63results about How to "Reduce filter range" patented technology

The invention discloses an online shopping method, device and system based on an image recognition technology. The online shopping method comprises the following steps of obtaining an image of an article to be purchased, wherein the image of the article to be purchased is sent by a client-side; extracting the characteristic of the image of the article to be purchased; matching the extracted characteristic and characteristics of the images of all articles in all shopping websites through cloud computing; seeking out articles within the set matching degree range and the related information of the articles; ranking the articles which are sought out according to the ranking condition; sending the ranked articles and the related information of the articles to the client-side. According to the online shopping method, device and system based on the image recognition technology, the articles to be purchased are sought out in an image characteristic matching mode, the articles which are located on different shopping platforms and meet the requirement of the matching degree are listed, users can compare and select the articles to be purchased in the shopping platforms at different areas, the articles to be purchased can be displayed for the users conveniently and rapidly, and the problem that due to manual classification and selection, the articles to be purchased are difficult to find out is avoided.

The invention discloses a geographic information data filtering method, electronic equipment, storage medium and system based on meshing. The method includes the steps of meshing a map, dividing the map into several mesh units along the longitude and the latitude; linking the geographic information of each region in the map to POI information points in the map; obtaining longitude coordinates and latitude coordinates of all the boundary points in a region to be analyzed, dividing the mesh units belonging to the region to be analyzed into the mesh units inside the boundary of the region to be analyzed and the mesh units on the boundary of the region to be analyzed; obtaining the first geographic information contained in the POI information points in the mesh units inside the boundary of the region to be analyzed; judging whether the POI information points inside the mesh units on the boundary of the region to be analyzed are in the region to be analyzed, and if so, obtaining the second geographic information contained in the POI information points, otherwise, excluding the POI information points; integrating the first geographic information with the second geographic information, and obtaining the complete geographic information of the region to be analyzed. According to the geographic information data filtering method, the range of screening is narrowed, and the efficiency of screening is improved.

The invention relates to a safe crop production technique in the polluted environment, in particular to a method for selecting crop varieties with low heavy metal accumulation. A crop is cultured in heavy metal-processed soil by way of pot experiment, the temperature is 5 DEG C to 40 DEG C, the soil moisture content is 40 percent to 90 percent of the field moisture capacity, the crop is cultured under the natural light for 50 to 170 days, plants are harvested, and the heavy metal content of edible parts is measured; the selected crop seeds, the edible parts of which have the heavy metal content lower than 30 percent of the heavy metal content of the edible parts of the conventional crop variety, are directly dibbled in the polluted field soil, three to six seeds are planted in each hole, the distance between holes is 10cm to 60cm, the distance between ridges is 10cm to 60cm, ripened plants are harvested, the heavy metal content of edible parts is measured, and if both the heavy metal enrichment coefficient and heavy metal transfer coefficient of the crop are less than 1.0, the crop is a crop with low heavy metal accumulation. The invention has the advantages of economy, high economic safety, high applicability and simple use method.

The invention relates to a rapid positioning method of cotton single locus quality gene in chromosome, and belongs to the technical field of biology. The method includes establishing a near-isogenic mixed pool by using land-sea F2 positioning group, screening the gene to obtain a primer with polymorphism, detecting the DNA of 9 single strains, P-1, P-2 and F1 in a recessive gene mixed pool, recording the banding patterns of the DNA, calculating the individual number among the 9 single strains in the recessive mixed pool which has the same banding pattern with the recessive parent P-1, calculating the coincidence rate, selecting the polymorphic primer having the banding pattern coincidence rate over 60% with P-1 and having dominant or codominant F1 banding pattern, detecting the F2 segregation population by using the small amount of the primer, calculating the banding pattern, and analyzing by software to obtain the SSR primer concatenated with the quality gene. Displayed by the embodiment, the method makes the range of screening target concatenated primer by using F2 positioning group reduce over 90%, can lock the SSR primer concatenated with the target quality gene rapidly, can reduce the workload of the gene positioning effectively, and can improve the positioning efficiency of the quality gene in the chromosome greatly.

