40results about How to "Many categories" patented technology

The invention provides a preparation method of spun-laced regenerated leather for an automotive interior surface. The preparation method comprises the following specific steps: (1) loosening and mixing fibers; (2) forming a net; (3) spreading the net; (4) pre-wetting; (5) spun-lacing; (6) drying; (7) carrying out vacuum plastic-absorption and molding. The requirements on material molding properties and mechanical properties by the automotive interior surface and industrial standards are sufficiently considered at the beginning of designing, and raw material ratios and a production process are reasonably optimized, so that collaboration of design, preparation and molding machining of a composite material is realized. Compared with traditional automotive interior surfaces including polyvinyl chloride, polyurethane and the like, the prepared automotive interior surface has the advantages of greenness and environmental friendliness, no odor, excellent elasticity, moderate strength, comfortable hand feeling, high cost performance and the like and is suitable for large-scale industrial production.

The invention relates to a multimedia resource recommendation method and device. The multimedia resource recommendation method comprises the steps that characteristics of multimedia resources to be recommended are respectively determined according to header information of multimedia resources to be recommended; the categories to which the multimedia resources to be recommended belong to are respectively determined; multimedia resource recommendation results are generated according to the categories to which the multimedia resources to be recommended belong to. By the adoption of the multimedia resource recommendation method, the categories of the multimedia resources in recommended information can be enriched, and the diversity of the multimedia resources is improved.

The invention discloses an entity identification method and device for a video scene, electronic equipment and a medium, and relates to the field of artificial intelligence. The method according to the invention comprises the steps of obtaining a to-be-processed target video and at least one target mode; extracting at least one target modal feature of the target video; determining a target entityidentification algorithm to be used from at least two candidate entity identification algorithms provided by the server according to the at least one target mode; wherein the at least two candidate entity identification algorithms are deployed in the server; and calling a target entity identification algorithm to perform entity identification on the at least one target modal feature to obtain a target entity included in the target video. According to the method and the device, identification of different modal entities of the target video is realized, the identification result accuracy is high, different service requirements can be met, and the universality is high.

The invention relates to the field of software systems for interior design and discloses a family management system used for information software of house interior design. The family management system comprises a browse module, a generation module and an addition module. The browse module comprises a classification display interface, a graphical interface and a search interface. The classification display interface is used for classifying and displaying objects of family data. The graphical interface is displayed through content selected by the classification display interface in the form of graphs. The search interface is displayed on the graphic interface by searching products. The classification display interface can be used for classification according to types of main items. Sub-items can be classified into the main items so that a user can distinguish classification information during family management. The graphical interface is used for displaying objects in the form of pictures and also displaying types, specifications and names below pictures. Therefore, the user can conveniently and directly select objects. The search interface is used for searching according to the search demand of the user so that search efficiency is improved.

The invention belongs to the technical field of entity linking, and particularly relates to a knowledge-driven query-based entity linking method which comprises the following steps of 1, identifying the named entity reference items in the user query statements on the basis of syntactic analysis; 2, based on incremental evidence mining, through an external knowledge source, carrying out informationexpansion on the entity reference items and a local knowledge base; and 3, performing link processing on the entity reference items by adopting an inference link algorithm. According to the method, the problems of lack of context, non-standard description and the like of the user query statements can be solved, the dependence on the local knowledge base is reduced, and the generation and discrimination of the candidate entities are accurately realized, so that the performance of the entity links is improved.

The invention relates to the field of intelligent sale, and especially relates to a monitoring method, device and system for goods of an intelligent sales container. The method includes the followingsteps: acquiring gravity change data of the goods in the intelligent sales container and image data of the goods before and after a gravity change; determining a set of hypothetical change conditionsof varieties and quantity of the goods according to the gravity change data; determining a difference in a category and the quantity of the goods before and after the gravity change according to the image data; and determining an actual change condition of the goods according to the set of the hypothetical change conditions and the difference in the category and the quantity of the goods. The monitoring method, device and system combine gravity and image analysis, overcome the defects that an existing algorithm cannot distinguish the goods with similar weight and similar appearance at the sametime, enable the accuracy rate of recognition of the goods by the intelligent sales container to be improved, further enrich types of the goods applicable to the intelligent sales container, and canmeet needs of different groups of people.

