Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Hiro

37results about How to "Increase the amount of training data" patented technology

Industrial product surface defect detection method based on sample enhancement

The invention discloses an industrial product surface defect detection method based on sample enhancement. The method comprises the following steps: 1) carrying out size standardization, normalization, cutting and classification on an industrial product surface image; 2) carrying out data enhancement of random flipping on the picture with the defect; 3) randomly splicing and enhancing the defective picture and the normal picture; 4) carrying out iterative training by using a Cascade-RCNN algorithm; 5) acquiring a Cascade-RCNN detection model; 6) carrying out sliding window detection on the industrial product surface picture to be detected and the texture template picture determining no defect through the Cascade-RCNN detection model, splicing results detected by a sliding window, and comparing the results obtained by the sliding window detection and the texture template picture to finally obtain the defect category and region annotation of the picture to be detected. According to the method, the influence of conditions such as illumination, exposure and displacement on defect detection can be effectively reduced, the detection stability is improved, the resolution capability of thetwo-stage target detector on patterns and backgrounds is improved, and the false detection rate is reduced.
Owner:SOUTH CHINA UNIV OF TECH

Surface defect detection method and device and computer readable storage medium

The invention is applicable to the technical field of image recognition, and provides a surface defect detection method, which comprises the following steps of: acquiring a surface image of an industrial part; partitioning the original large-size image of the surface image into a plurality of sub-images to obtain a sub-image training data set; constructing a defect detection network; inputting the sub-images in the sub-image training data set into the defect detection network for training to obtain a trained defect detection network; carrying out segmentation processing on a to-be-detected image, and segmenting the to-be-detected image into a plurality of sub-image block images; and inputting the plurality of sub-image-block images into a trained defect detection network to obtain detection results of the plurality of sub-image-block images. According to the method provided by the invention, the problem that the detection difficulty is increased due to poor surface defect detection precision, small data volume and small defects in the prior art is solved.
Owner:ZHENGZHOU JINHUI COMP SYST ENG

Oral cavity image multi-tissue full-automatic segmentation method and system

The invention provides a method and a system for performing full-automatic segmentation on multiple tissues of an oral cavity CBCT image. According to the obtained original oral cavity three-dimensional CBCT data, a deep learning method is utilized to label the data, a deep learning method is utilized again to segment a new image, the image processing process is simple, excessive manual labeling and intervention are not needed, the difficulty of existing medical data annotation is overcome, and meanwhile human differences in the image annotation process are reduced. In the network training process, the adversarial network is generated for the training data, the training data volume is increased, and the universality of the network can be effectively improved. The establishment of the method greatly promotes the segmentation of a medical image technology, and provides an effective technical support for the automatic diagnosis and prognosis of oral diseases.
Owner:PEKING UNIV SCHOOL OF STOMATOLOGY

3D point cloud semantic segmentation migration method based on meta-learning

The invention discloses a 3D point cloud semantic segmentation and migration method based on meta-learning, and relates to the technical field of robot navigation. The method comprises the following steps: constructing a Point Net network model; selecting a training data set; for each training data set, forming a training task set by utilizing different types of data; constructing a meta-learningframework; according to the meta-learning framework, training the Point Net network model through each training task set; selecting a test task set; and inputting the test task set into the trained Point Net network model for testing until the gradient update value of the model converges. According to the method, a trained model is loaded in a new environment task, optimal similar task parametersare used, and the training efficiency of the indoor scene semantic segmentation method of the new task is high; the semantic segmentation capability of different tasks is learned through a meta-learning framework, it is ensured that learning features can be suitable for different migration environments, and the generalization performance of the model is improved.
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA

Optic nerve simulation method and optic nerve simulation system based on deep learning

The invention discloses an optic nerve simulation method and an optic nerve simulation system based on deep learning, wherein the method comprises the steps of transmitting a visible light signal; acquiring a visual neural electrical signal for obtaining a first visual neural simulated electrical signal, performing two-dimensional imaging on a brain visual processing cortex for obtaining a first near-infrared brain functional area simulated two-dimensional image; performing filtering, prior modification and sampling on the received visible light signal, and furthermore obtaining a second visual neural simulated electrical signal and a second near-infrared brain functional area simulated two-dimensional image; and performing analysis and evaluation, using data after evaluation as parameters for correcting a deep leaning module, and transmitting the data to the deep leaning module in a multi-dimensional vector form, thereby reconstructing a deep learning network by means of the multidirectional vector. According to the optic nerve simulation method, medical simulation and rehabilitation training can be performed on a human brain and an optic nerve system by means of a deep learning method. Simple operation and high reliability are realized. Furthermore along with increase of training data volume, system gain can be remarkably improved.
Owner:TSINGHUA UNIV

