64 results about "Anisotropic diffusion filter" patented technology

A method for determining a candidate lesion region in a digital ultrasound medical image of anatomical tissue. The method includes the steps of: accessing the digital ultrasound medical image of anatomical tissue; applying an anisotropic diffusion filter to the ultrasound image to generate a filtered ultrasound image; performing a normalized cut operation on the filtered ultrasound image to partition the filtered ultrasound image into a plurality of regions; and selecting, from the plurality of regions, at least one region as a candidate lesion region.

A system and method of image processing for smoothing, denoising, despeckling and sharpening scanned document images which is performed prior to a compression. The scanned image is selectively smoothed by anisotropic diffusion filtering in a single iteration with a 3×3 kernel, which provides denoising, edge-preserving smoothing. The smoothed image data is then selectively sharpened using variable contrast mapping that provides overshoot-free variable-sharpening and despeckling. Image quality is improved while increasing compressibility of the image.

The invention discloses a method for making oral implant positioning guiding template based on CBCT data. The method comprises the steps: 1) scanning patient jaw bones through CBCT to mainly obtain jaw bone three-dimensional anatomic structure information of a implant region; 2) analyzing and processing CBCT data; 3) using the processed CBCT data to reconstruct a three-dimensional jaw bone model; 4) processing the reconstructed three-dimensional jaw bone model; 5) designing an oral implant positioning guiding template based on the three-dimensional jaw bone model; and 6) carrying out rapid prototype processing; Amount of influence data is effectively reduced through carrying out resampling, anisotropic diffusion filtering to original image data obtained through CBCT scanning, carrying out volume rendering through a ray-casting method, and noise of the data is further removed, thereby accurate reconstruction of the three-dimensional jaw bone model is guaranteed. Furthermore, accuracy of manufacturing the oral implant positioning guiding template is substantially improved through removing model fragments caused by image artifact and non-implant region models in the process of three dimensional jaw bone model reconstruction.

The invention discloses an SAR (Synthetic Aperture Radar) image change detection method based on a neighborhood logarithm specific value and anisotropic diffusion, relating to the field of remote sensing image processing and mainly solving the problem that a difference graph structure of SAR image change detection is seriously influenced by SAR image spot noises. The SAR image change detection method comprises the following steps: (1) structuring a difference striograph IL of two images I1 and I2 of different times and same terrain according to a neighborhood logarithm specific value method; (2) carrying out self-adaptation window anisotropic diffusion filtering processing on the difference striograph IL to obtain a final filtering result graph NI<t>[L] of the difference striograph; and (3) carrying out threshold segmentation on the final filtering result graph NI<t>[L] of the difference striograph by using an OSTU (Maximum Between-Class Variance) threshold algorithm to obtain a change detection result graph CNI<t>[L] for structuring the difference striograph by using the neighborhood logarithm specific value method. The histogram of the difference striograph can be compressed so as to effectively eliminate miscellaneous points in the change detection result graph; and the self-adaptation window anisotropic diffusion filtering has favorable edge retentiveness and cannot blurs the edges of the image, thus, an obtained change detection result graph is finer.

The invention discloses a mark point automatic registration method based on model matching. The method comprises the following steps of (1)obtaining image data containing mark points; (2) constructing a mark point model; (3) reading in the obtained image data, conducting anisotropic diffusion filtering on images, and automatically extracting a skin three-dimensional grid from the images; (4) adopting an ICP algorithm to be matched with the mark point model and the skin three-dimensional grid, and obtaining the coordinate of the center of each mark point on the skin three-dimensional grid in an image space coordinate system; (5) adopting the ICP algorithm to be matched with the center of each mark point of the image space and the center of each mark point of the actual space, obtaining a rotating matrix R and a translational vector quantity T between the image space coordinate system and the actual space coordinate system, and completing registration of the mark points. According to the mark point registration method, the multiple mark points can be rapidly registered, the obtained rotating matrix R and the obtained translational vector quantity T of the image coordinate system and the actual coordinate system are more accurate, manual interference is reduced, the precision of the registration of the mark points is improved, and the mark point automatic registration method has good robustness.

The invention discloses a method and a device for removing CT (computed tomography) image noises, and relates to the technical field of image noise removal. The method comprises the following steps of: estimating the tissue weight of an image, estimating the noise level of the image, calculating a noise removing parameter, carrying out anisotropic diffusion filtering on the image, carrying out edge enhancement on the image subjected to filtering output, enhancing details of the image and correcting the contrast ratio, cutting the image, and outputting the result. The device comprises a module for estimating the tissue weight of the image, a module for estimating the noise level of the image, a module for calculating the noise removing parameter, a module for carrying out anisotropic diffusion filtering on the image, a module for carrying out edge enhancement on the image subjected to filtering output, a module for performing detail enhancement and contrast ratio rectification on the image, and a module for cutting images and outputting the result. The method and the device can maintain the image edge and the original contrast ratio of the image while effectively removing the high-frequency noises of the CT images.

