63 results about "Level set segmentation" patented technology

A dynamic thresholding level set method combines two optimization processes, i.e., a level set segmentation and an optimal threshold calculation in a local histogram, into one process that involves a structure called a “propagating shell.” The propagating shell is a mobile 3-dimensional shell structure with a thickness that encompasses the boundary of an object, the boundary between two objects or the boundary between an object and a background. Because the local optimal threshold tends to shift to a value of a small region in a histogram, the shift can drive the propagating shell to an object boundary by pushing or pulling the propagating shell. The segmentation process is an optimizing process to find a balanced histogram with minimal threshold shift. When the histogram in the propagating shell is balanced, the optimal threshold becomes stable, and the propagating shell reaches a convergence location, i.e., an object boundary. This method can be applied to computer-aided organ and tumor volumetrics.

The invention provides an optical image and SAR image automatic registration method within multilevel multi-feature constraint. The optical image and SAR image automatic registration method within the multilevel multi-feature constraint comprises the following steps that optical images and SAR images are preprocessed, multi-scale level set segmentation is conducted, and a plane segmentation result is obtained; when similar plane targets exist, a coordinate set of centroid points of areas seemingly provided with the same name is calculated; when similar plane targets do not exist, multi-scale analysis is conducted on the images by means of wavelet transformation, extraction of lower-layer linear characteristics and point set matching are conducted on the thickest image, and lower-layer registration transformation parameters are extracted; high-layer linear characteristics is extracted, a control-point matching degree function is defined according to the lower-layer transformation parameters, and high-layer point set matching is conduced; finally, a matched point pair is precisely judged out through the KNN image from the structure, a wrong matched point pair is eliminated, transformation parameters of the matched point pair are obtained according to a polynomial transformational model, and a final registration result is obtained.

A method for segmentation of an image interactively with a user utilizes level set segmentation and includes selecting by user input respective areas of object and of background; initializing an embedding function implementing a segmentation boundary according to the selecting; computing intensity distributions and for the respective areas of object and of background; and performing repeatedly the steps below until convergence is reached: (a) evolving the embedding function, (b) recomputing the intensity distributions, and (c) checking for new user input and, if so: (d) updating labeling of the areas of object and background.

Provided is an SAR image road damaging information extracting method based on vector data. According to the range of SAR images after a disaster, vector data of a corresponding area are obtained; the vector data are projected to a coordinate system of the SAR images and then are registered on the SAR images; a suspected road damaging area of the SAR images is extracted, wherein a road detecting operator is used for line detecting, road width information and the road vector data are subjected to shape level set segmentation, and the road damaging area is obtained by fusion; and a Bayes network model is established to carry out further judging on the suspected road damaging area, and road damaging information is extracted. According to the method, the vector data are used as prior information for helping SAR image road changing detecting, detecting rate is high, a breaking zone can be well extracted through a method with line detecting and a shape level set being combined, the omission factor is low, interference information can be effectively removed through the established Bayes network model, and false alarm in road damaging extracting is lowered.

The invention discloses a semi-automatic medical sequence image segmentation method and system. The segmentation method includes the steps of: selecting an initial slice image containing a target to be segmented in the medical sequence image; generating the initial slice image according to user instructions. Target area: using the target area as the initial input image of the adjacent slice image, using the level set algorithm to iteratively segment the subsequent slice image until the subsequent slice image does not include the target to be segmented, and then obtain the segmentation result of the target area to be segmented . Thereby, the present invention can quickly and accurately segment the target region to be segmented in the medical sequence images.

The invention discloses a method for carrying out change detection on remote sensing images based on treelet fusion and level set segmentation, and mainly solves the problem that much pseudo-change information exists in the existing change detection methods. The method is implemented through the following steps: inputting two time-phase remote sensing images, then respectively carrying out mean shift filtering on each image so as to obtain two time-phase filtered images; respectively carrying out two-dimensional stationary wavelet decomposition on the two time-phase filtered images three times under different level numbers; carrying out subtraction on wavelet coefficient matrixes of corresponding directional son-bands of the filtered images with the same decomposition level number; carrying out enhancement and two-dimensional wavelet inverse transformation reconstruction on wavelet coefficient difference matrixes in horizontal and vertical directions by using a sobel operator; and fusing the reconstruction images with different decomposition level numbers so as to obtain a final difference map by using a treelet algorithm, then carrying out level set segmentation on the differencemap so as to obtain a change detection result. By using the method disclosed by the invention, the accuracy of the change detection result can be improved effectively, and the edge feature of a change area can be maintained better, therefore, the method can be applied to the fields of natural disaster analysis, land resource monitoring, and the like.