Method for generating large batch of point cloud data sets

The invention discloses a method for generating large batch of point cloud data sets. The method comprises the following steps: establishing a three-dimensional model for an object in a scene; re-sampling points for each vertex and each patch in the three-dimensional model to form a surface point cloud of the three-dimensional model; combining the surface point clouds of components in the scene togenerate scene point cloud information; processing the scene point cloud information by adopting a simulated camera model to form simulated scene point cloud information acquired by a simulated structured light system; generating a large batch of point cloud data sets by repeating for many times. According to the method, the training data volume of a neural network for point cloud processing is improved, and meanwhile, the method has pertinence for specific scenes and processing tasks, so that the performance of the neural network is improved.
Owner:SOUTH CHINA UNIV OF TECH

Model training method, relative relationship identification method, electronic equipment and storage medium

The embodiment of the invention relates to the field of face recognition, and discloses a model training method, a relative relationship recognition method, electronic equipment and a storage medium. The model training method comprises the following steps: constructing a family face feature extraction model for extracting family face features from family image samples; wherein the family image sample comprises a plurality of face images belonging to the same family, and the family face feature extraction model is provided with organ parameters for fusing individual face features of the plurality of face images to form family face features; the organ parameter represents the dominant degree of the face feature of the corresponding organ region in the individual face feature in the family face feature; and constructing a loss function according to the distance between the family face feature of each family and the corresponding family label and the distance between the family face feature and the face feature of each face image in the corresponding family image sample, and training a family face feature extraction model.
Owner:合肥的卢深视科技有限公司

Project performance guarantee insurance risk identification method and platform based on multi-party cooperation

ActiveCN111626886AImprove risk identification abilityGuarantee data securityFinanceCharacter and pattern recognitionPrivacy protectionGlobal risk
The invention discloses an project performance guarantee insurance risk identification method and platform based on multi-party collaboration, and the method comprises the steps: firstly enabling allparties participating in a collaboration training risk evaluation model to carry out preprocessing operation and value mapping operation of locally stored project project information data to acquire project business data after privacy protection; then, each participant and the cloud server complete training of the risk assessment model through distributed cooperation to obtain a global risk assessment model; and finally, project to-be-accessed project data is processed and the processed project data is inputted to the global risk assessment model for risk identification and prediction. According to the invention, cross-data-source multi-party cooperation risk identification can be carried out on the premise of protecting data privacy, and the risk identification capability of the risk assessment model is improved while project information data security is ensured.
Owner:工保科技(浙江)有限公司

Convolutional neural network training method, electroencephalogram signal recognition method and device and medium

The invention discloses a convolutional neural network training method, an electroencephalogram signal recognition method and device and a medium, and the method comprises the steps: executing a plurality of obtaining processes, obtaining an electroencephalogram signal in each obtaining process, and executing the time domain data enhancement and frequency domain data enhancement of the electroencephalogram signal, and training a convolutional neural network by using the enhanced electroencephalogram signal, and the like. The convolutional neural network trained by the method is a multi-input,multi-convolution-scale and multi-convolution-type hybrid convolutional neural network, the sizes of a multi-input convolution layer and a convolution kernel are reasonably designed, and the method has high recognition accuracy; a training set used for training the convolutional neural network is obtained by performing time domain data enhancement and frequency domain data enhancement expansion based on the acquired electroencephalogram signals, so the training data volume of the convolutional neural network can be increased, the over-fitting phenomenon can be reduced, noise interference in the electroencephalogram signals can be effectively coped with, and the recognition effect can be improved. The method is widely applied to the technical field of signal processing.
Owner:GUANGZHOU UNIVERSITY

A complex remote sensing land environment small sample and small target rapid detection and identification method