A method for reducing noise in a computed tomographic (CT) image includes acquiring both a first set of projection views and a second set of projection views, wherein for each projection view in the first set of projection views there is an associated projection view in the second set of projection views representing the same object scanned at substantially the same time from substantially the same position. The method further includes reconstructing the first set of projection views and the associated second set of projection views to obtain a first image and a second image, respectively. Next, the first image and the second image are combined to obtain a noise map and an amount of noise in a product image is estimated utilizing the noise map. The method also includes filtering using the noise map to perform noise reduction.

The invention discloses an adaptive denoising method for an ultrasonic image. The adaptive denoising method comprises the following steps: (1) obtaining the ultrasonic image; (2) performing Gaussian-Laplacian pyramid decomposition on the ultrasonic image and obtaining Gaussian layers and Laplacian layers at different scales; (3) calculating a structure tensor and a diffusion tensor of the Gaussian layer at each scale and performing anisotropic diffusion filtering treatment on the Gaussian layer at the scale; (4) according to a characteristic value of the structure tensor of the Gaussian layer, designing a grey mapping curve, and performing grey mapping on the Gaussian layer at the scale according to the grey mapping curve; (5) repeating the steps (3) and (4) for multiple times and performing the same treatment on the Gaussian layer and the Laplacian layer at each scale; (6) performing reverse reconstruction on the treated Gaussian layer and Laplacian layer and obtaining a denoised ultrasonic image; (7) outputting the denoised ultrasonic image. The adaptive denoising method can inhibit edge enhancement and spots of the image and is simple in algorithm and strong in adaptability.

The invention relates to a multi-scale anisotropic diffusion filtering method based on pre-stack CRP trace sets. The method includes the steps that firstly, the input pre-stack CRP trace sets are regularized; secondly, multi-scale decomposition is conducted through a two-dimensional Mallat algorithm, and each decomposed sub section is initialized; thirdly, a diffusion threshold value is obtained through a diffusion coefficient method, and the sixth step is executed; after diffusion tensor parameters are obtained based on anisotropic diffusion of a diffusion tensor method, the parameters are substituted into a nonlinear anisotropic equation so as to conduct anisotropic diffusion filtering, and the fourth step is executed; fourthly, the SNR, the MSE and the PSNR of the sub sections obtained after each iteration are calculated, and then the optimal earthquake sub section is optimized; fifthly, processing of the fourth step is executed on all the sub sections obtained after anisotropic diffusion filtering, and the sixth step is executed; sixthly, pre-stack CRP trace set nondestructive reconstruction is conducted on information of all the iterated sub sections through a two-dimensional Mallat reconstruction algorithm, and then the optimal pre-stack CRP trace set is output after reconstruction. The multi-scale anisotropic diffusion filtering method can be widely applied to the processing process of various oil exploration earthquake data.

A computer system for registering synthetic aperture radar (SAR) images includes a database for storing SAR images to be registered, and a processor for registering SAR images from the database. The registering includes selecting first and second SAR images to be registered, individually processing the selected first and second SAR images with an anisotropic diffusion algorithm, and registering the first and second SAR images after the processing. A shock filter is applied to the respective first and second processed SAR images before the registering. Elevation data is extracted based on the registered SAR images.

Methods and systems for extracting formation properties from formation logging data is disclosed. A method includes obtaining a deconvolution filter; and processing the formation logging data using the deconvolution filter to produce estimates of the formation properties. The estimates of the formation properties may be further processed with a diffusion filter, which may be an anisotropic diffusion filter. A system for extracting formation properties from formation logging data includes a central processing unit and a memory, wherein the memory stores a program having instructions for performing a method that includes designing a deconvolution filter; and processing the formation logging data using the deconvolution filter to produce estimates of the formation properties.

Methods and systems for extracting formation properties from formation logging data is disclosed. A method includes obtaining a deconvolution filter; and processing the formation logging data using the deconvolution filter to produce estimates of the formation properties. The estimates of the formation properties may be further processed with a diffusion filter, which may be an anisotropic diffusion filter. A system for extracting formation properties from formation logging data includes a central processing unit and a memory, wherein the memory stores a program having instructions for performing a method that includes designing a deconvolution filter; and processing the formation logging data using the deconvolution filter to produce estimates of the formation properties.