The invention discloses an oligonucleotide library classification and assessment method based on capillary zone electrophoresis, and belongs to the field of creature isolation analysis. The method comprises the following steps of: step one, carrying out capillary zone electrophoresis on an oligonucleotide library, and obtaining a secondary library within an oligonucleotide library electrophoresis time range according to transfer time slicing collection; and step two, respectively mixing each secondary library with a homogeneous target molecule, carrying out capillary zone electrophoresis, and comparing the strong or weak of each secondary library and target molecule interaction, thus obtaining the strongest secondary library of the target molecule combining capacity. The classification and assessment method can be used for realizing the fractionation of complicated constituent oligonucleotide library, and obtaining the strongest secondary library of the target molecule combining capacity; and the strongest secondary library of the combining capacity is utilized as a next CE (capillary electrophoresis)-SELEX(systematic evolution of ligands by exponential enrichment) technical screening library, the screening range is reduced, and the screening period of an adaptation body is shortened.

The invention provides a WKNN indoor positioning method based on spatial feature partitioning and front point constraint, and the method comprises the following steps: 1, dividing a region into subregions; 2, extracting a subregion identification sequence; 3, carrying out the subregion discrimination; 4, calculating a signal spatial distance; 5, obtaining position estimation through weighting. Through the steps 1-5, the method provided by the invention is executed, and enables a large target region to be divided into a plurality of subregions according to the spatial features. Meanwhile, a combined partitioning criterion of the recognition sequence and Euclidean distance is introduced, thereby solving a problem that a fingerprint database cannot achieve the coverage of the whole region. The method gives consideration to the spatial constraint relation of the positions of a pedestrian at adjacent moments, thereby reducing the screening range of final reference points, and greatly improving the precision of position estimation.

The invention discloses a topic model analysis and user short-long interest-based recommendation method. The method comprises the steps of performing hidden topic training on an activity text set to obtain a trained model, and performing calculation to obtain type characteristic topic distribution of all types; inputting short-term and long-term activity texts of a user to the trained model to obtain short-term and long-term activity topic eigenvectors of the user, and in combination with behavior weights and a time decay function, obtaining short and long interest vectors of the user; and according to the long interest vector of the user and the type characteristic topic distribution of all the types, obtaining a cosine similarity value of a long interest, selecting the type of TopM withthe highest cosine similarity value of the long interest, according to the short interest vector of the user and activities in the type of the TopM, obtaining a cosine similarity value of the short interest, and according to the cosine similarity value of the short interest, obtaining recommended activities of an activity platform for the user. The screening range is reduced when the to-be-recommended activities are selected, so that the recommendation calculation time is shortened and the recommendation accuracy can be improved.

The invention discloses an evaluation recording method based on credible friends. The evaluation recording method based on the credible friends comprises the steps that property information of the friends of a user is acquired based on property information of the user; purchase records of the friends and evaluation information of commodities purchased by the friends are acquired; display behavior options, corresponding to different display modes, for the user to select are set in a system according to system score display of the evaluation information; the user selects different display behaviors according to a user demand, and the evaluation information of the commodities purchased by all the friends is subjected to system module display according to the display behaviors selected by the user; and modules include a prior search module, a prior recommendation module and a prior activity check and selection module. Through the method, the commodities with high scores can be recommended or displayed to the user first through the purchase evaluation record information of the credible friends, screening pressure of the user is relieved, the credibility of commodity display is enhanced, and the success rate of purchase is increased.

The invention discloses a rapid separation method of antagonistic bacteria against pathogens of konjac soft rot disease, which comprises the following steps: A) taking konjac tissues infected by the soft rot disease; B) inoculating the tissues at the interface of diseased and healthy tissues onto a nutrient agar culture medium after carrying out surface disinfection; C) streaking the bacteria grown out on the sides of the tissues, and selecting a single colony for purification and preservation; D) adopting a distinctive Erwinia crystal violet polypectin selective culture medium (CVP) to screen the pathogens of the konjac soft rot disease; E) adopting the point connection method to carry out primary screening of antagonistic bacteria; and F) adopting the oxford cup method to carry out secondary screening of the antagonistic bacteria. The method has simple operation and easy operation, and can rapidly and effectively screen out the antagonistic bacteria against the pathogens of the konjac soft rot disease and simultaneously provide a concept for rapid screening of the antagonistic bacteria against other crop diseases.

