992 results about "Automatic segmentation" patented technology

The invention provides a multi-scale feature fusion ultrasonic image semantic segmentation method based on adversarial learning, and the method comprises the following steps: building a multi-scale feature fusion semantic segmentation network model, building an adversarial discrimination network model, carrying out the adversarial training and model parameter learning, and carrying out the automatic segmentation of a breast lesion. The method provided by the invention achieves the prediction of a pixel class through the multi-scale features of input images with different resolutions, improvesthe pixel class label prediction accuracy, employs expanding convolution for replacing partial pooling so as to improve the resolution of a segmented image, enables the segmented image generated by asegmentation network guided by an adversarial discrimination network not to be distinguished from a segmentation label, guarantees the good appearance and spatial continuity of the segmented image, and obtains a more precise high-resolution ultrasonic breast lesion segmented image.

The invention discloses a full text retrieval system based on natural language understanding, comprising: a database server, an information receiving judging module, a natural language processing module, a retrieving module, an indexing module, an index database and a result set processing module. The system of the invention provides two resolution strategies, that is, word classification static with semantic analysis associated with automatic segmentation and expanding inquired word static according to Hownet rule for low intelligence situation of current search engine. The deployed system converts information retrieval from current key word-based layer to knowledge (or concept)-based layer; the invention is capable of using techniques such as word classification, synonym, concept search, phrase identification, etc. with understanding and processing ability to knowledge. The search engine is provided with intelligence and humanization of information service. The user is allowed using natural language for information retrieval. The invention is capable of adding user selection behavior in interactive operation mode, so as to provide more convenient, more precise search service.

The invention belongs to the technical field of image processing and computer vision, and relates to a novel biomedical image automatic segmentation method based on a U-net network structure, including dividing a biomedical data set into a training set and a test set, and normalizing the test set and augmented test set; inputting the images of the training set into the improved U-net network model, and generating a classification probability map by output image passing through a softmax layer; calculating the error between classification probability diagram and gold standard by a centralized loss function, and obtaining the weight parameters of network model by a gradient backpropagation method; entering the images in the test set into the improved U-net network model, and outputting the image to generate a classification probability map through the softmax layer; according to the class probability in the classification probability graph, obtaining the segmentation result graph of theimage. The invention solves the problems that simple samples in the image segmentation process contribute too much to the loss function to learn difficult samples well.

A method for automatic left ventricle segmentation of cine short-axis magnetic resonance (MR) images that does not require manually drawn initial contours, trained statistical shape models, or gray-level appearance models is provided. More specifically, the method employs a roundness metric to automatically locate the left ventricle. Epicardial contour segmentation is simplified by mapping the pixels from Cartesian to approximately polar coordinates. Furthermore, region growing is utilized by distributing seed points around the endocardial contour to find the LV myocardium and, thus, the epicardial contour. This is a robust technique for images where the epicardial edge has poor contrast. A fast Fourier transform (FFT) is utilized to smooth both the determined endocardial and epicardial contours. In addition to determining endocardial and epicardial contours, the method also determines the contours of papillary muscles and trabeculations.

The invention relates to auxiliary medical diagnoses, and aims to provide a method for automatically identifying whether a thyroid nodule is benign or malignant based on a deep convolutional neural network. The method for automatically identifying whether the thyroid nodule is benign or malignant based on the deep convolutional neural network comprises the following steps: reading B ultrasonic data of thyroid nodules; performing preprocessing for thyroid nodule images; selecting images, and obtaining nodule portions and non-nodule portions through segmentations; averagely dividing the extracted ROIs (regions of interest) into p groups, extracting characteristics of the ROIs by utilizing a CNN (convolutional neural network), and performing uniformization; taking p-1 groups of data as a training set, taking the remaining one group to make a test, and obtaining an identification model through training to make the test; and repeating cross validation for p times, and then obtaining an optimum parameter of the identification model. The method can obtain the thyroid nodules through the automatic segmentations by means of the deep convolutional neural network, and makes up for the deficiency that a weak boundary problem cannot be solved based on a movable contour and the like; and the method can automatically lean and extract valuable feature combinations, and prevent the complexity of an artificial feature selection.

