47 results about "Form classification" patented technology

The invention relates to an image processing method, in particular to a detecting and classifying method of facial-form contours. The invention aims at solving the problems in the prior art that the accurate facial-form contour can not detected and the techniques can not be really applied to the facial-form classification. The invention provides the technical proposal that: the step 1, the human face is positioned; the step 2, the facial contour is detected: an egg-shaped face is set into a vector closed and symmetrical curve Lm that consists of one or a plurality of points, and then, the forehead height, the chin height, the abscissa position that keeps to the sides of the left cheek and the right cheek extremely and the gill shape are modified; the step 3, the facial forms are classified. The invention can provide the effective and automatic image processing measures for a plurality of industries such as image design, facial make-up, hairdressing, jewelry, face-painting, entertainment, and the like, and provides the automated and objective technical support for a plurality of applied industries.

A system and method to classify forms. An image representing a form of an unknown document type is received. The image includes line-art. Further, a plurality of template models corresponding to a plurality of different document types is received. The plurality of different document types is intended to include the correct document type of the unknown document. A subset of the plurality of template models are selected as candidate template models. The candidate template models include line-art junctions best matching line-art junctions of the received image. One of the candidate template models is selected as a best candidate template model. The best candidate template model includes horizontal and vertical lines best matching horizontal and vertical lines of the received image, respectively, aligned to the best candidate template model.

A searching method of Internet includes carrying out tree form classification on Internet information to set up relation table; setting attribute of key word and setting up key word bank; setting up column for information and setting up webpage mapping base to obtain webpage content; confirming classification attribution of each information according to title, content, key word bank and said mapping base; adding new result into result databank if searched result is new then demonstrating result of user's search.

A pulsed radar system uses simultaneous multiple frequency spectral response of life form targets to classify target type and movements. In particular classifying movements from human and animals such as walking and breathing.

Currently radar systems attempt to image targets with ultra high bandwidth (fine range resolution) and a large or synthetic aperture (SAR). These techniques slice up the scene into a three-dimensional radar picture. The radar picture data is again processed with target classifiers. Currently these classical techniques are beyond state-of-the-art at long standoff ranges.

This new radar system will use multiple sparse lower bandwidth frequencies to provide life form classification imaging based on the spectral response. The lower bandwidth dramatically improves the data quality while the spectral ratios and correlations will resolve sparse frequency ambiguities for a new type of imaging radar.

The invention discloses an automatic measuring and morphological classification method for particles precipitated from steel, comprising the steps as follows: firstly, the electron micrographs of the target particles precipitated from steel are subjected to image binary segmentation so as to obtain the binary images of the particles; the binary images of the target particles are denoised by a morphological filtering method, a seed filling method is adopted to fill holes, and the particles to be separated are determined by the domain value determined by experience criterion and the separation of agglomerate particles is carried out; the particles after separation are subjected to region labeling; finally, the neural network morphological classification models of the target particles precipitated from steel are established; and results are displayed and output in the form of graph files. The method can obtain ideal measuring and classification effect without omission inspection and re-inspection; the measurement accuracy of particle size can reach plus or minus 2 microns, the particle size distribution anastomotic rate can be more than 91.7 percent, and the anastomotic rate of morphological classification can be more than 90.5 percent; the particle measuring and classification of one view field cost only a few minutes; and the method has excellent universality and can be used in all the particle measuring and classification works with complex backgrounds and morphologies in the material field and biological field.

The invention discloses a method for determining parameters of a solid oxide fuel cell system, and belongs to the field of fuel cells. The solid oxide fuel cell system is analyzed in a static mode, and the tree-form classification statistical method is adopted to collect steady-state output operation points of the system in a traversal mode to form an operation point set; furthermore, the cubic convolution interpolation algorithm is used for obtaining a large number of operation point data; based on the expanded steady-state output operation point set and output feature parameters, a safety operation range meeting the system temperature restrain requirement is found out, and the optimal operation points under different power can be found out. The found system optimal operation points are the optimal parameter combination of the solid oxide fuel cell system working in a steady state and are the foundation for optimal control of the solid oxide fuel cell system, and stable and efficient operation of the system can be guaranteed.

