78 results about "Morphological operators" patented technology

The present invention provides a technique for automated selection of a parameterized operator sequence to achieve a pattern classification task. A collection of labeled data patterns is input and statistical descriptions of the inputted labeled data patterns are then derived. Classifier performance for each of a plurality of candidate operator/parameter sequences is determined. The optimal classifier performance among the candidate classifier performances is then identified. Performance metric information, including, for example, the selected operator sequence/parameter combination, will be outputted. The operator sequences selected can be chosen from a default set of operators, or may be a user-defined set. The operator sequences may include any morphological operators, such as, erosion, dilation, closing, opening, close-open, and open-close.

The invention relates to a level-set retinal vessel image segmentation method with fusion shape prior, comprising: (1) utilizing morphological operators and Gaussian convolution to enhance the retinal vessel image; (2) adopting the anisotropy characteristic of Hessian matrix and improving The vascular response function of the retinal vascular image is roughly segmented and used as shape constraints and initialization information; (3) using the shape prior and retinal image data information to construct a model that includes local area energy fitting items, shape constraint items, and level set function regularity maintenance A level set model for retinal vessel segmentation with length penalty term, weighted area constraint term. The segmentation result of the present invention has high accuracy, can replace manual segmentation, can play an important auxiliary role in the diagnosis and treatment of clinically relevant ophthalmic diseases, and has strong clinical application value.

The invention discloses a liver cancer CT image automatic segmentation method based on random forest and fuzzy clustering. The liver cancer CT image automatic segmentation method based on random forest and fuzzy clustering mainly has two processes: an image preprocessing process and a focus region segmentation process, and includes the following steps: in the preprocessing stage, removing the noise of a CT image through curvature filtering and at the same time enabling the edge of the image to be maintained unfuzzy; in the tumour image segmentation stage, training the random forest by extracting the character sample, classifying the pixel points one by one of the image to be segmented to obtain a coarse segmentation result, and then obtaining the final tumour segmentation result through fuzzy clustering and a morphological operator. The liver cancer CT image automatic segmentation method based on random forest and fuzzy clustering does not need adjusting parameters, can segment the liver cancer CT images in different period, has the advantages of being automatic, being high in accuracy and being wide in application, and has a wide application prospect in pre-operation programming and accurate treatment of liver cancer.

Disclosed is a method and system for detecting a structure (e.g., an airway) within an image. The method and system generate a segmented image from the original (reconstructed) image. The method and system then iteratively apply three dimensional morphological operators of varying sizes to the image to obtain a resulting image. The method and system obtain a candidate mask, apply the candidate mask to the resulting image to generate a masked resulting image, and update the segmented image using the masked resulting image to detect the structure. In one embodiment, the segmented image is generated from region growing. Also disclosed is a method to determine different regions of interest within a given segmented image based on computed projections and boundary lines of the segmented image.

The invention aims to provide a two-pass three-dimensional region growing and morphology combination-based lung airway tree segmentation algorithm for solving the problem to realize effective segmentation of a lung airway tree. The two-pass region growing and morphological reconstruction combination-based lung airway tree segmentation method provided by the invention comprises the following steps of automatically obtaining a seed point P1 of three-dimensional region growing from a serial section image; performing first-pass region growing by taking P1 as the seed point to obtain a lung main airway tree; setting a multi-scale morphological operator, and obtaining corresponding airway marking graphs in three anatomic positions of a cross section, a coronal plane and a sagittal plane by utilizing gray-scale morphological reconstruction operation; performing second-pass region growing by taking the lung main airway tree as a basis and the airway marking graphs in the three anatomic positions as limitation conditions to obtain corresponding segmented airway trees; and fusing the segmented airway trees to obtain a final lung airway tree segmentation result F.

A method, medium, and apparatus for eye detection. The apparatus for eye detection may include an image pre-processing unit pre-processing an input face image to divide it into a left image and a right image, an eye candidate extraction unit applying a morphology operator to each of the left and right images and extracting data on first eye candidates for a left eye and data on first eye candidates for a right eye, an eye candidate verification unit verifying the data on the first eye candidates using a binary classifier and extracting data on second eye candidates for each of the left and right eyes, and a final verification unit verifying data on pairs of left and right eyes, which is generated from the data on the second eye candidates for the left and right eyes, using a binary classifier and extracting final eye data.

The present invention discloses a target detection process that acquires imagery from a target; compares the acquired imagery with image metric data; applies criteria to eliminate false detections and reduce clutter; applies morphological operators on the acquired imagery; probability rank orders the target imagery; divides the imagery into a series of detected target windows; and displays the detected target windows. The imagery undergoes light target detection, dark target detection or both. A morphological operator isolates targets rom their background. Two concatenated morphological filter patterns are used to screen imagery data. Spatial discontinuities at the pixel level can be detected. The detected target window images are presented to a user in a mosaic format.

The invention provides a hyperspectral remote sensing image end member extraction method on the basis of a revised extended morphological operator, which is characterized in that an original image is filtered by the opening-and-closing operation and the closing-and-opening operation which are defined by the revised extended morphological operator so as to fulfill the aim of combining spectral information with space information to carry out end member extraction. The method mainly aims to solve the problem of limitation in the extended morphological operator; the revise is carried out by leading in a reference vector; for the provided revised extended morphological operator, the alternately replacing phenomenon is eliminated; the correct replacing direction is ensured; and a separating effect of pure pixels and mixed pixels is reinforced. An experimental result shows that the condition that the method provided by the invention can optimize the end element extraction effect is proved onthe aspects of the spectral curve similarity and a corresponding ground feature distribution map and the like; and the method has a moderate calculated amount and lays a foundation for application ofsubsequent analysis and classification of hyperspectral remote sensing image mixed end members and the like.