The invention relates to a member right and interest management method and device, computer equipment and a readable storage medium. The method comprises the following steps: receiving a member rightsand interests display request sent by a client for a target account, wherein the member rights and interests display request carries a member system identifier; according to the member system identifier, obtaining a plurality of rights and interests type lists under a member system corresponding to the member system identifier, and according to the first account information of the target account,determining a target rights and interests type list from the rights and interests type lists, the target rights and interests type list comprising at least one candidate rights and interests; and based on the second account information of the target account, screening effective rights and interests matched with the second account information from the candidate rights and interests of the target right and interest type list, generating an effective right and interest list according to the effective rights and interests, and sending the effective right and interest list to the user side. By adopting the method, the screening efficiency of effective rights and interests can be improved.

The invention relates to a virtual resource acquisition method and device based on a neural network, a computer device and a storage medium. The method comprises the following steps: receiving a resource acquisition request sent by a terminal; wherein the resource acquisition request carries a target attribute value; obtaining resource factors of the plurality of virtual resources, and inputting the target attribute value and the resource factor of each virtual resource into a regulation and control model to obtain a resource label of each virtual resource; selecting a corresponding expert model according to the resource annotations, and inputting the resource factors into the expert model to obtain predicted attribute values of the corresponding virtual resources; screening virtual resources with the prediction attribute values matched with the target attribute values, and recording the virtual resources as target resources; determining factor characteristics of the target resource according to the expert model corresponding to the target resource; and obtaining resource information of the target resource, and returning the resource information and the factor characteristics to the terminal. By adopting the method, the virtual resource acquisition efficiency can be improved.

The invention discloses a method and device for marking a to-be-shared video, a camera and a smart phone, and relates to the field of video processing. The method includes the steps: acquiring an acquired video file; identifying the video file to determine whether there is a joy factor or not; marking a video frame that is recognized to have a joy factor if there is the joy factor; cutting a videofile of a predetermined length of time forwards and / backwards based on the marked video frame; storing the cut video file in a preset storage space. The method can automatically screen out a video worthy of being shared from the captured video, saves the time cost of a user in searching and improves the user experience.

An information induction analysis method and an information induction analysis device belong to the technical field of IT application and are used for solving the technical problem that the existing information induction analysis method is long in analysis period, large in human cost input and low in recall ratio. The technical essential is as follows: S1, extracting characteristic data according to the type of input data information provided by a user, loading and releasing a corresponding sample model library; S2, dynamically loading a detection engine to track and recognize the characteristic data, and searching suspected objects in the sample model library; S3, presetting a correlation condition so as to obtain relevance among the suspected objects to determine the similarity of each suspected object; and S4, obtaining the output data information. The effect of the method and the device is as follows: the manual information filtering range is reduced, and the mass data information which needs to read is converted into effective information related to the object so as to improve the extraction efficiency of the useful information.

The invention provides a method for identifying reality and purity of pepper male sterile three-line mating hybrids based on InDel (insertion-deletion) molecular markers. The method includes the steps: culturing the pepper hybrids and parents to a bud and seedling stage; extracting DNA (deoxyribonucleic acid); screening the hybrids and the parents by core primers. Primers of co-dominant markers can serve as molecular markers for identifying the reality of the hybrids, and co-dominant markers of the pepper hybrids and the parents continues to be screened by combining a secondary core primer library until the co-dominant markers are screened out if the co-dominant markers of the pepper hybrids and the parents cannot be screened out in a core primer library. According to the method, a systemcapable of rapidly and simultaneously identifying the reality and the purity of the pepper male sterile three-line mating hybrids is built and cannot be limited by time and seasons, indoor rapid identification of the reality and the purity of the hybrids is achieved, and molecular biology bases are provided for protection and market supervision of the pepper male sterile three-line mating hybrids.