A remote sensing multispectral data semi-supervised labeling method based on self-learning relates to the field of data labeling, and comprises the following steps: acquiring remote sensing multispectral image data in a research area, determining category information of a to-be-classified target, and performing image fusion on the remote sensing multispectral image data; selecting a sample plot in the research area, recording category information of a to-be-classified target in the sample plot, determining a pixel corresponding relation between the to-be-classified target in the sample plot and the fused remote sensing multispectral image data in combination with the fused remote sensing multispectral image data, and obtaining pixel category information of the fused remote sensing multispectral image data; taking the seed point data as initial labeled data, and removing redundant information between wavebands by using a principal component analysis method; using the processed data with the labels to construct a classification model by adopting a random forest algorithm; the method comprises the following steps: classifying unlabeled data, removing abnormal points, and obtaining a self-labeling data set after multiple iterations; the method has the advantages of low requirement on manual labeling data volume, high precision, fast classification speed and strong anti-noise capability.

The invention discloses a forming process of a high light transmittance foamed elastic material. The forming process includes the following steps: (1) a to-be-foamed body is placed in a reaction kettle, supercritical gas is added to the reaction kettle, and the pressure in the reaction kettle is 10-60 Mpa; (2) the to-be-foamed body is made to saturate in the reaction kettle for a period of time until the to-be-foamed body is in the gas saturation state; and (3) an autoclave is subjected to pressure relief at the pressure relief speed of 0.1-1 Mpa / s, the gas in the to-be-foamed body in the gassaturation state is released gradually, and the high light transmittance foamed elastic material with forming holes at the diameter range of 0.1-3 mm. Compared with the prior art, the foamed elastic material obtained by the forming process has good light transmittance, and thus has widespread application in shoe sole lamplight function design.

The invention relates to a forestry IOT (Internet of Things) safety communication management method. The method includes the steps that an edge computing device performs the analysis of the data collected by an IOT device, wherein the edge computing device includes a communication module, an authentication module, a controller, and a strategy execution unit; the authentication module distributes dimension labels to the IOT device in a category related mode through a label unit so that each dimension label is corresponding to one type of IOT devices, wherein the dimension labels have public keycertificates, and the dimension labels are associated with the IOT devices and the category of the forestry IOT services; a controller stores and allocates a security rule, and the authentication module receives an authentication key from a TLS endpoint of the IOT device and determines whether the dimension label is allocated to the IOT device of the category or not; the strategy execution unit manages the communication of the IOT device according to the security rule. The method can perform communication management of the IOT device based on the security rule, thereby improving the securityof a system.

The invention is applicable to the fields of biotechnology and ophthalmic health, and provides a molecular marker of a myopia-related susceptibility gene as well as a detection primer group and application of the molecular marker. The detection primer group of the molecular marker comprises multiple PCR amplification upstream and downstream primer pairs of 37 SNP loci of 24 high myopia and pathological myopia-related susceptibility genes of human genome DNA, and a single base extension primer. According to the invention, sample nucleic acid mass spectrum detection can be carried out by using amultiplex PCR amplification technology, the 37 SNP loci can be detected at one time by double reaction holes, through correlation analysis of highly myopia susceptibility genes and genetic risks, a highly myopia genetic risk level assessment method special for Chinese population is provided by using a highly myopia genetic risk assessment system, and the method has the advantages of high detection flux, high efficiency, high accuracy, low cost and the like, and is suitable for large-scale population screening.