The invention discloses a complex remote sensing land environment small sample and small target rapid detection and identification method, and the method comprises the steps of constructing a detection and identification network for a complex remote sensing land environment vehicle target based on an improved Faster R-CNN convolutional neural network architecture; carrying out certain transformation and disturbance expansion on training data and carrying out repeated training on negative samples and difficult-to-distinguish samples, so that a network can fully learn the change of a target while the training data volume is increased, and the problems of weak model generalization ability and poor precision caused by small sample data volume are solved; adding small target features and mining difficult sample information, so that the problems that Faster R-CNN is poor in small target detection effect, high in false alarm rate and low in recognition precision are solved; the RPN and the Fast R-CNN sharing the same five-layer convolutional neural network, and adjusting and optimizing network model parameters, so that the whole detection process only needs to complete a series of convolution operations to complete the detection and identification process, and the operation time is reduced.
Owner:北京理工雷科电子信息技术有限公司

Training method, evaluation method, control method and device of automatic driving model

The invention discloses a training method, an evaluation method, a control method and a device for an automatic driving model. The driving track and driving behavior of each vehicle in an actual traffic scene can be mapped through a simulation system, further the driving information of a large number of vehicles in the actual traffic scene can be obtained, and the training data size of the model is increased. In the process of training the automatic driving model, the average deviation value between each target simulation vehicle and each automatic driving vehicle is continuously calculated, the target automatic driving model can be determined in time when the average deviation value is smaller than the first threshold value, and a training environment does not need to be redeployed when the average deviation value is not smaller than the first threshold value. The method improves the training efficiency, reduces the calculation amount, greatly reduces the cost, and can be widely applied to the technical field of automatic driving.
Owner:TENCENT TECH (SHENZHEN) CO LTD

Image text recognition method and storage medium

The invention relates to an image text recognition method and a storage medium. The method comprises the following steps: S1, receiving first image text information; the image text information comprises first text information and background image information; s2, extracting first text information and background image information, and determining parameter information corresponding to the first text information; s3, obtaining one or more pieces of text information from a text database, and processing the obtained text information by adopting the parameter information corresponding to the first text information to obtain second text information; and S4, synthesizing the second text information and the background image information into second image text information, and inputting the second image text information into the text detection model for training. According to the scheme, the training data volume of the detection model can be effectively expanded, and then the accuracy of the trained model on text detection is improved.
Owner:FUJIAN YIRONG INFORMATION TECH +2

Deep learning image registration method and device, electronic equipment and storage medium

Embodiments of the invention disclose a deep learning image registration method and device, electronic equipment and a storage medium. The method comprises the steps of obtaining deformation fields ofa first fixed image and a moving image through a first algorithm, acquiring a second fixed image according to the deformation field and the moving image, and training a deep neural network accordingto the moving image, the second fixed image and the deformation field to obtain a trained deep neural network. According to the deep learning image registration method provided by the embodiment of the invention, the simulated deformation field is generated by using the generation algorithm and is not artificially labeled, so that the registration network is better trained, and the problems that the registration standard is defined and the registration process is driven by depending on the artificially set registration metric in the prior art are solved. According to the method, the training data volume is greatly improved in the neural network, the registration accuracy is further improved, and the calculation speed is high.
Owner:SHENZHEN UNIV

Bill image layout analysis method and device

A bill image layout analysis method comprises the following steps: preparing a training layout sample for model training, and carrying out manual annotation; performing feature coding on a textbox in the training layout sample; performing feature splicing on the coordinate features of the textbox and text features to form splicing features of the textbox; splicing the splicing features of a plurality of candidate frames in a training layout sample to form a feature sequence vector of the training layout sample; training the model to obtain a layout analysis model; and inputting the feature sequence vector of the to-be-analyzed layout into the layout analysis model to obtain an analysis result of the to-be-analyzed layout. According to the method, a machine learning method is adopted, the end-to-end training and processing process is achieved, compared with a traditional feature engineering method, the manual workload is greatly reduced, meanwhile, the operation efficiency of a model framework is improved, and the accuracy of bill image layout analysis is remarkably improved.
Owner:北京玖安天下科技有限公司

Interpolation grouping sliding gravity center-based nationwide grain consumption prediction method with multiple regression model

The invention discloses an interpolation grouping sliding gravity center-based nationwide grain consumption prediction method with a multiple regression model. The method comprises the following stepsin sequence: (1) acquiring original data; (2) acquiring significant influence factors; (3) respectively pretreating grain consumption and two significant influence factors so as to obtain pretreatment data; (4) calculating a sliding gravity center of the pretreatment data; (5) inputting the data obtained in the step (4) into a multiple regression model so as to obtain a primary prediction resultof grain consumption; and (6) performing sliding data gravity center inverse operation on the prediction result obtained in the step (5), thereby obtaining an expected actual prediction value y'1(t) of grain consumption. By adopting the method, prediction errors can be smoothed, data gravity center decrement in the grouping sliding gravity center calculation process can be compensated by additional interpolation pretreatment, meanwhile due to addition of training data of prediction models, the prediction precision of the multiple regression model can be remarkably improved.
Owner:HENAN UNIVERSITY OF TECHNOLOGY