A classifying method according to an eye form considers the eye form feature as an important factor, to provide an eye cosmetic treatment method for appropriately and quickly making up the eye so that the eye balances and looks large and attractive according to the type categorized by the classifying method. A makeup tool is devised by integrating an eye form feature analyzing method with a form shaping method by markup. An eye form classifying method uses as indexes, three forms, i.e., the eye frame form showing the shape of the eye contour, the eye form showing the three-dimensional shape of the eye, and the angle form of the inner corner and outer edge, the eye forms are classified. The eye forms are classified by comparing the eye form with the eye form of a reference balance eye.

The invention discloses a modeling method of industry chain ecological big data model, comprising selecting a target industry chain, establishing an industry ecological system, establishing a big data component application framework model, establishing a big data structural form classification model, implementing link division for the industry chain based on the industry ecological system and according to the features of the industry ecological system where the target industry chain belongs, implementing standard design for the business process comprehensive data structure of the target industry chain and the comprehensive data structure of the industrial subject, meanwhile, implementing standard modeling of data interface system for the input data type and standard etc., related to different industry chain links, constituting the logical relationship among various data, and finishing big data modeling. The modeling method of industry chain ecological big data model is according to the requirements of the whole industry ecological system formed on the basis of the vertical industry chain and the horizontal correlative industry chain, and comply with the top-level design principle, and achieves automation, systematization and modeling of the big data model.

The invention discloses a retina vessel segmentation system based on combination of a hessian matrix and region growing. The retina vessel segmentation system comprises a retina image preprocessing unit which is used for extracting a retina image green channel and enhancing the extracted image to improve the contrast; a hessian matrix enhancement unit which re-enhances the image through the hessian matrix and extracts the vessel direction in the retina image; a connected domain classification unit which morphologically classifies the image enhanced by the hessian matrix so as to extract smallvessels; and a region growing unit which performs region growing on the classified image to connect the broken vessel structures in the image so as to enhance the segmentation image and improve the segmentation result. Vessel segmentation of the retina fundus image can be realized by the system, and the algorithm of combination of the hessian matrix and region growing is put forward by using the mode of combining the hessian matrix and region growing for aiming at the problem of appearance of breaking points in the segmentation result of the segmentation algorithm so that the situation of vessel breaking in the extracted image can be further improved and the accuracy of vessel segmentation can be enhanced.

The invention belongs to the technical field of medical science, and discloses a method and a system for detecting abnormal rate of urine erythrocyte. The method for detecting abnormal rate of urine erythrocyte comprises the following steps: firstly sample image acquisition and image processing are carried out; image reading and segmentation are executed; impurities are eliminated; marking is carried out; positioning is carried out; diagnosis and reporting are carried out in order to realize deformity form classification and abnormal rate calculation of urine erythrocyte. The system is stable,and continuous and error-free operation is more than or equal to 200 hours; the method and the system have high accuracy for identification of deform urine erythrocyte, and misreporting rate is lessthan 5%; the system can be used for treating some special cases, such as serious cytoadherence (multiple cells are overlapped together), or multiple cells in edges of acquired pictures or incomplete present of cells; the longest time consuming does not exceed 60 seconds from beginning of image acquisition to ending of analysis and reporting; the system has the advantages of simple interface, easyoperation, and good humanization.

The invention discloses a low-voltage prediction method for a power distribution network based on a support vector machine, and the method comprises the steps: 1), screening out different types of indexes from the formation factors and influence factor of the low voltage of the power distribution network; 2), extracting index sample data from various conventional information systems, and constructing different types of index sets; 3), respectively constructing a low-voltage prediction model for the power distribution network based on the support vector machine and different types of index sets; 4), carrying out the parameter optimization of to-be-optimized parameters in each prediction model through a particle swarm optimization algorithm; 5), substituting the optimized parameters into theprediction models, inputting the index data of a to-be-detected power distribution network into each prediction model, predicting the low voltage of the power distribution network through each prediction model, integrating all prediction results, and obtaining a final low voltage prediction result of the power distribution network. According to the invention, multi-source information is employedfor forming classification index sets, and the method can make the most of the advantages of information diversity of a big data platform, and also can effectively reduce the data dimension and training time of the prediction models.