Multiresolution decomposition of image data before scan conversion processing is hierarchically performed, low-frequency decomposed image data and high-frequency decomposed image data with first to n-th levels are acquired, nonlinear anisotropic diffusion filtering is performed on output data from a next lower layer or the low-frequency decomposed image data in a lowest layer, and filtering for generating edge information on a signal for every layer is performed from the output data from the next lower layer or the low-frequency decomposed image data in the lowest layer. In addition, on the basis of the edge information on each layer, a signal level of the high-frequency decomposed image data is controlled for every layer and multiresolution mixing of the output data of the nonlinear anisotropic diffusion filter and the output data of the high-frequency level control, which are obtained in each layer, are hierarchically performed.

Provided is an automatic retrieval method for calcified plaque frames in an intravascular ultrasound image sequence. The method comprises the specific steps of (a), carrying out anisotropic diffusion filtering on each frame of IVUS image, (b), carrying out polar coordinate conversion on each frame of IVUS image after the anisotropic diffusion filtering, (c), obtaining the radial gray level changing curve of each angle according to polar coordinate view, (d), initially retrieving the images containing calcified plaque, and (e), carrying out fine retrieval on the images containing the calcified plaque. According to the automatic retrieval method for the calcified plaque frames in the intravascular ultrasound image sequence, the original radio-frequency signals and backscattering signals of IVUS equipment do not need to be collected, the application conditions of the method are lowered, the retrieval process is completed full-automatically, doctors do not need to take part in manually, the doctors are liberated from onerous manual retrieval work, objective and high-repeatability retrieval results can be obtained fast, and reliable bases can be provided for computer-aided diagnosis and formulating interventional therapy schemes of coronary heart diseases.

A cervical caner image automatic partition method based on T2 weighted magnetic resonance imaging (MRI) T2-MRI and dispersion weighted (DW)-MRI includes that a DW-MR image is registered to a T2-MR image by using a non-linear register method, and the registered DW-MR image is sorted; the T2-MR image is filtered through non-linear anisotropic diffusion filtering technology, a bladder and a rectum are partitioned, and an interested area is partitioned through a partition result of the bladder and the rectum; and a combined maximum a posterior (CMAP) method is adopted for an interested area of the T2-MR image and the DW-MR image to conduct precise partition of a tumor. The cervical caner image automatic partition method fully uses effective information of the T2-MR image and the DW-MR image, can effectively overcome effects of noise, partial volume effect and strength overlapping in the T2-MR image, is precise and effective, and has important clinical and application value on prevention, diagnosis and treatment of the cervical cancer.

A computer system for processing synthetic aperture radar (SAR) images includes a database for storing SAR images to be processed, and a processor for processing a SAR image from the database. The processing includes determining noise in a SAR image to be processed, selecting a noise threshold for the SAR image based on the determined noise, and mathematically adjusting an anisotropic diffusion algorithm based on the selected noise threshold. The adjusted anisotropic diffusion algorithm is applied to the SAR image.

The invention discloses an image haze removal method based on a pixel dark channel and anisotropic diffusion filtering, belonging to the technical field of the image haze removal method. The technical key point is that the image haze removal method comprises the following steps of: (1) calculating a dark channel Idark(x) of each pixel point of haze image I(x); (2) calculating an atmospheric light intensity value A according to the dark channel Idark(x) of each pixel point; (3) performing anisotropic diffusion filtering on the dark channel Idark(x) of each pixel point; and (4) calculating transmittance t(x) of each pixel point according to a transmittance calculating formula; and (5) performing image restoration processing according to the atmospheric light intensity value A and the transmittance t(x). The invention aims at providing the image haze removal method, which is relatively small in calculated amount, less in stored resources occupied and quick in processing speed, based on pixel dark channel and anisotropic diffusion filtering. The image haze removal method is used for image haze removal processing.

The invention belongs to the image processing technical field and relates to a wavelet analysis-based image rain-removing method and a wavelet analysis-based image rain-removing system. The wavelet analysis-based image rain-removing method includes the following steps that: image layer decomposition is carried out on a video frame image according to wavelet analysis, and the image information of all decomposition image layers are analyzed; step b, a fusion coefficient matrix is calculated, wavelet fusion is carried out on decomposition image layers with different image information according to the fusion coefficient matrix, and image reconstruction is carried out according to a fusion result, so that a rain removed image can be obtained; and step c, filtering processing is performed on the rain removed image by using anisotropic diffusion filtering. With the wavelet analysis-based image rain-removing method and system of the invention adopted, the interference of dynamic characteristics can be avoided, rain droplets can be removed more effectively and accurately, the usage scope of the rain removing algorithm can be extended, the algorithm can have a sound rain-removing effect even under heavy rain; filtering processing is performed on the rain removed image through the anisotropic diffusion filtering, so that the image can be clearer and more nature, and the rain removing algorithm can be more practical.