The invention discloses an RFID (Radio Frequency Identification) anti-collision method based on a self-adaptive dividing mechanism. A required stack (req_stack), a divided marker bit (Div_bit) and a divided number (Div_num) are additionally added to a reader; further, the number of times of the reader to inquire a tag is reduced; meanwhile, after the tag is identified, the operation directly enters a silent state, the return to a root node for inquiry is not needed, and the number of inquiry times is further reduced; a collision bit (col-bit) is arranged at a tag end for recording the bits of the collision, and the transmission quantity in the communication process is effectively reduced; a state marker (state) is arranged at the tag end, so that the screening range of the tag is effectively reduced, and further, the number of the inquiry times is reduced; and the reader does not need to send a complete ID number, so that the data transmission quantity of an RFID system is effectively reduced. The method provided by the invention can be widely applied to the field of RFID.

The invention discloses an external point elimination method for a video stream based on an improved RANSAC method, and the method comprises the steps: extracting feature points, tracking, and updating the tracking frequency; counting and judging the continuous tracking times of the feature points, randomly selecting minimum data capable of calculating model parameters from the classified model construction point set to perform model fitting, and calculating a basic matrix model; judging all the other feature points by using the basic matrix model, and counting internal and external point information in all the feature points; counting the number of the inner points classified in the model construction point set, comparing the number with the previous model, and updating the model; obtaining the maximum number of iterations; after one iteration is completed, adding 1 to the number of iterations; jumping out after reaching an upper limit, and outputting interior point information; marking the input feature points, if the feature points are judged to be inner points, successfully tracking and retaining the feature points, and if the feature points are judged to be outer points, tracking unsuccessfully and clearing the feature points. The precision and the operation speed of the original algorithm are improved at the same time.

The invention provides a networking method for vehicle area communication. The method comprises steps of when the existence of the networking target vehicle is detected, starting short-distance networking equipment on the vehicle; and preparing networking communication, updating the network member relationship table after the vehicle networking succeeds, broadcasting and informing all vehicles inthe local area network, periodically broadcasting the own position and motion information of the networking target vehicle through the network, and judging whether networking is carried out or not according to whether the networking target vehicle has an adjacent vehicle or not and whether the networking target vehicle receives the information in the network or not. According to the invention, thevehicle-mounted sensor information is fused, the networking target capture accuracy is improved, the information does not need to be periodically broadcasted to the public network, the bandwidth pressure on the common network during vehicle networking is relieved, the screening range of the target vehicles is reduced, and the networking rapidity is improved.

The present invention provides a method for enriching and separating efficient phosphorus-accumulating bacteria in active sludge of a biological membrane reactor, and relates to the technical field ofphosphorus-accumulating bacteria domestication and enrichment. The method comprises pretreatment of the sludge, in vitro domestication and enrichment of the bacteria and separation of the bacteria. By the method, the interference of mixed bacteria on phosphorus-accumulating bacteria nutrition competition can be reduced, and the species richness of the separated phosphorus-accumulating bacteria isimproved; a dilution solution with the OD600 value of 0.1-0.5 is used for four-region streak inoculation, so that the workload can be significantly reduced; and four-region streaking and chromogenicscreening are carried out synchronously, and by two rounds of chromogenic screening, screening of the phosphorus-accumulating bacteria is more accurate, the screening range is narrowed, the workload is reduced and the accuracy of screening is improved.

The invention discloses a suspicious sample identification method based on power transmission line channel image continuous alarm. The method includes: using an image analysis model to identify an image acquired by certain visual inspection equipment in real time; and judging the identification result by using a continuous alarm model, and determining whether the continuous alarm is triggered or not: marking the image identified as having no hidden danger as a suspicious missed alarm sample if the continuous alarm is triggered, and marking the image identified as having hidden danger as a suspicious missed alarm sample if the continuous alarm is not triggered. According to the invention, continuous alarm identification and judgment is carried out on relatively fixed hidden dangers of staying areas such as machinery, smoke and foreign matters with limited operation ranges such as tower cranes and excavators, whether to trigger continuous alarm or not is judged in combination with imagedata continuously acquired by visual equipment, abnormal samples are marked, and suspicious missed alarm or suspicious false alarm samples are determined. According to the invention, the intelligent level of power transmission line operation inspection is improved by improving the recognition accuracy of the image analysis model.