The embodiment of the invention provides a video pushing method and device based on classification characteristics and electronic equipment, and belongs to the technical field of data processing, and the method comprises the steps: obtaining one or more target videos; analyzing different types of contents contained in the target video to obtain a content analysis result; classifying the content analysis result through a preset classification model to form classification information corresponding to the target video; and pushing a target video matched with the classification information to a target object. Through the processing scheme disclosed by the invention, the video pushing accuracy is improved.

The invention relates to a human body gait information collection and gait form classified identification system and method. The system is composed of a storage module, a display module, a power supply management module, a power supply/screen switching module, an original gait information collection module, an auxiliary function module, a processor module, a wired data transmission module and a wireless data transmission module. The method comprises the operation steps of human body gait form classified identification and gait form classified identification result data transmission. By adopting the system, the human body gait information is collected, and the classified identification of the gait forms in the walking process of a human body is realized. The human body gait information collection and gait form classified identification system and method have the advantages that the structure is simple, the operation is simple, the timeliness and the practicability are high, the system and method are suitable for various kinds of indoor and outdoor occasions needing human body gait information collection and gait form classified identification, the system and method are can be used for gait form recording in daily life and can also be applied in the fields of intelligent home, human body intelligent wearable devices and the like, in addition, the system and method can further be used in medical diagnosis or professional physical training.

A method for learning a rejector is provided. The method includes steps of: acquiring features from positive images and negative images of the classification tree; and deciding a specific feature ID and a specific threshold by which a minimum classification error is derived by referring to a histogram of features of positive images and that of features of negative images if a number of images classified to a k-th node is larger than that classified to a brother node of the k-th node as a result of classifying images acquired with respect to a parent node of the k-th node based on a determined feature ID and a determined threshold for the parent node and then allocating the specific feature ID and the specific threshold in nodes, which have depth equal to the k-th node of the classification tree.

The invention relates to a form development method and system for a WEB system. The method comprises the following steps: 1, designing forms according to different business requirements on a client, and configuring a corresponding service code for each designed form on a server side; 2, setting a plurality of IGrid interfaces on the server side, wherein each IGrid interface correspondingly executes a segment of service code; 3, loading a corresponding IGrid interface from the server side through a JS technology on the client, and representing the form which corresponds to the IGrid interface on an interface. Through form customization, form organization is well realized, meanwhile form classification centralized management is realized, and a field-oriented and business-oriented form assembly applicable to a business system is formed through management and control of form data.

The invention relates to an automatic high-efficiency fresh tea leaf classification device, and belongs to the field of tea leaf processing equipment. The automatic high-efficiency fresh tea leaf classification device comprises a vibration conveying groove, a tea leaf elevator, a conveying chute and a tea leaf classifier. The vibration conveying groove is connected with the tea leaf elevator. Thetea leaf elevator is connected with the conveying chute through a flange. The tea leaf classifier is provided with three net-shaped classifying rollers, and feeding hoppers are arranged at the inlet positions of the rollers separately. The lower portion of the conveying chute is divided into three chute outlets which are connected with the feeding hoppers correspondingly. A whole vibration discharging classification groove is formed in the lower portion of the tea leaf classifier. Partition plates are arranged in the vibration discharging classification groove to form classification areas. Bymeans of the automatic high-efficiency fresh tea leaf classification device, automatic feeding and discharging of the tea leaf classifier are realized; and moreover, the multiple rollers are used forclassifying tea leaf raw materials at the same time, and production efficiency is greatly improved.

The invention relates to a human body upper limb zone shape classification method based on main component analysis and layered clustering. The method includes: collecting human body data to build a human body database; preliminarily selecting human body part data related to upper limb zone shapes, and determining basic variables according to the linear dependence of the rest human body part data and the human body part data related to the upper limb zone shapes; using a main component analysis method to select variables representing the upper limb zone shapes from the basic variables; using alayered clustering method to further determine arm anteversion angle, upper arm circumference and shoulder length to serve as the variables for representing the upper limb zone shapes; classifying theupper limb zone shapes according to the arm anteversion angle, the upper arm circumference and the shoulder length. By the method, the upper limb zone shapes can be directly and effectively classified.

The invention relates to artificial intelligence, and discloses a fine-grained classification method, which comprises the following steps of creating an initial model, acquiring an original image, andpreprocessing the original image to form training data for training the initial model; acquiring loss data corresponding to the original image based on the initial model and the training data; performing random gradient descent back propagation based on the loss data until the initial model is converged within a preset range to form classification model training; and performing classification prediction on the to-be-processed image through the classification model. The invention also relates to a blockchain technology, and the original image is stored in the blockchain. According to the method, the local feature region can be adaptively selected, and the classification prediction accuracy is improved.

