72results about How to "Find quickly and efficiently" patented technology

The invention discloses a spatial spectrum attention hyperspectral image classification method based on Octave convolution, and solves the problems of large class spacing, small different class spacing and low classification accuracy in the prior art. The scheme is as follows: inputting images to be classified and preprocessing data, dividing a training set and a test set, constructing an Octave convolutional neural network, determining a loss function of the Octave convolutional neural network, training and updating the Octave convolutional neural network, testing the data of the test set, and completing hyperspectral image classification. According to the method, Octave convolution operation is used to reinforce feature representation, and a spatial attention mechanism and a spectral attention mechanism are introduced, so that the network can more accurately find an area which is more beneficial to classification and contains more comprehensive and detailed information. The method ishigh in classification precision and strong in robustness, and can be applied to analysis and management of hyperspectral image data.

The invention provides a method and equipment used for playing channel programs. The method comprises steps that, a program guide including multiple channels and a channel list of direct broadcast windows are displayed, and each direct broadcast window is used for playing pictures of direct broadcast programs of corresponding channels of each direct broadcast window; a control signal which is inputted by a user and is used for controlling a focal point for moving in the program guide of the multiple channels is received, and one position of the focal point corresponds to one channel of the multiple channels; according to the control signal, the focal point is controlled for moving from a first position to a second position on the channel list; video flows of direct broadcast programs of the first channel are played at the direct play window of the first channel corresponding to the second position. The direct broadcast pictures or the video flows of each channel are presented in the channel list, so the user is facilitated to visually know about program contents of each channel when checking the channel list, the user is facilitated to rapidly and efficiently find favorite programs, and thereby user satisfaction can be improved.

The invention discloses a fungus detection fluorescent dyeing liquid and a use thereof. The fungus detection fluorescent dyeing liquid comprises an independent dyeing liquid A and an independent re-dyeing liquid B. The dyeing liquid A comprises, by weight, 0.02-0.043% of a fluorescent brightener, 1-10% of potassium hydroxide, 10-20% of dimethyl sulfoxide, 0-5% of glycerin and 65-87% of water. The re-dyeing liquid B comprises, by weight, 0.1-0.5% of Evans blue and 99.50-99.90% of water. The invention discloses a use of the fungus detection fluorescent dyeing liquid. The invention provides a novel fast fungus infection detection product. The product can be used for detecting various fungi which may exist in a fresh or frozen clinical sample and paraffin and ethylene glycol methacrylate-coated tissue.

The embodiment of the invention discloses a tire data management method and device. The tire data management method comprises the following steps: receiving a tire marker reported by a data transmitting module, a data interaction module or a data monitoring module, and tire use data of a tire corresponding to the tire marker, correspondingly storing the tire use data and the tire marker, and rapidly and efficiently finding corresponding marker information and tire use data when the data are called by a user. Due to unified management of related data of tires, data support is provided for big data analysis on the tires on the basis of the data, and meanwhile unified management on multiple tires and vehicles is also achieved.

The invention discloses a scientific and technological achievement converting system which comprises a service issuing module, a packaging module, a weight identifying module and a request submitting module. The service issuing module is used for making expert users issue technological service information. The packaging module is used for packaging basic information, technological qualification and user evaluation of experts into an expert information packet and packaging service content depiction, technological qualification, success cases, user evaluation and statistics data of the success cases into a service information packet. The weight identifying module is used for setting a weight distribution scheme for items in the expert information packet and the service information packet and calculating corresponding values for the technological service packet or the service information packet according to the weight distribution scheme. The request submitting module is used for submitting technological request information to enterprise users. The system has the advantages that expert information is integrated, a reasonable and effective organization displaying mode is provided, scientific and technological circles are promoted to the wide market, and the efficient conversion of scientific and technological achievements is promoted; enterprises can rapidly and effectively find suitable experts and evaluate the experts for reference to more people.

The invention relates to the technical field of data storage, in particular to a data storage method and device based on a block chain, equipment and a storage medium. The data storage method based onthe block chain comprises the following steps: obtaining transaction data of any client, generating a transaction data packet, carrying out statistics on the data volume in the transaction data packet, and creating a block corresponding to the client according to a statistical result; obtaining to-be-stored transaction data of at least one client, and generating a to-be-stored transaction data sequence from the to-be-stored transaction data according to the generated time sequence; extracting an identifier of a client carried in the to-be-stored transaction data, and writing the to-be-storedtransaction data sequence into a block corresponding to the client according to the identifier of the client; And endowing each block with a key value, and writing the to-be-stored transaction data inthe block into a key value database according to the key value. According to the technical scheme, the data storage efficiency is improved, the data is not easily tampered, and the source of the datacan be quickly and effectively found when the historical data is traced.