The invention relates to a microorganism collection method in an ecological environment survey. The microorganism for the microorganism collection method is arbuscular mycorrhizal fungi. The collection method comprises the following steps: manufacturing a collection culture bag, preparing experimental matrixes, and packaging the experimental matrixes to the culture bag and sealing, and launching the culture bag. By the microorganism collection method in the ecological environment survey, the culture bag can be enriched in spore and mycelium, and the screening range is narrowed greatly and the workload is reduced.

The invention provides an urban rail line level close-contact person cross infection risk identification method and device. Based on behavioral science and data science, a risk exposure line station walking time probability distribution model is constructed from microscopic travel behavior analysis; from the perspective of an iterable risk period, the range of a close-contact person with the possibility of cross infection is continuously reduced; model solving is performed by utilizing an EM algorithm, cross infection risk exposure train risk probability and close-contact person cross infection risk probability are evaluated, and a close-contact person risk level evaluation model is established; and an urban rail risk level decision support and information query device is formed. Accordingto the invention, the method is an important means for improving suspected case screening precision and efficiency, provides a convenient and efficient query tool for the public to judge whether thepublic is a close-contact person, fills the blank of risk identification of the close-contact person in the field of public transportation, and has important decision support significance for public emotion guidance and emergency resource deployment in emergencies.

The invention discloses a low-voltage electricity stealing customer and electricity stealing means determination method based on big data analysis, and mainly relates to the field of electricity stealing prevention. The method comprises the following steps: 1) based on physical characteristic analysis of a processed electricity stealing case, establishing a plurality of electricity stealing early warning models and electricity stealing means judgment models from different analysis dimensions of electric quantity abrupt change, current abnormity and strong magnetic interference events; 2) performing big data analysis on the electric parameters, the electric energy meter events and the work order information data through a specified electricity larceny early warning model, determining suspected electricity larceny preliminary details of the electricity larceny means in the corresponding electricity larceny early warning mode, and screening out suspected electricity larceny clients; 3) comparing the screened suspected electricity larceny preliminary details by using an electricity larceny means judgment model, screening out electricity larceny data, and determining an electricity larceny customer and an electricity larceny means. The invention has the beneficial effects that the accuracy of judging electricity stealing customers and electricity stealing means can be improved, the judgment efficiency is greatly improved, and the loss of electric quantity is avoided.

The invention relates to an information retrieval method and apparatus, a computer device and a storage medium. The method comprises the steps of obtaining a target problem; obtaining preset retrievalpaths and calling sequences corresponding to the retrieval paths; according to the calling sequences, obtaining the retrieval path of the first sequence, and retrieving the target problem by utilizing the obtained retrieval path to obtain a temporary information library, wherein the temporary information library comprises multiple information pairs, and the information pairs comprise retrieval results; calculating the similarity between the retrieval results in the information pairs and the target problem; detecting whether the information pair meeting a condition that the similarity betweenthe information pair and the target problem reaches a threshold exists or not; if not, obtaining the retrieval path of the next sequence according to the calling sequences, and returning to the step of retrieving the target problem by utilizing the obtained retrieval path; and if yes, extracting the retrieval result with the similarity reaching the threshold to serve as a target result, and displaying the target result. By adopting the method, the information retrieval efficiency and accuracy can be improved.

The invention discloses a log extraction method and related equipment, and the log extraction method comprises the steps: carrying out the slice storage of a terminal log, and the attribute of each slice file comprising the file generation time; after a log request for extracting a certain time period sent by a server is received, extracting at least one slice file corresponding to the time period; generating a log file based on the at least one slice file; and uploading the log file to the server. The log extraction method is automatically operated and does not need to be manually grabbed. Moreover, the log data of the terminal is stored in the form of multiple slice files, so that the screening range is reduced, the log screening time is shortened, the log capture efficiency is improved,and the terminal problem checking time is shortened. The method is widely applied to remote extraction of the terminal log.

The invention discloses a diet recommendation method based on big data analysis, relates to the technical field of diet intelligent recommendation, and solves the technical problems of inaccurate dietrecommendation and small popularization range in an existing scheme. A range analysis module is arranged, a diet activity curve is drawn through diet records and search records of a user, the activity range of the user is obtained on the basis of the diet activity curve, diet is recommended to the user on the basis of the activity range, the screening range can be narrowed, and it is guaranteed that the recommended diet better conforms to the taste of the user. According to the invention, by means of a dish recommendation module, the setting is combined with the native place and the dwellingduration of the user in the current dwelling place to recommend dishes to the user, so that the dishes can be pushed to the user more accurately; in addition, a data interaction module is set, and thedata interaction module carries out the deep pushing of the dish screening range and the dish screening graph through the relation between a contact person and a user, thereby facilitating the distribution of excellent dishes to proper users.