The present invention discloses a kind of world clock. It is characterized by that it includes dial, hour hand, minute hand, second hand and mechanism of the clock, and its mainly has the following functions: time indication, calendar, names of 24 main countries and their geographic positions and time difference, etc.

The invention provides a method for identifying androgen active substances in an environmental sample based on characteristic liquid fragment matching. The method comprises the following steps of selecting an activity warning structure related to androgen activity, classifying the 'activity warning structure', and establishing a correlation between the 'activity warning structure' and the mass spectrum characteristic fragment structure, extracting characteristic fragments capable of representing activity, and verifying the characteristic fragments through a standard substance and a mass spectrogram database, constructing an androgen active substance screening method based on characteristic fragments and applying the androgen active substance screening method. Compared with a traditional method, the method is higher in flux, the covered compound range is wider, more categories are achieved, a direct relation between toxicity characteristics and mass spectrum fragments is established, high-throughput screening of a large number of unknown compounds can be conducted, determining of the complete structure of the compounds is not needed, possible androgen active substances can be rapidly screened out according to matching of secondary fragments, the non-target screening efficiency is greatly improved, and then a new technical scheme is provided for recognition of key toxic substances in the environment.

The invention provides a vacuum forming technology and device for regenerated leather skin for an automobile instrument panel. The technology comprises the following steps: uniformly mixing thermoplastic synthetic fiber, which adopts a skin-core structure, with real leather fiber, and producing skin-core thermoplastic synthetic fiber mixed regenerated leather for the automobile instrument panel by a non-weaving technology; and during a vacuum forming process, realizing multi-station integrated forming of the regenerated leather skin for the automobile instrument panel by use of the vacuum forming device provided by the invention via the stations of regenerated leather roll spreading, accurate temperature-control heating, formed skin fixing and vacuum forming, regenerated leather scrap reclaiming and the like, thus obtaining the instrument panel skin with excellent service performance and forming performance. The technology provided by the invention is simple, is convenient to operate and has high practicability.

The invention relates to a hydroelectric industry digital object identifier processing method, device and system. The method comprises the following steps: receiving an identifier generation request;when the identifier generation request contains the domain code, determining a newly added meta-class sequence number of the domain code, and allocating a meta-class identifier to the hydropower industry digital object according to the newly added meta-class sequence number; when the identifier generation request contains a meta-class identifier, determining a newly added class sequence number ofthe meta-class identifier, and allocating a class identifier according to the newly added class sequence number; when the identifier generation request contains a class identifier, determining a newlyadded instance sequence number of the class identifier, and allocating an instance identifier according to the newly added instance sequence number; and returning at least one of the meta-class identifier, the class identifier and the instance identifier to the user terminal. By adopting the method, the identification requirement of the hydropower industry field on the digital object can be met.

The invention relates to a forestry internet of things network. The system comprises a background management client, a background management server, edge computing equipment and multiple internet of things equipment. The edge computing equipment comprises a communication module, an authentication module, a controller and a policy execution unit; the controller is used for storing and distributinga safety rule, wherein each safety rule is corresponding to a class of internet of things equipment; the authentication module distributes a dimension label to the internet of things equipment according to a way associated with a category through the label unit, so that each dimension label is corresponding to one class of internet of things equipment; the authentication module is further used forreceiving an authentication key from a TLS end point of the internet of things equipment and judges whether the dimension label is distributed to the internet of things equipment of this category; the policy execution unit is used for managing the communication of the internet of things equipment according to the safety rule. The bandwidth occupancy is reduced by adopting the edge computation technology, and the communication management can be performed on the internet of things equipment based on the safety rule, and the system security is improved.

The invention relates to the technical field of table vinegar processing, and particularly discloses a method for brewing enzyme vinegar by malt and rice sprouts through liquid fermentation. The method is characterized in that barley, rice and millet are selected, and sprouted grains of the barley, rice and millet are taken as main raw materials; haihong fruit juice, grape juice and hawthorn juiceare added; a corresponding process is improved on the basis of a table vinegar liquid fermentation process; and fruit juice is used for blending to prepare the enzyme vinegar brewed by the malt and rice sprouts through liquid fermentation. The enzyme vinegar prepared through the method is moderate in taste, rich in enzyme active ingredients and high in nutritional value, and has the effects of improving the antioxidant effect and the like.