The invention provides a prediction method based on an orthogonal reverse bottleneck optimization algorithm, comprising the steps of loading a data set and performing standard processing on sample data; the penalty coefficients and kernel widths are optimized by using the orthogonal inverse bottleneck optimization algorithm, and a predictive support vector machine model is constructed. By adoptingorthogonal learning and reverse learning strategies to optimize the penalty factor and the kernel width on the basis of the traditional optimization algorithm of Styela bottlenecks, the invention caneffectively reduce the possibility of falling into local optimization in the optimization process, and obviously improve the quality of the solution and the convergence efficiency.

The invention relates to an indicating guide system of available parking spaces for planar car parks. The indicating guide system comprises a plurality of detecting units, a plurality of display units, an indicating guide server and a power supply element. The detecting units are fixed the parking spaces of a car park, respectively. The display units are fixed to the front edges of the parking spaces in the car park, respectively. The indicating guide server is connected with an entry barrier unit of the car park and the detecting units and display units in the parking spaces and is internally provided with a control program. The control program in the indicating guide server allows the display units at the front edges of the parking spaces to generate a light-on guide mode of dynamic merry-go-round action sequence, and each vehicle driver entering the car park is allowed to locate the available parking spaces fast.

The invention provides a metadata processing method and device, and the method comprises the steps: querying a physical address corresponding to a logic address of to-be-read data in pre-constructed LP metadata when a full-flash storage system carries out the data reading, and reading the to-be-read data based on the physical address; and when the full-flash storage system performs data correction, querying a logic address corresponding to the physical address of the to-be-corrected data in the pre-constructed PL metadata, and correcting the logic address. Wherein the pre-constructed LP metadata is used for indicating a mapping relationship from a logical address of the data to a physical address of the data, the LP metadata and the PL metadata are both of a B + tree data structure, and the pre-constructed PL metadata is used for indicating the mapping relationship from the physical address of the data to the logical address of the data. Based on the LP metadata and the PL metadata, the physical address and the logic address of the corresponding data can be quickly and effectively found, so that the data processing efficiency and performance of the all-flash storage system are remarkably improved.

The invention discloses a method for evaluating the influence of references in scientific documents. The method includes the steps of firstly, extracting the characteristic factors of the documents and the references, and calculating the weight of the characteristic factors; secondly, using probability measurement as influence scores; thirdly, training the weight of each characteristic factor by using a logistic regression model; finally, evaluating the influence evaluation model of the references. The method achieves quick, efficient and accurate searching of the most influential reference to the scientific documents, which facilitates researchers' searching of important references in documents.

The invention discloses a parking space reservation method and system. The method includes the following steps: acquiring a parking request of a user, wherein the parking request includes the locationinformation of a destination; searching for at least one parking lot near the destination according to the parking request; acquiring the parking space information of one parking lot, and sending theparking space information to a user terminal, wherein the parking space information includes at least one idle parking space; acquiring an idle parking space selected by the user, and locking the idle parking space selected by the user; generating a parking space reservation order according to the idle parking space selected by the user, and sending the parking space reservation order to the userterminal, wherein the parking space reservation order includes verification information; and acquiring verification information input by a user, and unlocking the idle parking space reserved by the user. Unified management and reservation of parking lots is realized. Vehicle owners can find a parking space quickly and efficiently while traveling. The problem of difficult parking is solved. The driving trip experience of users is improved. The situation in which the disorderly parking of vehicles affects the traffic order and the city appearance is indirectly improved.

The invention discloses an improved BHMC algorithm-based FenPtm alloy cluster structure optimization method. The method comprises the following steps of 1) initializing a population to obtain initial structures of an alloy cluster with a total atom number of N under the condition of various different ratios; 2) performing steepest descent local optimization on the initialized structures to obtain initialized local optimal structures; 3) performing disturbance processing on atoms of the local optimal structures, performing steepest descent local optimization, and storing lower-energy structures; 4) performing genetic local optimization processing on the structure with the lowest energy to obtain optimal element arrangement of the optimal structure in a current state; and 5) continuously repeating disturbance operator operation and genetic local optimization of the current optimal structures until a final optimal structure is searched. According to the method, the computer hardware overhead requirement is relatively low, the isomeride problem of the alloy cluster can be effectively handled, and the trapping in local optimum is avoided, so that the FenPtm alloy cluster structure optimization can be quickly and effectively realized.