Methods and systems for image segmentation are disclosed. A nonlinear statistical shape model of an image is integrated with a non-parametric intensity model to estimate characteristics of an image and create segmentations of an image based on Bayesian inference from characteristics of prior learned images based on the same models

Disclosed are a method and a device for sea surface oil spillage detection based on an SAR (synthetic aperture radar) image. The method includes: converting the SAR image into a binary image by performing threshold segmentation for gray of pixels in the sea surface SAR image; for a neighborhood in predetermined size of each pixel in the binary image, determining whether number and / or ratio of the pixels with the pixel value of 1 in the neighborhood is larger than a preset value or not; identifying the neighborhoods that whether the number and / or the ratio of the pixels with the pixel value of 1 is larger than the preset value or not; and using an image boundary formed by all the identified neighborhoods as an initial zero level set used for detecting level set segmentation so as to detect sea surface oil spillage.

A computer implemented method for performing bone segmentation in imaging data of a section of a body structure is provided. The method includes: Obtaining the imaging data including a plurality of 2D images of the section of the body structure; and performing a multiphase local-based hybrid level set segmentation on at least a subset of the plurality of 2D images by minimizing an energy functional including a local-based edge term and a local-based region term computed locally inside a local neighborhood centered at each pixel of each one of the 2D images on which the multiphase local-based hybrid level set segmentation is performed, the local neighborhood being defined by a Gaussian kernel whose size is determined by a scale parameter (σ).

The invention discloses a liver image segmentation method based on a dense feature pyramid network, and relates to the technical field of image processing. The method adopts a dense feature pyramid network to process a segmentation problem of a multi-stage liver CT image, the network introduces a feature pyramid based on a full-volume integral segmentation network, uses dense connection to enhancea feature flow, and realizes pixel-level liver segmentation on multi-stage CT; According to the method, the Dice value in a public 3DIRCADb database is 95.0%, so that the segmentation performance isimproved; A common data set and a clinical data set prove that the DPFN trained by using the CT image with the larger layer thickness can be seamlessly popularized to the CT image with the smaller layer thickness.

The invention provides an image level set segmentation method based on local gray clustering characteristics. The method comprises the steps that images to be segmented are read; linear weighting and fitting bias fields of orthogonal basis functions are used, and the weight value of each basis function is initialized; the level set function set of the images is initialized; the energy functional of image level set segmentation is established, and level set segmentation control parameters are set according to the images to be segmented; a clustering center set, the image level set function set and basis function weight column vectors are respectively updated until meeting the stop criterion for iteration so that the energy functional of iteration is obtained; the subordinating degree function of the images, i.e. the segmentation result of the images to be segmented, is constructed according to the currently updated image level set function set, and bias field estimation of the images to be segmented is obtained according to the updated basis function weight column vectors and basis function column vectors. According to the method, the adverse impacts of weak boundary, image noise and gray inconsistency on the accuracy of image segmentation can be overcome by the method so that the method has the effect of image gray correction.

The invention relates to medical apparatuses and medical images and provides an improved hybrid multi-mode brain tumour image segmentation method. A brain tumour area is extracted through employing an FFCM algorithm, a hybrid level set algorithm is utilized to correct a boundary problem existing in the tumour area. The FFCM algorithm and the level set algorithm can be more effectively applied to MRI brain tumour images. The method is characterized by comprising steps that firstly, the MRI images in three modes including T1C, T2 and FLAIR are inputted, median filtering is employed to carry out filtering and initial segmentation of the images to acquire pre-processed images, then linear fusion is employed, lastly, FFCM clustering segmentation is carried out for the fused images, areas with relatively large gray values are automatically extracted, and hybrid level set segmentation is carried out for the acquired tumour under-segmentation areas. The method is mainly applied to medical image acquisition and processing.

