61results about How to "Simplify the classification process" patented technology

The invention provides a network object processing method and device, and belongs to the technical field of computers. In the embodiment of the invention, at least two types of modal information can be firstly extracted from a target object, then feature vectors corresponding to all the modal information can be calculated, the feature vectors corresponding to all the modal information can be fusedfor obtaining a fused feature vector, and finally, the target object is classified on the basis of the fused feature vector. The fused feature vector can represent features corresponding to all the modal information, thus in the embodiment of the invention, a terminal can simplify a classification process while the target object is classified on the basis of all the information features of the target object, and classification costs are reduced.

The invention discloses an audio classification method, device and a computer readable storage medium, and belongs to the technical field of electronics. The method comprises: collecting an audio signal; intercepting or supplementing the audio signal to adjust the duration of the audio signal to a preset duration; converting the audio signal to a target audio according to the frequency informationof the audio signal; extracting audio features of the target audio through a convolutional network contained in a preset classifier; extracting time-order features of the audio features through a threshold circulation network contained in the preset classifier; and determining a probability that a category of the target audio is a preset category identified by each of multiple preset category identifiers through a fully-connected network contained in the preset classifier according to the time-order features; and determining the preset category identified by a preset category identifier having the highest probability among the multiple preset category identifiers as the category of the target audio. With the adoption of the method, segmentation of the target audio is avoided, the integrity of the target audio is preserved, and the classification accuracy is relatively high.

A method for facilitating semantic event classification of a group of image records related to an event. The method using an event detector system for providing: extracting a plurality of visual features from each of the image records; wherein the visual features include segmenting an image record into a number of regions, in which the visual features are extracted; generating a plurality of concept scores for each of the image records using the visual features, wherein each concept score corresponds to a visual concept and each concept score is indicative of a probability that the image record includes the visual concept; generating a feature vector corresponding to the event based on the concept scores of the image records; and supplying the feature vector to an event classifier that identifies at least one semantic event classifier that corresponds to the event.

The invention discloses an adjustable type household garbage crusher. The adjustable type household garbage crusher comprises a housing, a charging, horizontal-throwing and winnowing mechanism, a crushing mechanism and a power mechanism, wherein the crushing mechanism comprises a hammer head crushing combined cutter crushing group; and the charging, horizontal-throwing and winnowing mechanism comprises a housing, a charging and throwing conveyor belt, a resistance fan, a light material outlet and a heavy material outlet. Compared with the prior art, the adjustable type household garbage crusher has the advantages that: labor power is saved through the charging, horizontal-throwing and winnowing mechanism, and garbage sorting programs are simplified; materials with relatively great hardnessare subjected to secondary treatment, so that abrasion, on blades, of hard materials is alleviated, and crushing efficiency is improved; generally, the materials are only crushed by blades, so that equipment burden is alleviated; and when no hard materials and heavy materials are added, a hammer head crushing mechanism can be separated from a main shaft, so that electric energy is greatly saved,and equipment burden is also alleviated.

The invention provides an island micro power grid probabilistic load flow analytic calculation method. The calculation method is provided for calculating an operation state of an island micro power grid including intermittent DG such as photovoltaic electricity, wind power and the like; loads of the island micro power grid fluctuates randomly. A conventional island micro power grid steady state load flow analysis method is mainly based on load flow calculation, a load flow result of only a single operation point can be obtained, research on probabilistic load flow calculation is only based ona simulation method which requires computational statistics on a large amount of samples so as to obtain a probabilistic load flow result, online analysis can be adversely affected due to long computation time, and correlation of DG contribution in the micro power grid is mostly not taken into consideration. To overcome the above limitations, the invention provides the island micro power grid probabilistic load flow analytic calculation method via which rapid island micro power grid probabilistic load flow calculation can be realized, requirements for engineering accuracy can be met, and the method disclosed in the invention can be used for on-line analytical application.