The invention discloses a method and device for generating a dental panoramic image on the basis of a CBCT image. The method comprises steps of: determining a seed point in a current axial surface image of a CBCT apparatus, processing the seed point by using a curve fitting algorithm to generate a dental arch line; using a curve surface where the dental arch line is located as a central curve surface, computing body tissue density CT values corresponding to the pixels in the central curve surface according to a set sampling thickness, and generating initial dental panoramic images according to the body tissue density CT values; and determining the gradient factor of an anisotropic diffusion filter formula and processing the initial dental panoramic images with different scales by using the anisotropic diffusion filter formula to generate an final dental panoramic image. The method and device achieve a purpose of obtaining the two-dimensional panoramic image of the entire teeth by using the image processing algorithm on the basis of CBCT three-dimensional data.

The invention uses non-linear anisotropic diffusion filter to pre-treat the image based on the character of intracranial hematoma image, effectively removes noise and artefect and meeting the actual need of medical image treatment, adopts two-step image cutting process to skilfully remove the skull and non-brain tissue without image registering, guides the idea of entropy into the image cutting process to realize the two-dimensional entropy thresholding, adopts pixel grey and neighbourhood grey as parameters to cut the image. The invention reflects the grey information distribution and the image pixel neighborhood space related information, thereby effectively restraining the noise and removing the obstruction and unnecessary small structures; the operator of genetic algorithm is improved, and particularly the prematurity of genetic algorithm is prevented; and the self-adapting big variable operator is adopted to search the best threshold value by modified delaminated genetic algorithm.

Multiresolution decomposition of image data before scan conversion processing is hierarchically performed, low-frequency decomposed image data and high-frequency decomposed image data with first to n-th levels are acquired, nonlinear anisotropic diffusion filtering is performed on output data from a next lower layer or the low-frequency decomposed image data in a lowest layer, and filtering for generating edge information on a signal for every layer is performed from the output data from the next lower layer or the low-frequency decomposed image data in the lowest layer. In addition, on the basis of the edge information on each layer, a signal level of the high-frequency decomposed image data is controlled for every layer and multiresolution mixing of the output data of the nonlinear anisotropic diffusion filter and the output data of the high-frequency level control, which are obtained in each layer, are hierarchically performed.

The invention provides an automatic segmentation method for retina serous pigment epithelial layer detachment. The method comprises the following steps: a, pretreatment: inputting a three-dimensional retinal image obtained by an optical coherence tomography eye imager into a computer, and de-noising an image detached from a retina serous pigment epithelial layer by using a curve anisotropic diffusion filtering method; b, automatic segmentation: layering the image detached from the retina serous pigment epithelial layer by using a graph search algorithm, so as to obtain an initial segmented result; obtaining foreground and background seed points according to the initial segmented result by using a mathematical morphology algorithm; automatically segmenting a retina serous pigment epithelium detachment region by using a graph cut algorithm; c, post-treatment: optimizing the automatic segmentation result by using the mathematical morphology algorithm. According to the invention, the graph search algorithm, the graph cut algorithm and the mathematical morphology algorithm are effectively combined, so that the automatic segmentation of the retina serous pigment epithelial layer detachment region is realized.

The invention discloses a three-dimensional scattered point cloud smoothing denoising method of anisotropic diffusion filtering. A tensor matrix structure tensor matrix is obtained by tensor voting ona sampling point and effective neighboring points thereof, eigenvalues and eigenvectors are solved, according to the eigenvalues and eigenvectors of the structure tensor matrix, local characteristicsof the sampling point are analyzed, the eigenvalues of a diffusion tensor matrix are designed according to different geometric feature information of the sampling point, diffusion rates are designedaccording to the different geometric feature information so that the diffusion rates in different principal feature directions are different, a modified diffusion tensor matrix is reconstructed, finally, the reconstructed diffusion tensor is substituted into a three-dimensional diffusion anisotropic filtering equation for differential solution, and after a certain number of iterations, a filteringfactor is obtained for smoothing noise. The method can effectively remove noise of the scattered point cloud and maintain feature information of an original model at the same time, thereby avoiding excessive smoothing and local distortion.