The invention discloses a depth map assisted active contour image matting method and system, and the method comprises the following steps: obtaining a depth map of a color target image, carrying out the processing of the depth map through employing a repair algorithm, and obtaining a repaired depth map; carrying out confidence calculation on the repaired depth map and the target image to obtain adepth confidence map and a color confidence map; carrying out level set segmentation by utilizing the depth confidence map and the color confidence map, obtaining a final target contour, and obtaininga required tripartite graph; and respectively obtaining a matting result of the color image and a matting result obtained by the depth image, setting a color difference degree to measure a distance value between the foreground and the background, when the distance value is smaller than a given threshold T, adopting the color matting result, and otherwise, adopting the depth matting result. The method can better solve the problem of background similarity before and after the image, and is closer to the boundary of the to-be-segmented object in the segmentation process, so that the geometric topology of the object is better maintained, and the required binary segmentation result is obtained.

Described is a system for localizing and controlling a tool tip with a robotic arm. The system receives three-dimensional (3D) visual information of a scene having a tool with a tool tip proximate a target. A depth disparity map is generated based on the 3D visual information. The depth disparity map is filtered by depth continuity to generate boundaries of the tool. The boundaries of the tool are initially loose to ensure that the tool is contained within the boundaries. Thereafter, using level set segmentation, the boundaries of the tool are shrunk and extracted. Ends of the tool tip are identified based on the extracted boundaries, which are then used to generate a pose of the tool tip.

A process includes: generation of knowledge of the anatomical object to be separated from neighboring objects and identification of high level features (e.g., prior knowledge that includes: a mean shape template of the vertebra; and, high level features, e.g., readily identifiable boundary regions between the vertebrae neighboring objects, such as neighboring vertebra and rib structures); and the subsequent use of this prior knowledge to level set.

The invention discloses a water precise extraction method through polarized SAR image segmentation based on a level set, and the method comprises the steps: S1, carrying out the preprocessing of an SAR image; S2, extracting a weak scattering homogeneous region from a polarized SAR image based on the polarization total power of the polarized SAR image, and completing the preliminary extraction of water; S3, precisely extracting water from the weak scattering homogeneous region based on the scattering entropy and a covariance matrix of the polarized SAR image. The method gives full consideration to the similarity of weak scattering ground objects in the polarized SAR image, employs a multi-scale level set segmentation strategy, constructs different SAR models according to different segmentation contents, defines energy functions at different layers, and improves the water extraction timeliness and precision through a stepwise refinement method.

The invention belongs to the technical field of medical imaging, and particularly relates to a cardiac magnetic resonance image segmentation method. A cardiac anatomic structure-based left and right ventricle level set segmentation method specifically comprises the following steps: performing geometry decomposition to a cardiac structure; resolving a convex hull of an initialized contour; producing a signaled distance function; and generating intima and extima through a dual-level set algorithm, and the like. By using the method, the simultaneous segmentation of the ventricular intima and extima and cardiac muscle can be realized, and the segmentation results can meet the physiological structure characteristics of a heart as far as possible.

The invention discloses an automatic level set segmentation method and device for a blood vessel image. The method comprises the steps that an initial blood vessel model is acquired, the initial bloodvessel model is determined based on a center line and a radius function of a blood vessel corresponding to the initial blood vessel model, and the radius function is used for describing a blood vessel area; and on the basis of the local lumen statistical model, level set segmentation is carried out on the initial blood vessel model to obtain a segmented blood vessel model, the local lumen statistical model comprises a lumen threshold of a cross section of a blood vessel along a central line. The technical problem that in the prior art, a traditional global threshold level set cannot achieve accurate modeling of coronary artery blood vessels is solved.