The invention provides a polarimetric SAR image classification method based on feature attention and a feature improvement network. The polarimetric SAR image classification method mainly solves the problems that an existing polarimetric SAR image classification method based on deep learning is poor in intra-area consistency and inconvenient for end-to-end classification. The implementation schemecomprises the following steps of: 1) inputting a to-be-classified polarized SAR image and filtering the to-be-classified polarized SAR image; 2) synthesizing a pseudo color image and a classificationlabel of the polarized SAR image; 3) extracting initial features of the polarimetric SAR image and preprocessing the features; 4) respectively constructing an input representation layer, a feature attention sub-network, an encoder and a decoder, and sequentially connecting the input representation layer, the feature attention sub-network, the encoder and the decoder to form a feature attention and feature improvement network; 5) training fa eature attention and feature improvement network; 6) Inputting the polarization SAR image into the trained network to obtain the classification result. The polarimetric SAR image classification method is high in intra-area consistency, low in noise and high in classification precision, end-to-end learning and classification are realized, and the polarimetric SAR image classification method can be used for polarimetric SAR image classification.

The embodiment of the invention provides an application program downloading and classifying method and terminal equipment, relates to the technical field of communication, and aims to solve the problems that a user classifies application programs complexly, consumes time and is poor in man-machine interaction performance. The method comprises the following steps: receiving a first input of a userfor a target folder; in response to the first input, displaying a target identifier which is used for indicating an application program downloading entrance; receiving a second input of the user for the target identifier; displaying an application program downloading interface in response to the second input, the application program downloading interface being an interface corresponding to the application program downloading entry; downloading and installing a target application program through the application program downloading interface; and displaying the application icon corresponding tothe target application program in the target folder. The method is applied to a scene where a user classifies application programs.

The invention discloses a vehicle-mounted delay-tolerant network data forwarding method based on a semi-naive Bayesian classifier, comprising the following steps: S1, each node carries a data message,the current node obtains the current information, and a data message forwarding history table is maintained according to the semi-naive Bayesian classifier and the point; 2, the current node is movedand judging whether the meeting node is a destination node, if yes, directly forwarding the data message to the target node, and forwarding is finished; if not, the meeting node does not have a datamessage forwarding history table to calculate the current probability I of the current node successfully delivering the data message; S3, acquiring the probability I of the meeting node successfully delivering the data message; S4, comparing the current I and the meeting I, returning to S2 if the current encounter is greater than I, and forwarding the data message carried by the current node to the meeting node if the current I is less than the meeting I. The invention not only can significantly improve the message delivery rate, but also effectively reduce the network overhead, and has highuse and promotion value.

The invention discloses a chip surface defect classification method based on a generative adversarial network. The method comprises the following steps: S01, inputting a random vector into an image processing module, obtaining a composite image, combining a photographed image and the composite image into an image set, and enabling the image set to comprise a training image set; s02, inputting thetraining image set into a training unit for model training to obtain a recognition model; and S03, inputting a to-be-recognized image into the recognition model, and obtaining a corresponding defect type through recognition of the recognition model. According to the chip surface defect classification device and method based on the generative adversarial network provided by the invention, the synthetic image is generated by using the random vector, and the synthetic image is used as one part of the image set, so that a large number of training images and test images are formed, and the trainingaccuracy and generalization ability of a recognition model are improved.

An embodiment of the invention discloses a processing method and a device for portable execute (PE) files, relates to the field of program detection, can reduce influences of library functions on PE file classification and clustering, and improves accuracy of the classification and the clustering. The processing method for the PE files includes that disassembling process for the PE files is achieved to obtain function nodes of the PE files, and the function nodes divide the PE files into a plurality of function sections; a library function feature library is loaded, and whether the function sections are library functions is sequentially judged according to features in the library function feature library; and if the function sections are the library functions, the function sections are eliminated.

The invention discloses an observation vector difference based method for classifying SAR image textures. The method mainly solves the problem of terrain classification of SAR images. The method comprises the steps of (1) randomly selecting r images from a training set for partitioning processing, converting to obtain a column vector difference matrix P; (2) observing the P with an observation matrix to obtain a texture observation vector difference matrix X, and conducting clustering on the X to obtain a texture dictionary D; (3) calculating images of the training set according to Step (2) to obtain an observation vector difference matrix Xtr; (4) projecting the Xtr onto the texture dictionary D to form a training image texture column diagram h; (5) representing images of a test set by a test image texture column diagram he; (6) calculating the distance between the he and the h, and determining the classification to which the he belongs according to the distance; and (7) calculating all test images according to Step (6) to obtain a final classification rate. According to the method, the latest compressed sensing theory is applied, the process is simple, the classification identification rate is high, and the method is applicable to terrain texture classification of SAR images.