The present invention provides a method for star point extracting of a navigational star in cloud environment in daytime, and the method comprises firstly, respectively performing the following operation on each frame of star map of a plurality of frames of daytime cloud star maps which are consecutively taken: (1) using an improved morphological TopHat operator to filter an original start map to obtain a first star map; (2) using a RobinsonGuard filter to filter the first star map to obtain a second star map; (3) using an adaptive threshold method for threshold cutting of the second star map, and storing coordinates and grayscales of all pixels higher than a threshold; and (4) using a clustering method for four-connected domain star point extracting; secondly using a multiframe comparison technique to eliminate pseudo-star points in the connected domain extracted from each frame of star map; and thirdly, calculating the centre of mass of each successfully-compared connected domain to complete the star point extraction. The method can effectively eliminate the interference of daytime sky clouds and other backgrounds, meanwhile avoids mistaken extraction caused by dust and other air floating objects, and ensures the effective and stable star point extraction.

Some embodiments provide techniques and systems to identify locations in a target mask layout for placing assist features. During operation, an embodiment can determine a spatial sampling frequency to sample the target mask layout, wherein sampling the target mask layout at the spatial sampling frequency prevents spatial aliasing in a gradient of a cost function which is used for computing an inverse mask field. Next, the system can generate a grayscale image by sampling the target mask layout at the spatial sampling frequency. The system can then compute the inverse mask field by iteratively modifying the grayscale image. The system can use the gradient of the cost function to guide the iterative modification process. Next, the system can filter the inverse mask field using a morphological operator, and use the filtered inverse mask field to identify assist feature locations in the target mask layout.

The invention discloses a novel noise extraction method based on mathematical morphology. A root mean square error value is utilized to analyze filtering effects achieved after structural elements of different types and scales are used for performing morphological filtering; the corresponding structural element enabling the root mean square error value between a filtered signal to be output and an input noise-containing signal to reach a maximum value is regarded as the optimal structural element for noise extraction; a morphological operator is constructed based on the optimal structural element, and the optimal structural element is used in the operator to perform noise extraction on the noise-containing signal obtained through collection. According to the method, the optimal structural element is found based on the change law of the root mean square error value between the filtered output signal and the input signal obtained by using different structural elements, and meanwhile the morphological operator used for noise extraction is constructed; the method is simple, complicated iterative computation is not needed, random noise can be accurately extracted out of a sampled signal, and support is provided for noise analytical processing in electrical signal collection.

The invention discloses a morphological graphic processing-based point cloud data smoothing method. The method comprises the steps of performing voxelization on point cloud data based on an octree algorithm; traversing and extracting boundary voxels on the basis of the voxelized point cloud data, and selecting a point set of surface and corner points as candidate feature points for smoothing; andopening operation and closing operation are performed on the candidate feature points by using a three-dimensional morphological operator, so that smoothing, denoising and surface repairing of the point cloud data are realized. Smoothing, denoising and surface repairing effects on the point cloud data can be achieved.

Disclosed is an unsteady-state signal detection method based on improved self-adaptive morphological filtering. The method is used for feature extraction of fault signals of various kinds of rotating mechanical equipment, and includes drawing a time domain chart of original signals, searching and marking all local maximums of the signals from the original signals, and calculating a contour area defined by the local maximums and a time axis in the time domain chart to serve as a reference area; performing morphological filtering on the original signals through structure elements with different lengths and a close-open morphological operator, calculating a contour area defined by obtained local extremums of the signals and the time axis after the morphological filtering of the different structure elements, and calculating a difference between the area and the reference area to determine the length of the structure element corresponding to the a minimum difference; and using the obtained structure element for the morphological filtering of the signals, and determining faults according to frequency spectrums of the signals after analysis processing. The unsteady-state signal detection method based on the improved self-adaptive morphological filtering improves efficiency and accuracy of fault determination of rotating equipment.

The invention discloses a method and device for abdomen soft tissue nuclear magnetism image segmentation. The method comprises the steps that an original outline is initialed near an objective outline; a morphological operator is used for carrying out expansion and corrosion operations on the original outline, and a band-shaped closed area is formed in and outside the objective outline to be segmented; KPCA training is conducted on a collected shape template and prior shape information is obtained through a statistics model; the prior shape information is combined with data items of an energy function of a nuclear magnetism image segmentation model, and an energy function is constructed; a kernel Graph cuts algorithm is used for carry out segmentation on the band-shaped closed area and an objective outline is obtained. The method and device for abdomen soft tissue nuclear magnetism image segmentation can achieve semi-automatic segmentation, the device is simple, the robustness of the nuclear magnetism image segmentation algorithm is effectively improved so as to enable the segmentation result to be more accurate, and the method and device for abdomen soft tissue nuclear magnetism image segmentation can be applied to most nuclear magnetism image segmentation.

The present invention provides a visual saliency-based complex environment face detection method. According to the invention, a visual attention mechanism is utilized to obtain a face saliency map; the saliency map is segmented, so that a binary template can be obtained; a morphological operator is utilized to adjust the binary template, so as to make the binary template cover an entire face area; and the template is matched with an original image, so that a face target area can be detected out. According to the visual saliency-based complex environment face detection method, the visual attention mechanism is utilized to obtain the face saliency map; the effective face target area is accurately positioned according to the saliency map; and therefore, the influence of factors such as illumination, poses and occlusion in a complex environment can be eliminated, and manual intervention-free, automatic and accurate detection of a non-restraint face target area can be realized.