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

3852 results about "Feature (machine learning)" patented technology

In machine learning and pattern recognition, a feature is an individual measurable property or characteristic of a phenomenon being observed. Choosing informative, discriminating and independent features is a crucial step for effective algorithms in pattern recognition, classification and regression. Features are usually numeric, but structural features such as strings and graphs are used in syntactic pattern recognition. The concept of "feature" is related to that of explanatory variable used in statistical techniques such as linear regression.

Cloth defect detection method and device based on machine learning

The invention discloses a cloth defect detection method based on machine learning. The method particularly comprises the following steps of image segmentation, image enhancement, image denoising, sub image feature extraction, defect point area segmentation, offline cloth learning, online cloth detection, and cloth defect point classification. In the offline cloth learning stage, a BP neural network is used to train standard image characteristic parameters, and a standard value is obtained. In the online cloth detection stage, the BP neural network is used to detect the feature parameters of the sub image. In the cloth defect point classification stage, a depth learning algorithm based on a convolutional neural network is used to classify cloth defects. The cloth defect detection method based on machine learning has a self-learning function and can meet the continuously-developing industry needs. The invention also provides a cloth defect detection device based on machine learning, which comprises an image acquisition unit, an image processing unit, a data communication unit and an action execution unit. The detection device can realize high-efficiency and accurate detection, and workers can be freed from heavy and tasteless physical labor.
Owner:FOSHAN NANHAI GUANGDONG TECH UNIV CNC EQUIP COOP INNOVATION INST +1

Vehicle type identification method based on support vector machine and used for earth inductor

The invention relates to a vehicle type identification method based on a support vector machine and used for an earth inductor. The vehicle type identification method includes the following steps: vehicle type waveform data which require to be identified are collected by the earth inductor; a plurality of numeralization features are extracted from waveforms, effective data are screened out, and the features are normalized; multilayer feature selection is performed according to the extracted features, and an optimal feature combination is picked out; a vehicle type classification algorithm based on the clustering support vector machine is built, and parameters in a classification function are optimized by adopting a particle swarm optimization algorithm; a binary tree classification mode is built, classifiers on all classification nodes are trained, and a complete classification decision tree is built; and earth induction waveforms of a vehicle type to be identified are input to obtain identification results of the vehicle type. The vehicle type identification method builds a waveform feature extraction and selection mode, simultaneously adopts the classification algorithm based on the support vector machine and the particle swarm optimization algorithm, greatly improves machine learning efficiency, and enables a machine to identify vehicle types rapidly and accurately.
Owner:TONGJI UNIV

Image feature-based image subject identification method

The invention discloses an image feature-based image subject identification method. According to the method, primary processing of an image is performed firstly, image features are deepened through image enhancement, and a foreground is roughly separated from a background; morphological processing is mainly used for extracting the image features, and a segmentation process is used for dividing the image into elements or target objects; the search on the image feature extraction is that the extracted image elements or target objects are represented in a numerical form suitable for subsequent processing of a computer, and finally features capable of being directly used for a classifier model generated by machine learning are formed; a distributed environment improves search efficiency and parallel computing capability; and the input image is identified through the method to obtain feature data, an image with highest similarity with the input image is searched for and output, and whether the two images are matched or not is judged. The invention provides a stable and feasible image search method. The semanteme of the image is deeply analyzed and learnt, so that the time and speed of a current search algorithm are shortened and increased, the network restrictions are avoided, and the universality is very high.
Owner:NANJING UNIV OF SCI & TECH

Crowdsourced and social media IP search and analytics platform with startup/industry partnerships and virtual incubator/accelerator including automated patent valuation system