The invention provides an X-ray film bone age prediction method and system based on deep learning. According to the invention, a bone age prediction result is finally formed through performing preprocessing on an X-ray image of a teenager, automatic segmentation of a hand bone area and bone age prediction. The method comprises a hand bone X-ray film preprocessing and sample data enhancement method, a hand bone X-ray film image block sampling method, a hand bone automatic segmentation algorithm and a transfer learning based hand bone X-ray film bone age assessment algorithm, and finally designsa hand bone X-ray film bone age prediction system taking Dicom data as input. Users only need to select a hand bone X-ray film with the bone age to be predicted, the segmentation and prediction process is completely automatic, and doctors are not required to perform region marking or selecting, and a powerful tool is provided for bone age assessment in scientific research and clinical practice.

The invention discloses a nasopharyngeal-carcinoma (NPC) lesion automatic-segmentation method and nasopharyngeal-carcinoma lesion automatic-segmentation systems based on deep learning. The method comprises: carrying out registration on a PET (Positron Emission Tomography) image and a CT (Computed Tomography) image of nasopharyngeal carcinoma to obtain a PET image and a CT image after registration;and inputting the PET image and the CT image after registration into a convolutional neural network to carry out feature representation and scores map reconstruction to obtain a nasopharyngeal-carcinoma lesion segmentation result graph. The method carries out registration on the PET image and the CT image of the nasopharyngeal carcinoma, obtains a nasopharyngeal-carcinoma lesion by automatic segmentation through the convolutional neural network, and is more objective and accurate as compared with manual segmentation manners of doctors; and the convolutional neural network in deep learning isadopted, consistency is better, feature learning ability is higher, the problems of dimension disasters, easy falling into a local optimum and the like are solved, lesion segmentation can be carried out on multi-modal images of the PET-CT images, and an application range is wider. The method can be widely applied to the field of medical image processing.

The invention discloses a CT image kidney segmentation algorithm based on a residual double-attention deep network. According to the method, the advantage that the residual error unit can repeatedly utilize the features and the excellent feature learning capability of the double attention mechanism are combined; a residual double attention module is designed; and a residual double attention moduleis used as a basic module to construct a U-shaped deep network segmentation model, and a loss function for segmentation is designed at the same time, so that the U-shaped deep network segmentation model can pay more attention to kidney region features, can effectively cope with shape changes of the kidney with cystic lesions, and can maintain robustness for the shape changes of the kidney with cystic lesions. Therefore, the boundary of the kidney area is accurately positioned, and automatic segmentation of the kidney area in the CT image is achieved, and a good segmentation effect is achieved.

The invention discloses an image segmentation method for a dynamic contrast-enhanced mammary gland MRI sequence, pertaining to the field of magnetic resonance image processing techniques, which is characterized by comprising the following steps: a three-dimensional magnetic resonance image sequence of the section of the mammary gland is put into a computer; the image is divided into two parts including a mammary gland-air interface and a mammary gland-chest interface; a breast-air boundary is obtained by a splitting transaction in which a dynamic threshold controls the regional growth; an initial profile of the mammary gland and the chest is obtained in the same way, the complex profile of the breast and the chest is obtained with a method of controlling a level set; a three-dimensional magnetic resonance image sequence of a point-in-time is obtained by split jointing the segmentation results and taken as an initial position of the next group three-dimensional image segmentation. The image segmentation method of the invention increases the segmentation speed, solves the problem that a level set algorithm can not easily determine the initial profile and the velocity function and realizes an automatic segmentation of the complex dynamic contrast-enhanced mammary-gland magnetic resonance image with plenty of data.