Lung pathological sound automatic analysis method based on multi-task classification

The invention discloses a lung pathological sound automatic analysis method based on multi-task classification, and relates to the technical field of lung pathology analysis. The method comprises the following steps: inputting extracted audio features into a multi-task classification model of a convolutional neural network MobileNetV2, wherein the multi-task classification model of the convolutional neural network comprises the steps of outputting a lung pathological sound recognition task and outputting a lung illness prediction task. According to the method, the training data volume can be implicitly increased by adopting a multi-task learning method, the generalization performance of the model is improved through domain knowledge of multiple pieces of label information of the same data, so that the prediction accuracy of the multi-task classification model of the convolutional neural network MobileNetV2 is improved, and in addition, the lightweight multi-task classification model of the convolutional neural network MobileNetV2 is adopted, so that the number of parameters is small, the requirements for the computing power and the memory size of training equipment are small, and the prediction classification task can be completed on mobile or embedded equipment.
Owner:NANKAI UNIV

Machine learning model training method and device and image classification method and device

The invention relates to a machine learning model training method and device and an image classification method and device, and relates to the technical field of artificial intelligence. The training method comprises the steps that a first machine learning model is utilized to extract a feature vector of a to-be-processed picture, a first classification result of the to-be-processed picture is determined according to the feature vector, and the to-be-processed picture belongs to a first data field or a second data field; according to the feature vector, a second machine learning model is utilized to determine a second classification result of the to-be-processed picture, and the second classification result comprises a classification result of the to-be-processed picture in the first data domain and a classification result of the to-be-processed picture in the second data domain; and according to the first classification result and the second classification result, adversarial training is performed on the first machine learning model and the second machine learning model, so that the accuracy of the second classification result is lower than a threshold, and the trained first machine learning model is used for picture classification.
Owner:BEIJING WODONG TIANJUN INFORMATION TECH CO LTD +1

Image data enhancement method and device, electronic equipment and storage medium

The invention relates to an image data enhancement method and device, electronic equipment and a storage medium, and the method comprises the steps: S1, obtaining an original image, carrying out the annotation of an annotation box of the original image, and obtaining an annotated image; S2, rotating the marked image, and obtaining a rotated image; S3, reducing and / or amplifying the rotated image, wherein the reduction processing comprises the steps that two adjacent sampling points serve as segmentation lines, the rotated image is divided into multiple sub-block images, the average value of the sum of the pixels of the corresponding sub-block images is taken as the pixel of the multiple sub-block images, and the amplification processing comprises the steps that a Bi Cub i c function is constructed, performing interpolation calculation on neighborhood points near the pixel points to be interpolated; and S4, obtaining the annotation data subjected to the zooming-out processing and / or the zooming-in processing. The diversity of data can be greatly improved, and the number of training samples can be quickly increased on the basis of not increasing the manual labeling cost.
Owner:SHENZHEN WEITESHI TECH

Face attribute analysis method and device

The embodiment of the invention provides a face attribute analysis method and device. The method comprises the following steps: acquiring a basic two-dimensional face image and a basic three-dimensional face point cloud of a target face; rendering the basic three-dimensional face point cloud to a two-dimensional image to obtain a rendered two-dimensional face image; superposing the basic two-dimensional face image and the rendered two-dimensional face image to obtain target input data; inputting the target input data into a preset analysis model to obtain target attribute values of a plurality of preset face attributes of the target face, the preset analysis model being a model obtained by training a preset neural network by using a preset training set, the preset training set comprising sample two-dimensional face images and sample three-dimensional face point clouds of a plurality of sample faces, and the sample attribute values of the multiple preset face attributes of each sample. By applying the technical scheme provided by the embodiment of the invention, the accuracy of face attribute analysis is improved.
Owner:HANGZHOU HIKVISION DIGITAL TECH

