9 results about "Diffusion filter" patented technology

A method of processing a mammogram image, including receiving a digital mammography image of a breast of a patient, and separating the digital mammography image into three channels. The method includes removing noise and uneven illumination to form a first set of three uniform background images, and separating a luminance channel and a chrominance channel from the first set with a transfer function to form a second set of three uniform background images. The method includes applying principal component analysis to the second set; and assembling a single grayscale image and applying a first subspace projection. The method includes removing at least a portion of a pectoral muscle from the grayscale image, and normalizing the muscle removed image using a second order Gaussian Laplacian and an oriented diffusion filter to form a processed image. The method includes identifying cancerous and / or noncancerous regions in the processed image.

The invention relates to the technical field of seismic exploration data processing, in particular to a depth domain seismic data volume high-wavenumber imaging noise iterative filtering method and a depth domain seismic data volume high-wavenumber imaging noise iterative filtering device. The method comprises the following steps: acquiring parameter information of an initial depth domain seismic imaging data volume and a velocity model; carrying out numerical discretization on the diffusion filtering equation according to the parameter information according to a set time step length, a diffusion coefficient and a grid step length, and constructing a first filtering coefficient; according to the parameter information, carrying out numerical discretization on the seismic wave equation according to a set time step length, a speed model and a grid step length, and constructing a second filter coefficient; constructing a filtering operator according to the filtering coefficient; and carrying out imaging noise iterative filtering processing on the initial depth domain seismic imaging data volume by using a filtering operator to obtain a filtered depth domain seismic data volume. The problems of trace processing channel by channel and high-wave-number imaging noise amplitude-preserving suppression are effectively solved, so that the false appearance of reservoir imaging can be eliminated, and depth domain structure interpretation and reservoir research are facilitated.

A method of processing video data is disclosed. The method includes performing a conversion between a current block of a video and a bitstream of the video by using at least one of a combined inter-intra prediction (CIIP), a diffusion filtering, a bilateral filtering, a transform domain filtering, or another type of post-reconstruction filtering that is different from the diffusion filtering, the bilateral filtering and the transform domain filtering. The reconstructed neighboring samples of the current block used in at least one of the CIIP, the diffusion filtering, the bilateral filtering, the transform domain filtering, or the another type of post-reconstruction filtering are replaced by approximated samples generated from corresponding samples of the reconstructed neighboring samples in one or more reference pictures.

The invention discloses a fault automatic detection method based on multi-feature fusion and structural guiding filtering, and relates to the technical field of geological data detection, and the method comprises the steps: obtaining a three-dimensional coherent body which is corresponding to seismic data of a target region and reflects the stratum discontinuity; processing the fault features in the three-dimensional coherent body through a multi-scale multi-direction filter to generate a fault boundary enhanced image; according to the fault inclination angle in the three-dimensional coherent body, direction diffusion coefficients corresponding to the principal direction gradient and the vertical direction gradient are calculated; based on the directional diffusion coefficient, performing iterative diffusion filtering processing on the three-dimensional coherent body to generate a fault continuity enhanced image; and generating a fault detection result based on a fusion image between the fault boundary enhanced image and the fault continuity enhanced image. According to the invention, the false detection rate during fault identification is reduced, and the identification precision is improved.

The invention relates to an anti-glare filter module (1) for a projector (20), the anti-glare filter module (1) having: • at least one filter disc (2, 3), which is made as a diffusion filter; • at least one frame module (4), on which the filter disc (2, 3) is mounted so as to be able to slide along a direction of travel (100); • wherein the frame module (4) is made to position the anti-glare filter module (1) on an optical path (200) of the projector (20) such that the direction of travel (100) is oriented transversely to the optical path (200); • wherein the frame module (4) has a rotatably mounted drive element (5), which is made to be rotated by a user; • wherein the frame module (4) has a force transmission system (6), which is made to transmit a force between the drive element (5) and the disc filter (2, 3) for sliding,By rotating the attack element (5), the filter disk (2, 3) moves, in whole or in part, along the optical path (200) in the direction of movement (100). Figure 1,

