or C-BM3D method (see page 3150, Fig. 9; “Fig. 9. Average PSNR improvement over BM3D/CBM3D with respect to different noise levels by our DnCNN-B/CDnCNN-B model. The results are evaluated on the gray/color BSD68 dataset”). Accordingly, it would have been obvious to one of ordinary skill in the art before the effecting filling date of the invention to modify a kind of method and device of video noise estimation of Liu et al. in view of method an improved video denoising method allowing for spatiotemporal motion resilient video denoising of Gong et al. and machine learning to perform improved image fusion and/or noise reduction of Tico et al. and further in view of residual learning of Deep CNN for image denoising of Zhang in order to speed up the training process as well as boost the denoising performance (see page 3150, Fig. 9).
Regarding claim 17, the rejection of claim 16 is incorporated herein.
Zhang et al. in the combination further teach wherein the artificial-intelligence-based denoising is performed using a diffusion model or a WDnCNN, FFDNet, DnCNN, BM3D, or C-BM3D method (see page 3150, Fig. 9; “Fig. 9. Average PSNR improvement over BM3D/CBM3D with respect to different noise levels by our DnCNN-B/CDnCNN-B model. The results are evaluated on the gray/color BSD68 dataset”).
Claims 4, 11, and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Liu et al. and Gong et al. in view of Tico et al. and Zhang et al.as applied in claims 1-3 above, and further in view of Xue et al. (US 20190180443 A1).
Regarding claim 4, the rejection of claim 3 is incorporated herein. The combination of Liu et al., Gong et al. Tico et al. and Zhang et al. as a whole does not teach wherein the sampled patch is a fifty pixel by fifty pixel patch.
In the same field of endeavor, Xue et al. teaches wherein the sampled patch is a fifty pixel by fifty pixel patch (see para [0123]; “processing logic generates patches from the image… The image patches may be ROIs for the image, and may have a predefined size. Some example sizes for image patches are …..50×50 pixels”). Accordingly, it would have been obvious to one of ordinary skill in the art before the effecting filling date of the invention to modify a kind of method and device of video noise estimation of Liu et al. in view of method an improved video denoising method allowing for spatiotemporal motion resilient video denoising of Gong et al. and machine learning to perform improved image fusion and/or noise reduction of Tico et al. and further in view of residual learning of Deep CNN for image denoising of Zhang and A machine learning model is trained to define bounding shapes around teeth in images of Xue et al. in order to use deep learning with low-cost edge detection for tooth detection (see para [0123]).
Regarding claim 11, the rejection of claim 10 is incorporated herein.
Xue et al.in the combination teaches wherein the sampled patch is a fifty pixel by fifty pixel patch (see para [0123]; “processing logic generates patches from the image… The image patches may be ROIs for the image, and may have a predefined size. Some example sizes for image patches are …..50×50 pixels”). Accordingly, it would have been obvious to one of ordinary skill in the art before the effecting filling date of the invention to modify a kind of method and device of video noise estimation of Liu et al. in view of method an improved video denoising method allowing for spatiotemporal motion resilient video denoising of Gong et al. and machine learning to perform improved image fusion and/or noise reduction of Tico et al. and further in view of residual learning of Deep CNN for image denoising of Zhang and A machine learning model is trained to define bounding shapes around teeth in images of Xue et al. in order to use deep learning with low-cost edge detection for tooth detection (see para [0123]).
Regarding claim 18, the rejection of claim 17 is incorporated herein.
Xue et al.in the combination teaches wherein the sampled patch is a fifty pixel by fifty pixel patch (see para [0123]; “processing logic generates patches from the image… The image patches may be ROIs for the image, and may have a predefined size. Some example sizes for image patches are …..50×50 pixels”).
Claims 6, 13, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Liu et al. in view of Gong et al. as applied in claims 1 and 5 above, and further in view of Baeza et al. (US 20090002379 A1).
Regarding claim 6, the rejection of claim 5 is incorporated herein.
Liu et al. in the combination further teach further comprising, after the denoising the first sampled patch (the noise value determining module is used for respectively determining the difference value of the pixel value of each noise reduction area of the current frame video image before noise reduction processing and the pixel value of each noise reduction area after noise reduction processing). However, the combination of Liu et al. and Gong et al. as a whole does not teach further comprising, after the denoising the first sampled patch: determining sufficient processor cycle availability or sufficient control loop time to perform the denoising of the second sampled patch, wherein the denoising the second sampled patch is based on the determining sufficient processor cycle availability or sufficient control loop time.