InactiveUS20200250776A1Facilitate crowdsource functionalityEasy to spotFinancePublic key for secure communicationSocial mediaSoftware system
The analytics embodiment uses artificial intelligence (AI), machine learning, predictive analytics and computerized analytics. The analytics embodiment generates patent analytics and intelligence to analyze and value patents. The analytics embodiment can have smartphone applications. Smartphone applications can provide notifications, alerts, reminders, etc. Smartphone applications can include key features, including navigation tools. The analytics embodiment can have a social media feature, so users can connect with each other (as in Facebook, LinkedIn, etc.) and/or follow other users (like in Twitter, LinkedIn and Medium). The analytics embodiment will facilitate crowdsource functionality for validity, relevance and valuation.
A corresponding IP platform could allow startups to partner with big companies, to get mentorship and industry partnerships. This can help the startups with connections and eventually get acquired. This is common at incubators/accelerators. Therefore, we can add a virtual incubator/accelerator. The virtual incubator can have experts teach the startups and advise them on raising angel and venture capital money. Angel investors and venture capitalists can be guest lecturers and provide advice to startups. Teachers can help startups prepare paperwork to apply for government grants, such as small business and STTR/SBIR tech transfer government grants.
Furthermore, a corresponding IP platform can have an automated patent valuation system. A user can enter 1. An individual patent or 2. A pool of multiple patents. Then, the IP platform will automatically calculate the patent valuation. The analytics embodiment is the IP platform search and analytics software system. Therefore, this automated patent valuation system can be part of the analytics embodiment. The automated valuation can show multiple different patent valuations, based on different valuation techniques. Also, the automated valuation can show all of the individual patent valuations. This automated valuation can include the mean, median, and/or average patent valuations from multiple different valuations, based on different valuation techniques.
Owner:IPWE INC

Camera image quality evaluation method and device

ActiveCN105825500AComprehensive Image Perceptual FeaturesGood consistency of subjective feelingImage enhancementImage analysisImaging qualityLinearity
The invention provides a camera image quality evaluation method and device, and belongs to the technical field of image quality evaluation. According to the method, a natural scene statistical model and local resolution features formed by three sets of features are utilized. As for the first set of features, four features are extracted by utilizing the linear relation between free energy and a structure degradation model. As for the second set of features, the statistical features of natural images are utilized, and generalized Gaussian distribution is utilized to measure deviation of distortion images and the natural scene statistical model so as to extract four features. As for the third set of features, three features are extracted by utilizing discrete wavelet decomposition to measure resolution of the images. The previous two sets of features measure naturalness of the images and are different in focus, wherein the first set of feature consider local autoregression, and the second set of features consider global histogram statistics. Finally, the quality score of the camera images is obtained by the result of the evaluation algorithm through the machine learning tool of a support vector machine, and the quality of the camera images is evaluated according to the quality score.
Owner:JIANGSU VOCATIONAL COLLEGE OF BUSINESS

Data privacy protection method and system in machine learning

ActiveCN108717514AThe degree of ciphertext expansion is smallPracticalDigital data protectionTransmissionPlaintextData privacy protection
The invention relates to a data privacy protection method and system in machine learning. The method is characterized by comprising the following steps of: 1) selecting a to-be-used encryption algorithm and system parameters to generate a secret key; 2) encrypting original data to generate corresponding cyphertext data; 3) carrying out training and parameter adjustment on a to-be-used machine learning model by using the cyphertext data so as to obtain an optimal machine learning model; and 4) encrypting to-be-predicted or classified original data by using the secret key in the step 1) by adoption of the method in the step 2) , and inputting the to-be-predicted or classified original data into the optimal machine learning model to obtain a prediction or classification result. According to the method and system, an order preserving/distribution property preserving encryption algorithm and the machine learning model are combined, so that the original data and the machine learning model can be protected. The swelling degree of cyphertext output by the order preserving/distribution property preserving encryption algorithm is far lower than that of a full-homomorphic encryption algorithm, and certain distribution features in plaintext features can be kept, so that the machine learning is relatively high in efficiency and has relatively good expansibility.
Owner:RENMIN UNIVERSITY OF CHINA
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products