The invention discloses a mouth model training method, a mouth recognition method, device and apparatus and a medium. The method includes obtaining a face image sample, marking the face image sample to obtain face image sample data, extracting a feature vector of the face image sample, dividing the face image sample data into training sample data and verification sample data; training a support vector machine classifier by the training sample data, and obtaining the critical surface of the support vector machine classifier; calculating the distance between the eigenvector of the verification sample and the vector of the critical surface in the verification sample data; obtaining a preset real class rate or a preset pseudo-positive class rate, obtaining a classification threshold accordingto the vector distance and the labeling data corresponding to the verification sample data, and obtaining a mouth judgment model according to the classification threshold. Through the mouth model training method, a mouth judgment model with high accuracy of judging whether the mouth is occluded or not can be obtained.

The invention discloses a question type automatic sorting method based on an elementary mathematics-oriented question-solving idea. After questions which need to be sorted are inputted, the knowledgepoint used in the answering process of the standard answer of the questions is taken as the question-solving idea, and the question-solving idea of the questions are extracted; the questions are roughly sorted into a plurality of main classes with determinate quantity according to priority; then the questions of each of the main classes are subtly sorted into a plurality of subclasses with indeterminate quantity; then the questions in the subclasses are calibrated, and the sorting result is outputted and saved. The invention further discloses a question type automatic sorting system based on an elementary mathematics-oriented question-solving idea. The system comprises a question inputting module, an idea extracting module, a rough sorting module, a subtle sorting module and a calibratingmodule. The method and system achieve more accurate pushing for users' exercises, improve user experience, and use the automatic sorting method to accelerate sorting process and to save labor and time.

The invention provides a cloud-based intelligent garbage identification and cleaning method and a related system. And through the unmanned sweeper for detection, uninterrupted inspection and classification of garbage in a fixed area are realized. And directional classification treatment and recycling of the garbage are achieved through the unmanned sweeping vehicle set. Garbage is classified and recognized through a cloud server network, intelligent and efficient scheduling is conducted on unmanned sweeping vehicles used for detection and unmanned sweeping vehicle number grouping configuration, unmanned operation in the whole process of classified sweeping and classified recycling and transferring of garbage is achieved, the electricity utilization cost of equipment is reduced, and the labor cost is reduced.

The invention discloses an HBV classifying method based on Gaussian blur integrals. The HBV classifying method comprises the following steps that 1, DNA sequences of HBV patients are screened from an HBV database; 2, the database is divided into four small data sets B1, C1, C2 and C3 according to a clustering method; 3, the data sets are classified depending on classification of a classifier and true categories of cases; 4, the classifier constructed through the Gaussian blur integrals is applied to the HBV database and classifies HBVs. Based on the blur integrals of Gaussian distribution, Gaussian functions represent integrand functions to finish projections of the blur integrals, then linear classification is carried out according to virtual integral values obtained through the projections, the HBV classification precision is improved, and the HBV classification process is simplified.

The invention belongs to the technical field of laser probe component analysis, and discloses a method for improving classification accuracy of laser probe by utilizing image characteristics. The method comprises the following steps: S1, collecting plasma spectrum of a sample by adopting a spectrum collecting device, wherein the plasma spectrum is a multi-dimensional spectrum; S2, performing imageprocessing on the plasma spectrum to obtain a 16-bit gray scale image; S3, extracting image features based on the gray image, wherein the image features are multi-dimensional spectral features; S4, training the image features as an input of a classification algorithm to obtain a classification model based on the image features; S5, inputting the image features of the sample to be classified to the classification model, and outputting the classification result by the classification model, thereby completing the classification. The invention improves the splitting precision and does not need manual or automatic route selection, simplifies the classification process and improves the classification efficiency.

The embodiment of the invention provides an application program control method and electronic equipment, relates to the technical field of communication, and aims to solve the problems of tedious andtime-consuming application control process and poor man-machine interaction performance. The scheme comprises the following steps: receiving a first input for indicating icon classification of an application program; responding to the first input, an icon of the first application program and an icon of the second application program are displayed on a target interface, and the target interface isa target desktop interface of the electronic equipment or an unfolding interface of a target folder; wherein the second application program comprises at least one application program authorized to login based on the first account information, and the first account information is account information in the first application program. The application program control process is simplified, the interaction duration is shortened, and the man-machine interaction performance is improved.