The invention discloses a technical method of Sichuan embroidery silk, and relates to the technical field of silk printing and dyeing. The technical method comprises the following processes: matching color for the dyeing pigments, designing a composition, painting the basic color, coloring by the way of baking, hanging and steaming, water washing and surface dyeing, and finishing the formed products. The technical method has the advantages of unique manufacturing technology and high artistic value; the drawn patterns on silk cloth have the effects of shape, color, elegance, and charm, besides the patterns on silk cloth are natural and vivid, and have a soft hand-feeling and no hard feeling. The silk cloth obtained through the technical method is capable of generating complicated color glazing changes, has very abundant layers, does not fade after water washing or after a long time; after the patterns on the silk cloth being subjected to the technology processing of water washing and surface dying, the pattern colors are more transparent, vivid, and brighter, and defects of color block clots and hard hand-feeling existing in the conventional ways are overcome.

The invention provides an intelligent express delivery sorting machine. The intelligent express delivery sorting machine comprises a scanning device, a transport device, a sorting device and a sortingbasket frame, wherein the scanning device, the transport device, the sorting device and the sorting basket frame are sequentially arranged according to the station sequence. The intelligent express delivery sorting machine further comprises a control center which is in signal connection with each device and the sorting frame. The invention further provides a corresponding sorting method. By meansof the intelligent express delivery sorting machine, the sorting speed can be effectively increased. The intelligent express delivery sorting machine is particularly applicable to the distribution ofmedium and small express delivery packages. Sorting is convenient, express delivery sorting is rapid and accurate, the occupied space is small, the labor cost is reduced, express delivery sorting isin order, the sorting efficiency is high, and the error rate is very low. Moreover, because a large number of sorting baskets exist, a larger number of categories of express delivery sorting are available, and more convenience is brought. Meanwhile, by means of the express delivery sorting method, the positions of the sorting baskets can be confirmed in advance, so that a large number of packagesto be delivered in the same delivery station can be sorted into the sorting baskets as soon as possible, and the sorting baskets are carried away by transportation tools after being fully loaded.

The invention provides a man-machine verification method based on common knowledge, which improves the security of a verification code on the basis of ensuring user friendliness so as to effectively improve the accuracy of man-machine verification. The method comprises the following implementation steps: 1, constructing an image quantity list and a keyword list; 2, obtaining a verification code based on common knowledge; 3, generating a user verification interface of the man-machine verification system; 4, acquiring click behavior information of the user; 5, obtaining a man-machine verification result. According to the invention, the common knowledge is applied to the generation of the verification code by utilizing the difference of the understanding degrees of people and machines for thecommon knowledge, the difficulty of cracking the verification problem by an artificial intelligence algorithm is increased when the man-machine verification system performs man-machine verification on the user, and the security of the verification code is improved, so that the accuracy of the man-machine verification is effectively improved, and the risk that the Internet is maliciously attackedbased on an artificial intelligence algorithm in network scenes of login, registration and the like is reduced.