The invention discloses a wavelet-decomposition-based SAR image change detecting algorithm of a multi-scale level set, and belongs to the field of remote sensing image processing. The wavelet-decomposition-based SAR image change detecting algorithm mainly solves the problem that a speckle noise effect is serious in an SAR image change detecting process. The wavelet-decomposition-based SAR image change detecting algorithm is implemented in the following steps that (1) a difference image of two registered SAR images in the same region and at different time phases is obtained by adopting logarithm ratio operators; (2) multilayer wavelet decomposition is conducted on the difference image through SWT so as to obtain images with different resolution ratios; (3) the images with the low resolution ratios are preliminarily segmented through a level set algorithm, and the outlines of obtained segmented images are used as initialization curves of the level set algorithm of the images with the higher resolution ratios; (4) the step (3) is repeated in a layer-by-layer mode until final segmented images are obtained by conducting level set segmentation on the images with the high resolution ratios. According to the wavelet-decomposition-based SAR image change detecting algorithm, the multi-scale application overcomes the defect that a closed curve is prone to getting into local optimum in a level set evolutionary process, and robustness on noise is enhanced; the wavelet-decomposition-based SAR image change detecting algorithm is applied to change detecting of the SAR images, the detecting effect and detecting accuracy are improved obviously, and the change detecting process is accelerated.

Methods and systems for image segmentation are disclosed. A nonlinear statistical shape model of an image is integrated with a non-parametric intensity model to estimate characteristics of an image and create segmentations of an image based on Bayesian inference from characteristics of prior learned images based on the same models.

The invention, which belongs to the field of digital surface model (DSM) data segmentation processing by applying a level set algorithm, relates to a level set-based method for constructing a level of detail 2 (LOD2) building model, so that a problem that the construction precision is not high due to a rough top surface structure in the existing two-dimensional image-based building model construction method can be solved. More specifically, the invention comprises the following steps: extracting a building outline mask omega m, selecting DSM data and distributing the building outline mask omega m and the DSM data into a unified coordinate system; obtaining building top surface data T; obtaining a characteristic space of the building top surface data T; carrying out multi-phase level set segmentation to obtain sub-areas; obtaining point sets of all the sub-areas, detecting a boundary point of each fragment and obtaining an image coordinate of an angular point of each primitive of the building; establishing a topological structure of the building top surface data T; and according to an aerial visible image, extracting texture data of the building surface and enabling the data to correspond to different primitives of the building, so that construction of the LOD2 building model is completed. The provided method is applied to a three-dimensional construction task of a large building with an LOD2 level.

The invention discloses an inner-ear three-dimensional level set segmentation method based on a statistical shape model. The method comprises the following steps of step1, establishing the statisticalshape model of an inner ear; step2, through a rigid registration method from volume data to volume data, acquiring an area of interest of the inner ear, and then through the rigid registration methodfrom the model to the volume data, registering an average shape model of the statistical shape model to the area of interest of the inner ear, and acquiring an initial surface contour of the inner ear; and step3, using the three-dimensional level set segmentation method based on a threshold area, carrying out level set evolution and acquiring a target contour surface, and then combining originalbrain MRI volume data so as to finally calculate and acquire a needed inner ear contour. In the invention, a real inner ear boundary can be rapidly converged; a complexity of the method is lower thana method for acquiring an initial contour through combination of various pre-segmentation methods; and the method possesses high robustness and accuracy during a inner ear segmentation process.

The present invention discloses an automatic segmentation method of brain gray matter nuclei in magnetic resonance imaging. The method comprises that by using standard reference images and a matched image set, seed spots located in each gray matter nucleus can be obtained through calculation; and a to-be-segmented image is preprocessed by using unconnectedness between the seed spots, fuzzy regions between adjacent nuclei are eliminated, and the segmentation contours of the target nuclei are automatically obtained by combining with the level set segmentation method. According to the method disclosed by the present invention, the variability of the morphology of the nuclei in different images can be more effectively coped with, and the segmentation results are more accurate; and by using the method disclosed by the present invention, specific nuclei such as the substantia nigra, the red nucleus and the like in the brain can be precisely segmented, diagnosis and pathological study on diseases such as Parkinson disease can be facilitated to a great extent.

A method for segmentation of an image interactively with a user utilizes level set segmentation and includes selecting by user input respective areas of object and of background; initializing an embedding function implementing a segmentation boundary according to the selecting; computing intensity distributions and for the respective areas of object and of background; and performing repeatedly the steps below until convergence is reached: (a) evolving the embedding function, (b) recomputing the intensity distributions, and (c) checking for new user input and, if so: (d) updating labeling of the areas of object and background.