The invention relates to a method for determining cross-allergy-acting epitopes related to eggs. The method comprises the following steps: S1, retrieving amino acid sequences of main allergens of eggs through NCBI; S2, manually inputting an amino acid sequence of a main allergen of a certain food, comparing the similarity of the input allergen sequence and an allergen sequence of eggs through a tool Align in a page Uniprot, and predicting whether cross-allergy is present or not; S3, in the presence of cross-allergy, predicting a secondary structure and antigenic epitopes of the egg allergen presenting cross-allergy through a tool Protean in software DNAstar; S4, based on the prediction in S3, analyzing similar sequences of the egg allergens with cross-allergy and allergen proteins by using a tool web page Blastp; S5, preparing a corresponding monoclonal or polyclonal antibodies with the proteins presenting cross-allergic reaction in the step S2, identifying cross-allergens by using an immunoblotting method and the ELISA method, further synthesizing the similar sequence determined in the step S4, and identifying whether the similar sequence is a reaction epitope or not by using serology.

The invention belongs to the technical field of biological information, and discloses a method for predicting functional genes for regulating and controlling yeast autophagy, which comprises the following steps: S1, carrying out time sequence transcriptome and proteome analysis on normal yeast samples before and after starvation induction treatment and yeast samples from which autophagy key regulation and control genes are knocked out; s2, screening genes with interaction, and recording the genes as a set A; s3, extracting transcriptional expression change and protein expression change of each gene in the set A, and the interaction degree with the known autophagy gene as features; s4, establishing and training a prediction model; and S5, optimizing the model, and scoring the genes in the set A so as to predict the functional genes for regulating and controlling yeast autophagy. By utilizing the prediction method provided by the invention, the gene screening range can be effectively narrowed before verification of a traditional experimental method, so that the workload of experimental verification is reduced, and the functional gene for regulating and controlling yeast autophagy is accurately predicted.

The invention discloses a method for integrating multi-omics data to presume key protein kinases in cell transdifferentiation. The method comprises the steps of: preparing a cell sample during whole cell transdifferentiation, dividing the cell sample into a transcriptome group and a protein phosphorylation group, and conducting omics tests separately to obtain corresponding original data of each group; pre-processing the original data, screening the protein kinases expressing significant differences through a transcriptome, and screening the protein kinases with activity significant differences and regulation significant differences through a phosphoproteome; and integrating all protein kinases with significant differences, and conducting comprehensive measure to obtain key protein kinases. The method of the invention can provide small-scale candidate kinases for the research of cell transdifferentiation mechanism, provide a guiding direction for the research of the cell transdifferentiation mechanism, greatly narrow the screening range, improve the research efficiency and the accuracy of the experiment, shorten the research and development cycle, increase the success rate of experiments, and reduce the cost of research on cell transdifferentiation mechanisms.

The invention relates to the technical field of automobile structure design, in particular to a design method and a design device for a door protection plate pushing cushion block. The method comprises the following steps: S1, determining a chest compression amount target value and a pelvic resultant force required value; s2, constructing a two-dimensional relation curve among the chest compression amount, the pelvic resultant force and the pushing cushion block span point distance; s3, determining a theoretical value range of the cross-point distance of the pushing cushion block; s4, determining a theoretical value range of the action area of the pushing cushion block; s5, constructing a three-dimensional relation curved surface of the chest compression amount, the pushing cushion block span point distance and the pushing cushion block action area; and S6, selecting a proper design value of the cross-point distance of the pushing cushion block and a design value of the action area of the pushing cushion block, and determining the material type of the pushing cushion block. The design method disclosed by the invention is simple and efficient, and the design parameters of the pushing cushion block can be quickly obtained on the premise of meeting the Chinese new vehicle evaluation regulations and vehicle model project development targets.