A method for facilitating semantic event classification of a group of image records related to an event. The method utilizing an event detector system for providing: extraction of a plurality of visual features from each of the image records; generation of a plurality of concept scores for each of the image records using the visual features; generation of a feature vector corresponding to the event based on the concept scores of the image records; and supplying of the feature vector to an event classifier that identifies at least one semantic event classifier that corresponds to the event, wherein the visual features include segmenting an image record into a number of regions, in which the visual features are extracted, each concept score corresponds to a visual concept and each concept score is indicative of a probability that the image record includes the visual concept.

The invention discloses a method and device for classifying network traffic data. N flow data samples are obtained from network data stream, and the data size of each flow data and the identification of each flow data are obtained. According to the data size of each flow data, N flow data are divided into K categories. According to the identification of each flow data, the multiple times of sampling of each flow data in each category are carried out, the number of times of successful sampling and the number of times of failed sampling are obtained, and the numbers of times of sampling of the flow data in a same category are the same. According to the number of times of successful sampling and the number of times of failed sampling, the expectation and uncertainty probabilities of each flow data in a corresponding category are obtained. The weight of the category where each flow data is located is calculated according to the expectation and uncertainty probabilities. According to the weight of the category where each flow data is located, the flow data is reclassified according to a preset classification rule. The problem that the all flow data cannot be randomly classified in the prior art is solved.

The embodiment of the invention discloses a data flow classification method, which is applied to message forwarding equipment between an internal network and the Internet. The method provided by the embodiment of the invention comprises the following steps: message forwarding equipment acquires a plurality of data streams, and extracts address information and time information of each data stream in the plurality of data streams; according to the source IP address of each data stream, a data stream set generated is screened when a first client device accesses the plurality of services from theplurality of data streams; according to the destination IP address and the destination port number of each data stream in the data stream set, a service set which is accessed by the first client device and comprises a first service and a second service is confirmed; the correlation among services is determined in the service set according to the time information of each data flow in the data flowset so that first service and second service are used for realizing a first application. therefore, the message forwarding device determines that the data streams corresponding to the first service and the second service are the data streams of the first application.

The invention relates to a rotatable shelf bottom that has a tablet function (1) and is composed of a shell which is rotatably mounted on the bottom side (5b) of a bottom part (5). The shell comprises a bottom part (5) and a frame (1a) which are positively glued together by means of an adhesive duct that is filled with an adhesive. A frame (1a) is substantially L-shaped, forms part of the bottom side (5) of the bottom part (5a) at the outer edge, and is removably placed on the top side of horizontal furniture components (30) located inside or outside a furniture carcass.

A method and plant for thermal treatment and chemical transformation of waste comprising natural and synthetic carbonaceous materials for generation of a fuel gas for further use is described. Pyrolysis gas and solid waste from a thermolysis and pyrolysis reactor (40), is further processed to produce a fuel gas having a substantially stable WOBBE index.

The invention discloses an analogical electronic circuit fault diagnostic method based on an M-ary-structure classifier, which belongs to the field of analogical circuit network test. The invention comprises the following steps: firstly, analyzing the testability of an analogical circuit to determine proper test stimulus and a test point; secondly, acquiring to-be-tested outputting signals at a testable node of an electronic circuit; thirdly, compressing acquired circuit fault information to extract fault characteristic samples; finally, designing the tags of the samples in an M-ary encoding way, establishing a hyperspherical sub-classifier in the way of a SVDD classifier, training the samples, and storing information to form a fault dictionary after training. The invention has the advantages of less classifiers, simple method, high reliability, and the like, thereby improving the automation degree and the efficiency of the online fault diagnosis of a power electronic circuit.

The invention relates to an image classification method and device, equipment and a storage medium, and the method comprises the steps: preprocessing a to-be-processed original image to acquire a corresponding input image; performing feature extraction on the input image by adopting a pre-training model to obtain a plurality of feature data; wherein each piece of feature data corresponds to different image layers of the input image; performing dimension reduction on the extracted feature data, so that a feature set after dimension reduction can be obtained; connecting the feature data of different dimensions in the feature set in series to form a feature change curve; and classifying an original image by utilizing the trained classification network model according to the characteristic change curve. A pre-training model is directly adopted for extraction when feature extraction is carried out, a large number of samples do not need to be manufactured for training the network model for feature extraction, dependence on the training samples is effectively reduced, the image classification process is simplified, and overall efficiency of image classification is improved.