The invention belongs to the technical field of wine brewing, and discloses a complex fruit wine and a preparation method thereof. The complex fruit wine is prepared by adopting navel oranges, glorious oranges and blood oranges in sweet oranges as main raw materials, carrying out compound fermentation on three fruit juices of the navel oranges, the glorious oranges and the blood oranges and glutinous rice wine grains, and blending with a Sichuan-type solid-state Xiaoqu kaoliang spirit. According to the complex fruit wine and the preparation method thereof provided by the invention, sugar in the glutinous rice wine grains is used as the supplement for natural sugar of a sweet orange fruit wine, so that no extra exogenous sugar needs to be added; the glutinous rice wine grains and the sweetorange fruit juice are mixed for fermentation, and nutritional ingredients in fruits and sticky rice are fully utilized, so that the loss of the nutritional ingredients is reduced, and the product isrich in the components such as vitamin and amino acid; and according to the product characteristic requirements, the Sichuan-type solid-state Xiaoqu kaoliang spirit is added for blending, so that thecomplex fruit wine has the characteristics of the sweet orange fruit wine and the Sichuan-type solid-state Xiaoqu kaoliang spirit, the alcoholic strength of the product can reach to 5 percent to 40 percent, the product categories are rich, the flavor is unique, and the production cost is reduced by more than 10 percent.

The invention relates to a communication management method for an Internet of Things security system, which analyzes and processes data collected by an Internet of Things device related to security through an edge computing device, wherein the edge computing device comprises a communication module, a controller and a policy execution unit. The authentication server assigns dimension labels to theInternet of Things devices in a manner associated with a category through its label unit so that each dimension label corresponds to a class of Internet of Things devices. The dimension label has a public key certificate and is associated with a class of Internet of Things devices and Internet of Things security services. Internet of Things security services include fire services, access control services and video surveillance services. The authentication server receives the authentication key from the TLS endpoint of the Internet of Things device through the controller storing and distributing the security rules. The policy execution unit manages the communication of the IOT device according to the security rules. The invention improves the security of the security system by classifying the communication management of the internet of things equipment.

The invention provides an unmanned aerial vehicle operation scene modeling method. The method comprises the following steps of obtaining unmanned aerial vehicle operation scene information; defining composition elements of an unmanned aerial vehicle operation concept in a scene based on the obtained scene information; defining a structural relationship of the unmanned aerial vehicle operation concept in the scene based on the scene information and the composition element information; describing the composition elements and the structural relationship by utilizing a formalized language, and constructing a model of the unmanned aerial vehicle operation scene; and matching the model of the unmanned aerial vehicle operation scene with an unmanned aerial vehicle supervision requirement in the scene. According to the method, the information acquisition means is rich, the definition of the composition elements of the unmanned aerial vehicle operation concept is tightly combined with the requirement of specific operation risk assessment, various composition elements of the unmanned aerial vehicle operation scene are covered, the definition of the structural relationship of the unmanned aerial vehicle operation concept is tightly combined with the composition elements, the mutual relationship of the elements is constructed, classification is accurate and clear, and systematic support isprovided for specific operation risk assessment and unmanned aerial vehicle supervision.

The invention provides a convolutional neural network training method based on network images. A network car purchase website has a large number of rear-shot images of cars with category labels, the images can provide basic car brand information, but feature distribution of the images is different from that of car images shot by real monitoring, so that the images cannot be well applied to real monitoring scenes. Therefore, the recognition model is trained in advance through the network image and then applied to the monitoring image, and the image with the high score in the monitoring image is used as the training data to train the recognition model again, so that the effect of continuously promoting the recognition accuracy is achieved.

The invention provides a preparation method for needle-punched regenerated leather used as surface leather for automobile interior. The preparation method comprises the following concrete steps: (1) opening and mixing of fibers; (2) net formation; (3) net laying; (4) reinforcing through needle-punching; (5) drying and shaping; and (6) vacuum forming. In the beginning of the design of the needle-punched regenerated leather, requirements of the automobile interior surface leather on the molding performance and mechanical properties of materials and industrial standards are taken into consideration, and raw material proportioning and production process are reasonably optimized, so cooperation among the designing, preparation and molding of a composite material is realized. Compared with traditional polyvinyl chloride and polyurethane automobile interior surface leather, the automobile interior surface leather prepared in the invention has the advantages of environmental protection, no odor, excellent elasticity, moderate strength, comfortable handle, high cost performance, etc., and is suitable for large-scale industrial production.