The invention discloses an automatic segmentation method for a point cloud of a facade of a large scene city building. The method comprises steps of (1) performing fusion registration on airborne LiDAR point cloud data and on-vehicle LiDAR point cloud data, (2) extracting airborn LiDAR building point cloud data from the airborne LiDAR point cloud data which goes through registration in the step (1), (3) performing segmentation on the point cloud data of a single building based on the airborne LiDAR point cloud data extracted from the step (2), (4) performing contour tracking on the single building segmented by the step (3), (5) performing simplification and normalization processing on a contour line extracted from the step (4), (6) performing rough segmentation on the point cloud of the facade of the building based on the contour line which goes through the simplification and normalization processing in the step (5), and (7) performing fine segmentation on the building facade point cloud which is roughly segmented in the step (6). The automatic segmentation method of the invention can fast and accurately separate the building facade point cloud from the on-vehicle LiDAR point cloud.

The invention discloses a retina eyeground image segmentation method based on a depth full convolutional neural network. The retina eyeground image segmentation method includes the following steps of:selecting a training set and a test set, extracting retina eyeground images to obtain optic disk positioning area images, and performing blood vessel removal operation on the optic disk positioning area images; constructing the depth full convolutional neural network, taking the optic disk positioning area images as the input of the depth full convolutional neural network, and performing the training of an optic disk segmentation model on the training set based on trained weight parameters as initial values to fine tune model parameters, and performing fine tuning on parameters of an optic cup segmentation model based on trained optic disk segmentation model parameters; and performing optic cup and optic disk segmentation on the test set by utilizing a trained optic cup segmentation model, performing ellipse fitting on final segmentation results, calculating a vertical cup-disk ratio according to optic cup and optic disk segmentation boundaries, and taking a cup-disk ratio result as important basis for a glaucoma auxiliary diagnosis. The retina eyeground image segmentation method achieves optic disk and optic cup automatic segmentation of the retina eyeground images, has high precision and fast speed.

The “repeating object extractor” described herein automatically segments and identifies audio objects such as songs, station identifiers, jingles, advertisements, emergency broadcast signals, etc., in an audio stream. The repeating object extractor performs a computationally efficient joint segmentation and identification of the stream even in an environment where endpoints of audio objects are unknown or corrupted by voiceovers, cross-fading, or other noise. Parametric information is computed for a portion of the audio stream, followed by an initial comparison pass against a database of known audio objects to identify known audio objects which represent potential matches to the portion of the audio stream. A detailed comparison between the potential matches and the portion of the audio stream is then performed to confirm actual matches for segmenting and identifying the audio objects within the stream, an alignment between matches is then used to determine extents of audio objects in the audio stream.

The present invention provides a dermoscopy image automatic segmentation method based on a full convolutional neural network. The method comprises the following four steps: 1: obtaining of dermoscopy images and true value graphs; 2: performing structure design of a full convolutional neural network; 3: performing design of feature fusion and a per-pixel segmentation method; and 4: performing network training and segmentation. According to the steps mentioned above, an end-to-end depth convolutional neural network is obtained through training to perform accurate segmentation of the dermoscopy images and allow a small area skin lesion area to be effective so as to solve actual problems that the skin lesion area segmentation is not good to influence the subsequent diagnosis accuracy in the dermatology computer auxiliary diagnosis system.

The invention relates to a deep convolutional neural network-based traditional Chinese medicine tongue image automatic segmentation method and belongs to the computer vision field and traditional Chinese medicine tongue diagnosis field. According to the method of the invention, a convolutional neural network structure is designed; collected sample data are adopted to train the network, so that a network model can be obtained; and the model is adopted to automatically segment a traditional Chinese medicine tongue image. The method includes an offline training phase and an online segmentation phase. The method can be applied to both closed type and open tongue image acquisition environments and can effectively improve the accuracy and robustness of the automatic segmentation of the traditional Chinese medicine tongue image. The method of the present invention specifically relates to deep learning, semantic segmentation, image processing and other technologies.