The invention relates to the technical field of material physical data classification, and discloses a material physical data classification system based on the computer big data. The system comprises a central processing unit, the input end of the central processor is electrically connected with the output ends of an information acquisition unit and an information feedback unit. The output end of the central processing unit is electrically connected with the input ends of a data calculation unit, a deletion module, a data classified storage unit and an identification judgment unit, and the output end of the data calculation unit is electrically connected with a data statistics unit. A first information data acquisition module, a second information data acquisition module, a third information data acquisition module and a fourth information data acquisition module in an information acquisition unit are utilized; material data in each field are collected, and are calculated, compared, identified and judged by the central processing unit after being collected, and then are stored in a classified manner, so that the program of material physical data classification is simplified, and the working efficiency is improved.

The invention relates to a seal image recognition and classification method and device, computer equipment and a storage medium. According to the invention, different seals are classified based on the rotation change condition of the circumscribed rectangles of the seal connected domain; and meanwhile, correction processing of the seal is achieved. Compared with a traditional classification method based on edge detection, the invention does not need to carry out collection training on prior data; the classification process is simplified, implementation is easy, and the efficiency of stamp image classification is further improved. The method comprises the following steps: acquiring an image containing a seal; solving a seal connected domain for a seal area in the image; rotating the seal connected domain for multiple times at a preset angle in a fixed direction, and solving the circumscribed rectangles of the rotated seal connected domain each time to obtain a plurality of circumscribed rectangles; calculating the area of each external rectangle; and classifying the seals according to the size relation of the areas of all the external rectangles.

The invention discloses a classifying method of crop data based on fuzzy C mean value utilizing improved gene expression programming. The method comprises following steps: using customized similarity to measure a calculating formula and combining with information entropy to calculate weight vector of data attribute; and using a weighed distance calculating formula to replace conventional Euclidean distance and combining a gene expression program with a fuzzy C mean value algorithm to solve a optimal cluster center for classification of crop data. The classifying method of crop data based on fuzzy C mean value utilizing improved gene expression programming has following beneficial effects: on one hand, distance between crop data can be measured so that accuracy of the classification result is increased; and on the other hand, the optimal classification result can be obtained by lower iteration frequency.

The invention discloses an event subject recognition model optimization method, device and equipment and a medium, and the method comprises the steps: inputting a training text into an information extraction module, and extracting character information and event information in the training text; inputting the character information and the main body label of the training text into a main body identification module to obtain a main body identification result and main body identification loss; inputting the event information and the subject recognition result into an event and subject matching module to obtain a matching result, and calculating matching loss based on the matching result and an event subject matching label corresponding to the training text; inputting the event information into an event classification module to obtain an event classification result, and calculating classification loss based on the event classification result and the event type label; and optimizing the subject identification loss, the matching loss and the classification loss to optimize the event subject identification model. According to the method, the classification problem is simplified, so that the error probability of the model is reduced, and the accuracy of event classification and event subject identification of the model is improved.

The invention discloses a chip surface defect classification method based on deep learning. The method comprises the following steps of S01, inputting shot images into an image preprocessing unit, theimage preprocessing unit preprocessing the shot images to form processed images, combining the shot images and the processed images into an image set, and the image set comprising a training image set; S02, inputting the training image set into a training unit for model training to obtain a recognition model; and S03, inputting a to-be-recognized image into the recognition model, and obtaining acorresponding defect type through recognition of the recognition model. According to the chip surface defect classification device and method based on deep learning provided by the invention, the original shot image is preprocessed by using an image processing method including scaling, rotation and wrong cutting, so that a large number of training images and test images are formed, and the training accuracy of a recognition model is improved.

The invention discloses a power load prediction method based on a grey correlation degree and a support vector machine. The method comprises the following steps: firstly, screening out historical daysconsistent with the date type of a prediction day; according to the meteorological factors, calculating gray correlation degrees between the prediction days and the historical days; and selecting a historical day with a high correlation degree as a sample of the support vector machine, and establishing a support vector machine model. The historical day most similar to the prediction day is selected according to the date type and the meteorological factors to establish the support vector machine model, and the effectiveness and accuracy of the prediction method can be effectively improved.