In the same field of endeavor, Baeza et al. teach determining sufficient processor cycle availability or sufficient control loop time to perform the denoising of the second sampled patch, wherein the denoising the second sampled patch is based on the determining sufficient processor cycle availability or sufficient control loop time (see para [0709]; “a decoder may encounter content that is harder to decode (e.g., because it has a higher complexity or is encoded at a higher quality) or the decoder may experience an unexpected resource shortage (e.g., lack of available processor cycles or memory because of other processes running)”). Accordingly, it would have been obvious to one of ordinary skill in the art before the effecting filling date of the invention to modify a kind of method and device of video noise estimation of Liu et al. in view of method an improved video denoising method allowing for spatiotemporal motion resilient video denoising of Gong et al. and video decoding innovations for multithreading implementations and graphics processor unit of Baeza et al. in order to (see para [0709]).
Regarding claim 13, the rejection of claim 12 is incorporated herein.
Liu et al. in the combination further teach further comprising, after the denoising the first sampled patch (the noise value determining module is used for respectively determining the difference value of the pixel value of each noise reduction area of the current frame video image before noise reduction processing and the pixel value of each noise reduction area after noise reduction processing).
Baeza et al. in the combination further teach determining sufficient processor cycle availability or sufficient control loop time to perform the denoising of the second sampled patch, wherein the denoising the second sampled patch is based on the determining sufficient processor cycle availability or sufficient control loop time (see para [0709]; “a decoder may encounter content that is harder to decode (e.g., because it has a higher complexity or is encoded at a higher quality) or the decoder may experience an unexpected resource shortage (e.g., lack of available processor cycles or memory because of other processes running)”). Accordingly, it would have been obvious to one of ordinary skill in the art before the effecting filling date of the invention to modify a kind of method and device of video noise estimation of Liu et al. in view of method an improved video denoising method allowing for spatiotemporal motion resilient video denoising of Gong et al. and video decoding innovations for multithreading implementations and graphics processor unit of Baeza et al. in order to (see para [0709]).
Regarding claim 20, the rejection of claim 19 is incorporated herein.
Liu et al. in the combination further teach further comprising, after the denoising the first sampled patch (the noise value determining module is used for respectively determining the difference value of the pixel value of each noise reduction area of the current frame video image before noise reduction processing and the pixel value of each noise reduction area after noise reduction processing).
Baeza et al. in the combination further teach determining sufficient processor cycle availability or sufficient control loop time to perform the denoising of the second sampled patch, wherein the denoising the second sampled patch is based on the determining sufficient processor cycle availability or sufficient control loop time (see para [0709]; “a decoder may encounter content that is harder to decode (e.g., because it has a higher complexity or is encoded at a higher quality) or the decoder may experience an unexpected resource shortage (e.g., lack of available processor cycles or memory because of other processes running)”).
Claims 7, and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Liu et al. in view of Gong et al. as applied in claims 1 and 5 above, and further in view of Po et al. (US 20190362484 A1).
Regarding claim 7, the rejection of claim 5 is incorporated herein.
Liu et al. in the combination further teach further comprising: sampling third and fourth contiguous-pixel portions of the frame, thereby providing third and fourth sampled patches, respectively (see page 3, 5th para; “The video image is divided into N noise reduction areas, wherein N is an integer not less than 2. In general, all pixels of a video image are equally divided into N noise reduction regions. One special case is that one pixel acts as a noise reduction region” Note; N is open ended not less than 2, and disclose more than two regions including at least four sampled regions), wherein the estimate of noise in the digital video signal is based on a statistical combination of (see page 3, 2nd para; “performing histogram statistics on the difference value of each noise reduction region; and step 100, respectively determining the difference value of the pixel value of each noise reduction area of the current frame video image before the noise reduction processing and the pixel value of each noise reduction area after the noise reduction processing”): the first-patch noise metric, the second-patch noise metric, a third-patch noise metric based on the comparison of the third denoised patch and the third sampled patch, and a fourth-patch noise metric based on a comparison of the fourth denoised patch and the fourth sampled patch (see page 3, 9th para; “determining the difference between the pixel values of the noise reduction region before the noise reduction processing and the pixel values of the noise reduction processing may be: and determining the difference value of the pixel value of each pixel point included in the noise reduction area before and after the noise reduction processing. That is to say, the difference value of the noise reduction region is a set of difference values between pixel values of each pixel included in the noise reduction region before and after the noise reduction processing”, see also page 3, 2nd para; “performing histogram statistics on the difference value of each noise reduction region; and step 100, respectively determining the difference value of the pixel value of each noise reduction area of the current frame video image before the noise reduction processing and the pixel value of each noise reduction area after the noise reduction processing”).