Medical images are automatically segmented by customizing the morphological segmentation of features identified in the image based upon statistical analysis of the features within each region to be analyzed. The statistical description of the features, as reported through a feature vector, informs the system as to which input variables to select for further segmentation analysis for features residing within the region of the image analyzed. By customizing the automatic segmentation analysis to produce an enhanced image, features within the image are characterized more efficiently and precisely. False positive identification of lesions are minimized without sacrifice of true positive identifications.

Content items may be segmented and labeled by topic to provide for the capture, analysis, indexing, retrieval and/or distribution of information within information rich media, such as audio or video, with greater functionality, accuracy and speed. The segments and other related information may be stored in a database and made accessible to users through, for example, a search service and/or an on-demand service. Automatic segmentation may include receiving a text representation, calculating relevance intervals based on the text representation, determining a nodal representation based on the relevance intervals, and determining segments of the content item based on the nodal representation.

The invention relates to a large-length voice full-automatic segmentation method which is a zero-labeling sentence automatic segmentation algorithm having higher accuracy. The algorithm enables a Force-alignment non-supervision algorithm and a semi-supervised learning method based on an HMM to be blended, automatic expansion is carried out on a few precise labeling sets provided for the zero-labeling sentence segmentation algorithm by a semi-supervised learning minimization labeling sentence segmentation algorithm through establishment of an iteration mechanism based on a timer shaft, the purpose of the maximization of the precise labeling sets is realized, and then according to obtained correct periods, voice of an original length is cut into smaller paragraphs or sets of sentences. According to the method, a Force-aligned method under the HMM and a Co_training method in semi-supervised learning are blended together, so the facts that in a large-length voice sentence segmentation process, manual intervention is not needed, and segmentation accuracy is high are guaranteed. The large-length voice full-automatic segmentation method can be applied to rapid and automatic construction of a voice corpus.

The invention provides a three-dimensional MRI image-based breast tumor automatic segmentation method. The method comprises the following steps of performing image preprocessing: providing an initial MRI image and performing preprocessing on the initial MRI image by adopting a non local mean filter; performing breast tumor locating: building a multilayer processing model for a training set, performing hierarchical abstraction on characteristics of a training object by adopting a convolutional neural network, automatically extracting segmentation characteristics, and outputting a probability distribution graph of a tumor position; performing breast tumor boundary segmentation: providing a breast three-dimensional MRI image, determining a seed point based on the probability distribution graph of the tumor position, and finishing segmentation initialization to obtain an initial region C0 of a tumor; and performing accurate segmentation on the tumor by using a region growth algorithm.

An image processing method includes a segmentation step that segments an input image into a plurality of regions by using an automatic segmentation algorithm, and a computation step that calculates a saliency value of one region of the plurality of segmented regions by using a weighted sum of color differences between the one region and all other regions. Accordingly, it is possible to automatically analyze visual saliency regions in an image, and a result of analysis can be used in application areas including significant object segmentation, object recognition, adaptive image compression, content-aware image resizing, and image retrieval.

The invention discloses an automatic segmentation method for a lung-area CT sequence. The method is characterized by comprising the steps of 1) an image I of the lung-area CT sequence is input; 2) the image I is segmented via interactive region growth; 3) according to a segmentation result, an initial contour is obtained, and seed-point coordinates of adjacent images are calculated; 4) based on the seed-point coordinates of the image I, a present image II in the sequence is segmented via the interactive region growth in step 2; and 5) step 2, 3, 4 are repeated to determine whether all images in the CT sequence are segmented, and if no, the step 3 is turned to. The automatic segmentation method automatically calculates mapping of the seed-point areas by combining the image context object characteristic continuity of the sequence to realize sequence segmentation, thereby obtaining complete three-dimensional area data of the lung, and providing basis for VOI extraction and classification of a suspected lung tubercle for a CAD system.