The invention relates to a printing method, in particular to a general printing method based on winform software. The general printing method comprises the following steps of 1, form creation; 2, data table creation; 3, data table data input; 4, automatic regulation and management on data in a data table; 5, automatic customization of printing formats; and 6, printing. The method is characterized that the a form user-defined method is used for form creation, and the user-defined steps includes 1, data model definition or form entity definition, 2, form model definition, 3, form classification application definition, 4, form storage process definition, 5, application form application definition, and 6, form practical application. The printing method has the advantages that an automatically printed form conforms to the requirements of a national specification; the manual intervention is not needed in the whole report printing process; the involuntary modification probability of data is reduced; the form formats are flexibly defined; and the formats and the contents of the form can be flexibly regulated according to the change of a national statistical system.

The invention provides a data processing method and device for a classification interaction interface, and solves the technical problem that the interaction interface cannot adapt to a flexible retrieval classification technology. The method comprises the steps of retrieving classification category data according to first interaction data to form a first classification category topological structure of related classification categories; forming a first classification result data set matched with the retrieval result according to the classification type first topological structure, and displaying classification result data in order; and according to the second interaction data, forming classification type combination logic, according to the classification type combination logic, forming a classification type second topological structure, and according to the classification type second topological structure, forming orderly display of a second classification result data set matched withthe retrieval result. Timely classification, classification combination and data matching display of mass retrieval data are formed, and the defects that the data processing process in an existing classification interaction interface is influenced by the interaction process, the matching degree of the search process and the data classification dimension is limited, and data positioning and data matching combination cannot be rapidly formed are overcome.

The invention discloses a high-dimensional data function selection algorithm based on a reverse elimination method and application of the high-dimensional data function selection algorithm in medical treatment. The high-dimensional data function selection algorithm comprises three stages. In the first stage, according to a variable selection method based on interaction, influential factors which can interact with other factors to form function modules are firstly recognized. In the second stage, function modules are generated through the reverse elimination method, through the influential factors generated in the first stage, influential function modules which can form high influences with the influential factors are selected, the factors in the function modules interact with each other, and therefore a strong correlation on dependent variables is generated. In the third stage, classifiers are combined, one function module forms one classifier, and the classifiers are combined to form classification rules on the dependent variables. The high-dimensional data function selection algorithm can provide quantitative results for genetic diagnosis and treatment in medical treatment and health, and predication accuracy is greatly improved.

The invention discloses a method for replacing prompt pictures displayed during startup and shutdown of a windows system. The method for replacing the prompt pictures displayed during startup and shutdown of the windows system comprises the following steps that original resources of the prompt pictures involved are organized and classified according to the different states or stages during startup and shutdown, so that the original resources in the same state or stage are located in the same category, and a formed classification structure is displayed; a command inlet for replacement is provided for each original resource file in each category; resource files of prompt pictures used for replacing the original prompt pictures are prepared; the resource files are led into a system file relevant to the prompt pictures; the system file obtained after replacement is used for displaying the prompt pictures during startup and shutdown. In addition, the invention provides a device for replacing the prompt pictures displayed during startup and shutdown of the windows system. By the adoption of the method and device for replacing the prompt pictures displayed during startup and shutdown of the windows system, the prompt pictures displayed during startup and shutdown can meet individual requirements, and more information can be displayed.

The invention discloses a declaration form classification method. The method comprises the steps of utilizing a deep learning neural network in advance to train a classification model; obtaining eachtarget project name of a target declaration form; and inputting each target project name into the preset classification model, and obtaining a corresponding classification result. The classification model learns the characteristics of different types of financial statements in advance, so that compared with the mode of determining the report type of the target declaration form by performing project name matching on the target declaration form and a standard financial report form in the prior art, the method has the advantages that the influence of irrelevant symbols, messy codes, difference inexpression and the like on classification can be eliminated, and the accuracy of classifying the declaration forms is further improved. The invention furthermore discloses a declaration form classification device and equipment and a computer readable storage medium, which all have the above beneficial effects.