Gong et al. in the combination further teach and denoising the third and fourth sampled patches using the artificial-intelligence-based denoising, thereby providing third and fourth denoised patches, respectively (see para [0025]; “a deep learning framework that can perform real-time video quality enhancement or video denoising during live video acquisition”, see also para [0082]; “DeepFlow, Farneback algorithm or LiteFlowNet may be applied to the live video to estimate motion at a point in time and/or location of patches within an image frame, then window size for processing the input image stream or a selected patch of the image frame may be automatically adjusted”). However, the combination of Liu et al. and Gong et al. as a whole does not teach wherein the first, second, third, and fourth sampled patches are sampled at random or constrained-random spatial locations within the frame.
In the same field of endeavor, Po et al. teach wherein the first, second, third, and fourth sampled patches are sampled at random or constrained-random spatial locations within the frame (see para [0017]; “extracting one or more image patches randomly from the input image; measuring the moment of the extracted image patches; …..repeating the image patches selection process until sufficient extracted image patches are accepted, with adjusted settings if necessary”). Accordingly, it would have been obvious to one of ordinary skill in the art before the effecting filling date of the invention to modify a kind of method and device of video noise estimation of Liu et al. in view of method an improved video denoising method allowing for spatiotemporal motion resilient video denoising of Gong et al. and a method for image patch selection for training a neural network for image quality assessment of Po et al. in order to generate quality score estimation (see para [0017]).
Regarding claim 14, the rejection of claim 12 is incorporated herein.
Liu et al. in the combination further teach wherein the operation further comprising: sampling third and fourth contiguous-pixel portions of the frame, thereby providing third and fourth sampled patches, respectively (see page 3, 5th para; “The video image is divided into N noise reduction areas, wherein N is an integer not less than 2. In general, all pixels of a video image are equally divided into N noise reduction regions. One special case is that one pixel acts as a noise reduction region” Note; N is open ended not less than 2, and disclose more than two regions including at least four sampled regions), wherein the estimate of noise in the digital video signal is based on a statistical combination of (see page 3, 2nd para; “performing histogram statistics on the difference value of each noise reduction region; and step 100, respectively determining the difference value of the pixel value of each noise reduction area of the current frame video image before the noise reduction processing and the pixel value of each noise reduction area after the noise reduction processing”): the first-patch noise metric, the second-patch noise metric, a third-patch noise metric based on the comparison of the third denoised patch and the third sampled patch, and a fourth-patch noise metric based on a comparison of the fourth denoised patch and the fourth sampled patch (see page 3, 9th para; “determining the difference between the pixel values of the noise reduction region before the noise reduction processing and the pixel values of the noise reduction processing may be: and determining the difference value of the pixel value of each pixel point included in the noise reduction area before and after the noise reduction processing. That is to say, the difference value of the noise reduction region is a set of difference values between pixel values of each pixel included in the noise reduction region before and after the noise reduction processing”, see also page 3, 2nd para; “performing histogram statistics on the difference value of each noise reduction region; and step 100, respectively determining the difference value of the pixel value of each noise reduction area of the current frame video image before the noise reduction processing and the pixel value of each noise reduction area after the noise reduction processing”).
Gong et al. in the combination further teach and denoising the third and fourth sampled patches using the artificial-intelligence-based denoising, thereby providing third and fourth denoised patches, respectively (see para [0025]; “a deep learning framework that can perform real-time video quality enhancement or video denoising during live video acquisition”, see also para [0082]; “DeepFlow, Farneback algorithm or LiteFlowNet may be applied to the live video to estimate motion at a point in time and/or location of patches within an image frame, then window size for processing the input image stream or a selected patch of the image frame may be automatically adjusted”).
Po et al. in the combination further teach wherein the first, second, third, and fourth sampled patches are sampled at random or constrained-random spatial locations within the frame (see para [0017]; “extracting one or more image patches randomly from the input image; measuring the moment of the extracted image patches; …..repeating the image patches selection process until sufficient extracted image patches are accepted, with adjusted settings if necessary”). Accordingly, it would have been obvious to one of ordinary skill in the art before the effecting filling date of the invention to modify a kind of method and device of video noise estimation of Liu et al. in view of method an improved video denoising method allowing for spatiotemporal motion resilient video denoising of Gong et al. and a method for image patch selection for training a neural network for image quality assessment of Po et al. in order to generate quality score estimation (see para [0017]).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to WINTA GEBRESLASSIE whose telephone number is (571)272-3475. The examiner can normally be reached Monday-Friday9:00-5:00.
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, Andrew Bee can be reached at 571-270-5180. 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.
/WINTA GEBRESLASSIE/Examiner, Art Unit 2677
Prosecution Insights