The invention provides a CBCT tooth image segmentation method based on center point detection. The CBCT tooth image segmentation method comprises the steps of preprocessing; roughly segmenting; carrying out double-level-set fine segmentation; performing inter-layer iterative segmentation; and obtaining a tooth three-dimensional structure. According to the tooth image segmentation algorithm based on the center point detection, the detected center point information is fully utilized to perform the coarse segmentation of an initial layer, and a coarse segmentation result is utilized to replace the manual initialization, so that the full-automatic image segmentation under the guidance of experts is realized. According to the invention, the detected central point information is considered to beused as the prior information of a segmentation algorithm, the tooth image segmentation algorithm based on the central point detection is provided, and a double-level set segmentation algorithm and threshold optimization constraint processing are introduced in the segmentation process, so that the tooth three-dimensional structure can be obtained more quickly, more conveniently and more accurately, and the segmentation is more accurate and better in robustness.

The invention discloses a text extraction method based on level set segmentation. The method includes reading image data information and determining a border curve; performing graying on the read image; extracting graying feature values; dividing the image into two areas by adopting a level set function according to the graying feature value; performing binaryzation on the two areas obtained through division; performing connection element calibration on the two areas subjected to binaryzation; filtering connection elements calibrated in the two areas; performing polarity judgment on the areas subjected to filtering and determining a text pixel area and a background pixel area; filtering the text area and performing filtering for background noise removal; and outputting a text extraction result. By adopting the method provided by the invention, text information in a complex background can be extracted. Besides, extraction of image text containing outline letters is also very accurate. The method has certain universality and practicability.

The invention provides a multi-scale blood vessel enhanced level set segmentation system and method, and relates to the technical field of image processing. The system comprises an input module, a preprocessing module, a blood vessel enhancement module, a level set module and an output module. According to the invention, a to-be-segmented blood vessel and a background area often slightly differentin the grayscale, the phenomenon of uneven gray level distribution is avoided, influence of unobvious gray level difference between the blood vessels in the blood vessel image and the background on the segmentation precision is overcome, the segmentation capability of the level set segmentation method on the blood vessels with different thicknesses is improved under the multi-scale blood vessel enhancement, and the precise segmentation of the micro blood vessels is realized.

The invention discloses a quantitative detection method for surface defects of a large-aperture telescope lens, which comprises the following steps of: firstly, adding a weight item, a regular item and an inter-class dispersion penalty item in a clustering objective function to optimize the clustering objective function, classifying an optical defect image by adopting an iteration method, and enhancing a defect contour while removing noise; evolving the classified image by a geodesic line active contour model through a level set function, and taking a zero level set to segment a defect image; and finally, quantifying the extracted defect image, and marking the defect contour line of the defect image by adopting a binary chain code technology so as to quantify the area, the gravity center, the long diameter, the short diameter and the perimeter of the defect region. According to the method, iteration of the optimized clustering objective function, defect image segmentation of the active contour model and defect quantification are combined, so that the damage features of the defects on the surface of the satellite telescope lens are enhanced, the contour feature information of the defects is shown, and quantitative analysis of the defects is completed while the detection precision is improved.

The invention provides an effective method for segmenting a solar photovoltaic panel picture under the condition that the image contrast is too low, which mainly comprises the following parts: the target picture is equalized by a histogram to obtain an enhanced gray scale picture and a v-channel picture converted from RGB to HSV is obtained; the obtained histogram equalization picture and the v-channel picture are segmented by using a level set segmentation method respectively. In order to improve the level set operation efficiency, the invention adds an edge stop function, so that the evolution can be stopped quickly in the region with large gray value change, and the level set formula can be rapidly evolved in the region without boundary. The level set segmentation results of the two images are calculated by local entropy operation to obtain the local entropy of the image. The local entropy of the two images is superposed in different ratios and binarized according to the entropy value. After the binary image is processed to eliminate noise, the whole board is obtained and whether the board is qualified or not is judged. Finally, the binary image is used as a mask and the original image is copied to get the final segmentation result.