The utility model discloses a tempered glass stress spot detection lamp box, and belongs to the technical field of tempered glass detection. A tempered glass stress spot detection lamp box comprises a working box with a clamping assembly and further comprises a transparent plate fixedly connected to the working box, a supporting plate is fixedly connected to the working box, the supporting plate is perpendicular to the transparent plate, a diffusion filter plate is arranged on the transparent plate, and the bottom end of the diffusion filter plate is attached to the supporting plate; the transverse movable block is connected into the working box in a sliding mode, a longitudinal movable block is connected to the transverse movable block in a sliding mode, and a direct light is fixedly connected to the longitudinal movable block; according to the utility model, the direct illuminating lamp can move up and down and left and right in the working box to irradiate different positions of the toughened glass in a concentrated manner, so that when a worker wears a polariscope to observe, the worker can carefully observe the local positions of the toughened glass which may have fuzzy stress spots, and the operator can clearly see the stress spots; and the observation difficulty is reduced.

An automotive lighting unit (1) comprises: a rear housing (2) adapted to be fixed on a vehicle; a front flap (3; 30) arranged to close the opening of the rear housing (2) to form a hard container (5) and provided with a rectangular through-seam (4) in the center; a light guide plate (6) made of a light guide material and inserted in a penetrating manner into the through-slot (4) of the front baffle (3; a lighting assembly (7) emitting light according to instructions and arranged within the rigid container (5) facing the rear side wall (6a) of the light guide plate (6) to guide the generated light rays through the rear side wall (6a) into the body of the light guide plate (6) such that the light rays propagate within the light guide plate (6) by total internal reflection to the front side wall (6b) of the same light guide plate (6); and a diffusion filter (8) interposed between the lighting assembly (7) and the light guide plate (6).

The invention discloses a multi-scale fractal deep network infrared false alarm source classification method and system, belongs to the technical field of remote sensing image processing, and solves the problems that in the prior art, false alarm sources of different scenes and different features are difficult to classify by utilizing texture information with effective generality, a set of general algorithm is difficult to design to solve multi-category false alarm source classification, and the classification efficiency is low. And a classification method is poor in adaptability in a complex background environment and relatively high in training data volume requirement. The method comprises the following steps: reading an infrared remote sensing image, extracting edge contour features of the infrared remote sensing image by using an anisotropic differential diffusion filter, and stacking the edge contour features and an original infrared remote sensing image in an image channel number dimension to be input into a Swin Transform network; and depth features output by each Stage layer in the Swin Transformer network are input into the false alarm source classification network to obtain different false alarm sources, the different false alarm sources are classified and labeled by adopting different identification colors, and the result is labeled on the infrared remote sensing image, and the method is used for multi-scale fractal depth network infrared false alarm source classification.

The invention relates to the technical field of seismic data processing, and discloses a signal-to-noise ratio optimization method and system for post-stack two-dimensional seismic data. The method comprises the following steps: performing well-to-seismic joint calibration on original two-dimensional seismic data to determine main target stratum distribution, and performing spectral analysis to identify a noise frequency band; based on the main target layer distribution, performing segmented Q compensation processing on the original data to obtain a seismic data volume after Q compensation processing; on the basis, spectrum shaping processing is carried out according to the identified noise frequency band so as to suppress the noise and preliminarily improve the signal-to-noise ratio; and finally, diffusion filtering processing is carried out on the data volume after spectrum shaping processing, and a final two-dimensional seismic data volume with an improved signal-to-noise ratio is obtained. According to the method, through series application of Q compensation, spectrum shaping and diffusion filtering, the continuity of the event is effectively improved, the definition of fracture imaging is improved, and a more reliable data basis is provided for structural interpretation and trap evaluation.