Optic nerve simulation method and system based on deep learning

The invention discloses an optic nerve simulation method and an optic nerve simulation system based on deep learning, wherein the method comprises the steps of transmitting a visible light signal; acquiring a visual neural electrical signal for obtaining a first visual neural simulated electrical signal, performing two-dimensional imaging on a brain visual processing cortex for obtaining a first near-infrared brain functional area simulated two-dimensional image; performing filtering, prior modification and sampling on the received visible light signal, and furthermore obtaining a second visual neural simulated electrical signal and a second near-infrared brain functional area simulated two-dimensional image; and performing analysis and evaluation, using data after evaluation as parameters for correcting a deep leaning module, and transmitting the data to the deep leaning module in a multi-dimensional vector form, thereby reconstructing a deep learning network by means of the multidirectional vector. According to the optic nerve simulation method, medical simulation and rehabilitation training can be performed on a human brain and an optic nerve system by means of a deep learning method. Simple operation and high reliability are realized. Furthermore along with increase of training data volume, system gain can be remarkably improved.
Owner:TSINGHUA UNIV

A Text Classification Algorithm for Telephone Appeal Based on DenseNet for Power Field

The invention discloses a telephone appeal text classification algorithm based on DenseNet facing the electric power field, belonging to the technical field of text classification algorithms, through preprocessing the text to be classified, data augmentation, establishing a vocabulary dictionary, word vector id matching, and word vector reduction Dimensions, concatenated eigenvalues, and random permutations and combinations of concatenated eigenvalues ​​are used to obtain a text classifier, and the text classifier is used to classify text. The DenseNet-based phone appeal text classification algorithm for the electric power field provided by the present invention can effectively make up for the shortcomings of traditional algorithms, and can well adapt to the characteristics of strong professionalism, large length differences, and mixed characters and numbers in electric power appeal texts. On the premise of ensuring the classification accuracy, the complexity of the model is reduced, and the rapid and accurate classification of the telephone appeal text in the electric power field is realized, which satisfies the classification requirements well.
Owner:STATE GRID ZHEJIANG HANGZHOU XIAOSHAN POWER SUPPLY CO +3

An automatic analysis method for lung pathological sounds based on multi-task classification

The invention discloses a method for automatic analysis of lung pathological sounds based on multi-task classification, relates to the technical field of lung pathology analysis, and comprises the following steps: inputting the extracted audio features into a multi-task classification model of a convolutional neural network MobileNetV2, The multi-task classification model of the convolutional neural network MobileNetV2 includes an output for a lung pathological sound recognition task and an output for a lung disease prediction task. The present invention uses the multi-task learning method to implicitly increase the amount of training data, and improves the generalization performance of the model through the domain knowledge of multiple label information of the same data, thereby improving the prediction accuracy of the multi-task classification model of the convolutional neural network MobileNetV2 In addition, the multi-task classification model using the lightweight convolutional neural network MobileNetV2 has fewer parameters and requires less computing power and memory size for training equipment, so that the prediction and classification tasks can be completed on mobile or embedded devices.
Owner:NANKAI UNIV

False wake-up audio acquisition method, system and device and storage medium

The invention discloses a false wake-up audio acquisition method, system and device and a storage medium. The method comprises the following steps: S100, receiving and storing environment audio; S200, receiving a plurality of wake-up logs, wherein the wake-up logs are configured to be a plurality of wake-up logs generated after a plurality of processors with voice recognition functions are used for executing wake-up operation on the environment audio of the same sound source in the same environment according to a preset wake-up triggering strategy; and S300, based on the same time period in the time sequence, according to a preset false wake-up strategy, combining the multiple wake-up logs into a data source for analysis and judgment, and according to a judgment result, extracting each segment of audio data causing false wake-up from the environment audio. By adopting the false wake-up audio obtained by the method, the audio duration is shortened, the audio occupied space is reduced, the training data volume is increased, and the precision of the recognition result is improved.
Owner:深聪半导体(江苏)有限公司 +1

NLP-based listed company financial fast message automatic generation method and system

The invention discloses an NLP-based listed company financial fast message automatic generation method and system in the technical field of natural language processing, and the method comprises the steps: carrying out the structural processing of an announcement file, and obtaining text analysis information; extracting financial data and text data from the text analysis information to obtain key financial information; and inputting the key financial information into the trained improved seq2seq model to generate a financial message. The method has the characteristics that the generated financial message text is rich in style, simple to maintain and use, good in text matching effect and the like.
Owner:JIANGSU SECURITIES
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products