The invention provides a frequency division diffusion filter fault strengthening method based on optimized Morlet wavelet. The method comprises the following steps: converting post-stack seismic datainto an MATLAB identifiable data format; carrying out spectral analysis on seismic data and extracting the wavelet; improving a Morlet wavelet generating function and constructing a Morlet wavelet function composition; carrying out wavelet frequency division processing by utilizing the optimized Morlet wavelet; calculating a structural tensor of the post-stack seismic data and further constructingnew fault factors based on the structural tensor; under the protection of the structural tensor, carrying out anisotropic diffusion filter on frequency division seismic data; and carrying out waveletinverse transformation on a frequency division filter result and reconstructing to obtain improved signal-to-noise ratio and resolution ratio and protect the seismic data of fault information. The method has the advantages that the resolution ratio of the seismic data is improved while denoising and edge protecting are realized, so that the low-order fault recognition ability of the seismic datais improved.

The invention provides a Retinex underwater image enhancement method based on gray value mapping. The Retinex underwater image enhancement method comprises five processes of color correction, filtering and denoising, establishment of a gray mapping function, searching of optimal parameters and contrast enhancement. The method comprises the following steps: firstly, carrying out color correction ona collected underwater degraded image by adopting a multi-scale Retinex algorithm with a color recovery factor to obtain a first output image; secondly, performing contrast enhancement on the first output image by adopting anisotropic diffusion filtering to obtain a second output image; establishing a gray value mapping function; then, according to the information entropy and the global brightness mean value of the image, combining a simulated annealing algorithm to quickly search for an optimal parameter value, and perfecting a gray value mapping function; and finally, carrying out gray value mapping on the second output image to obtain a final enhanced image. Under the condition of not needing any prior information, underwater image color correction and contrast enhancement are effectively achieved, the image visual effect is improved, and the method can be applied to underwater image preprocessing.

The invention discloses a facial expression recognition method with noise robust. The method comprises steps of improvement on filter pre-treatment and improvement on a feature extraction operator. According to each anisotropic diffusion filtering model of a relative brightness difference adjustment factor, defects that each original anisotropic diffusion filtering method smoothes out details while noise is filtered can be remedied, facial noise and weak detail expression information are distinguished, and integration of expression image information can be ensured; and in addition, the improved HOG operator is adopted to extract expression features, the feature vector dimension is reduced, interference of redundant information is reduced, the algorithm operating time is shortened, classification precision and noise robust are improved, and good application prospect is provided.

An ultrasonic diagnostic apparatus and an ultrasonic image processing apparatus are provided. The ultrasonic diagnostic apparatus comprises: a data generating unit that executes transmission and reception of an ultrasonic wave in a B-mode with respect to a predetermined region of a tested body and generates ultrasonic image data; a decomposition unit that hierarchically performs multi-resolution decomposition of the ultrasonic image data and acquires low-frequency decomposed image data with first to n-th levels; a filtering unit that calculates a filter coefficient using the high-frequency decomposed image data in each layer and uses the filter coefficient to perform nonlinear anisotropic diffusion filtering on the low-frequency decomposed image data or output data from a next lower layer and generate edge information on a signal, for every layer, from the high-frequency decomposed image data; a high-frequency level control unit that controls a signal level of the high-frequency decomposed image data for every layer on the basis of the edge information of each of the layers; and a mixing unit that acquires ultrasonic image data by hierarchically performing multi-resolution mixing of output data of the filtering unit or output data from a next lower layer and output data of the high-frequency level control unit.

An image processing method for conducting anisotropic diffusion filtering on a pixel value Ii,j in a two-dimensional image, comprises: finding conduction coefficients Cn, Cs, Cw, Ce, Cnw, Csw, Cne and Cse in eight surrounding directions for each pixel based on a pixel value gradient ∇I to produce a two-dimensional distribution image of the conduction coefficients; and finding pixel value first partial differentials ∇nI, ∇sI, ∇wI, ∇eI, ∇nwI, ∇swI, ∇neI and ∇seI in the eight surrounding directions for each pixel to calculate a pixel value in an output image according to: Ii,jn+1=⁢Ii,jm+⁢λ[[Cn·∇n⁢I+Cs·∇s⁢I+Cw·∇w⁢I+Ce·∇e⁢I]+⁢12⁡[Cnw·∇nw⁢I+Csw·∇sw⁢I+Cne·∇ne⁢I+Cnw·∇nw⁢I]]i,jn,(n: the number of times of repetition, λ: a constant).