The invention provides a method used for mining synonyms of search words of direct broadcasting rooms. The method comprises the steps of S1, obtaining information of clicked direct broadcasting roomscorresponding to to-be-queried search words and search words in a to-be-mined search word library in a preset time period; S2, calculating cosine similarity values between the to-be-queried search words and the search words in the to-be-mined search word library by utilizing the information obtained in the step S1; and S3, sorting the search words in the to-be-mined search word library from high to low according to the cosine similarity values, and selecting first N search words as synonyms of the to-be-queried search words, wherein N is a preset positive integer. The invention furthermore relates to related computer readable storage medium and electronic device.

The invention discloses a 3D guidance method comprising the following specific steps of 1) making a 3D scene model; 2) achieving a walking interactive function; 3) writing data into a device to form aguidance map; and 4) using the guidance map. The method achieves 3D space guidance. A user can quickly and efficiently find a route by moving in a real 3D environment. The method is easy to operate and has good guidance practicability and good experience.

The invention relates to the technical field of vehicle machine equipment and provides a vehicle, vehicle-mounted equipment and a personalized vehicle-mounted equipment interest point searching and displaying method thereof. The vehicle-mounted equipment receives interest point information input by a user; and acquires the interest point information, searches a target interest point conforming tothe interest point information in map data according to the interest point information, performs priority sorting on the searched target interest point by using the personality hobbies of the user, and displays the target interest point subjected to priority sorting to the user in a predetermined mode. The target interest points are screened and sorted by using the personalized hobbies of the user, so that an effective and applicable search result display mode can be provided for the user, the hobbies of the user can be conveniently, quickly and effectively found, a proper choice is made, andvisual and convenient use experience is brought to the user.

The invention discloses Beidou emergency rescue equipment which comprises a shell, a first fixing band and a second fixing band. A partition arranged in the shell has the shell divided into a left-half area and a right-half area, a gasbag, a gas generator, a first switch, a communication module, a positioning module, a first storage battery and a return spring are arranged in the right-half area, and a first cylinder, a moving rod, a retaining board and a second switch are arranged in the left-half area. The Beidou emergency rescue equipment has the advantages of convenience in use, high security and small size.

The invention discloses a layout planning area optimization method based on a neural network and a sequence pair, and the method comprises the steps: firstly constructing a layout planning database through employing a layout planning optimal solution, and solving the problem of insufficient machine learning data size for solving a layout planning problem in integrated circuit electronic design automation; meanwhile, the layout represented by a sequence pair in layout planning is converted into a classification problem in machine learning, finally, a neural network model is built through a multi-layer perceptron, training is conducted, and the trained model is obtained and used for predicting the optimal solution of the layout planning area; according to the method, the optimal scheme of the layout planning area can be quickly and effectively found.

The invention relates to an AGV path planning method, a storage medium and a terminal. The method comprises the steps of S1, obtaining a map, and setting the vertex angle of an obstacle in the map as a jump point; S2, carrying out path searching by utilizing an improved A* algorithm, and in the path searching process, if an expansion point comprises a jump point, moving the expansion point by one unit in the oblique direction of the jump point; S3, fitting the found path to obtain a fitting path; and S4, carrying out AGV path planning by using the fitting path. Compared with the prior art, the calculation amount can be reduced, the algorithm operation speed can be improved, and path planning can be rapidly and effectively realized.

A method for intelligently arranging and displaying jobs on a job search website includes comprehensive ranking for job search by weight and intelligent ranking of records of job search of job seekersand records of resume submitting of the job seekers. The method ranks the positions that best meet the needs of job seekers at the top, so that job seekers can find jobs quickly and efficiently without repeatedly viewing the jobs for which resumes have been submitted or which have been browsed on search result pages when looking for jobs for a continuous period of time. At the same time, the jobexposure amount of each enterprise can be maximized, and the problem that the exposure amount of some enterprise jobs is high and the jobs of some enterprises have no exposure amount is solved; job seekers can mark the jobs of interest which are displayed in the first place in the search of other job seekers, so that the others can find a job conveniently and the efficiency is high.

The invention discloses a full-text search intelligent sorting method for job-seeking website, including job searching, and intelligently sorting according to a comprehensive ranking of weights, a record of job searched by a job seeker, a job seekers' submitting record, which not only solves the problem of ranking the jobs that best meet the needs of job seeker, but also solves the problem that job seeker can find jobs quickly and efficiently without repeatedly viewing the submitted or browsed jobs when searching for jobs for a continuous period of time. At the same time, each enterprise position exposure can maximally increased, the high exposure of some enterprise positions and no exposure of some enterprise positions are reduced.