Abstract: A vehicle damage detection method based on image analysis, an electronic device, and a storage medium are provided. In the vehicle damage detection method, query images are obtained by filtering received images through a pre-trained Single Shot MultiBox Detector (SSD) object detection model, and a feature vector of each of the query images is obtained by inputting each of the query images into a residual network. Target output data is obtained using a Transformer model, similar images of the query images are obtained by processing the target output data using a pre-trained similarity judgment model. Loss of a current vehicle damage assessment case is evaluated based on similar cases, and evaluated loss is outputted. By utilizing the vehicle damage detection method, effectiveness of the vehicle damage detection is improved, and automatic evaluation of a loss is achieved.

Abstract: Disclosed are a method and a user behavior data processing device, as well as a computer-readable storage medium. The user behavior data processing method includes: obtaining log information stored in a preset server at a fixed time, wherein, the log information comprises user behavior data which is captured and sent to the preset server by client; determining that whether row data satisfying a preset format exists in each row data of the log information, wherein, the row data is the data corresponding to each row of the log information; setting the row data satisfying the preset format as user behavior information; performing data statistics on the user behavior information according to a preset rule to obtain statistical information of the user behavior information.

Abstract: In a face swap method carried out by an electronic device, a first head image is segmented from a destination image. First facial landmarks and a first hair mask are obtained according to the first head image. A second head image is segmented from a source image. Second facial landmarks and a second hair mask are obtained according to the second head image. If at least one eye landmark in the second facial landmarks is covered by hair, the second head image and the second hair mask are processed and repaired so as to obtain a swapped-face image with eyes not covered by hair.

Abstract: The present application discloses a method and a device for customer resource acquisition, a terminal device and a storage medium. The method for customer resource acquisition includes: acquiring user's access request which includes a session identifier, user information and a progress identifier; determining whether the progress identifier is a completed identifier; determining whether the user's access request is a first access request corresponding to the session identifier; if the progress identifier is a completed identifier and the user's access request is not a first access request, determining a corresponding administration process based on the session identifier; storing the user information in an address space corresponding to the administration process, generating a first resource data based on all user information in the address space, uploading the first resource data to server, and destroying the administration process.

Abstract: The present disclosure discloses a remote interview signature agent matching method, an electronic device and a computer memory readable storage medium, said method comprising the following steps: Step 01, an user end inquiring whether there is an interview signature task, if so, then going to Step 02, otherwise ending the method; Step 02, the user end sending an interview signature request to an agent end, wherein the interview signature request includes user information and business information, the agent end confirming the user information and the business information; Step 03, the agent end allocating the interview signature request to a matching agent according to preset allocation strategy; Step 04, the matching agent performing matching check of user information and business information on the interview signature request, if matching is successful, connecting to incoming call user, otherwise returning to Step 03 for reallocation.

Abstract: Scoring information matching method and device, storage device and server. This scoring information matching method comprises: obtaining a target scoring information and a target scoring message which corresponds to the target scoring information; obtaining a first telephone number which sends out the target scoring message; obtaining the second telephone number which sends out the target scoring information; extracting a first identity number from the first telephone number; searching in preset service records for a service record of which an identity number is the same as the first identity number, a telephone number of a recipient of a corresponding scoring message is the same as the second telephone number, and a transmission time of the corresponding scoring message satisfies a preset condition; and determining the searched service record as a target service record that matches with the target scoring information.

Abstract: The present application relates to a data chain archiving method. The method includes: dividing an original data table in a database into partitions, and writing finished data in the partitions into a first temporary table according to a predetermined first processing mode by utilizing preset parallel threads; obtaining data with superior and subordinate association in the first temporary table by utilizing a single thread after the data is written in the first temporary table, and writing the data with the superior and subordinate association into a second temporary table; and dividing the second temporary table into preset partitions after all the data in the first temporary table is processed, archiving the data in the partitions of the second temporary table into a pre-established archive table in the database according to a predetermined second processing mode by utilizing preset parallel threads.

Abstract: A method of page displaying includes: obtaining page data of a current page of an application; the page data includes a screenshot and view identifiers and view names of a plurality of views; adding the plurality of view identifiers to a plurality of arrays having different levels according to a preset rule; building a multi-fork tree corresponding to the current page of the application using the array; generating hierarchical paths corresponding to the plurality of views according to the multi-fork tree, adding corresponding burial point frames to the corresponding views according to the hierarchical path, and transmitting the screenshot provided with burial point frames to the preset terminal, so that the preset terminal displays the screenshot with burial point frames.

Abstract: A machine learning based medical data classification method is provided. The method includes: a medical data classification request including medical record information is received; a preset medical term base is obtained, and word segmentation is performed on the medical record information according to medical terms in the medical term base to obtain multiple text vectors; features of the multiple text vectors are extracted to obtain multiple text vectors and corresponding feature dimension values; a target classifier is trained with multiple pieces of medical data, and the multiple text vectors and the corresponding feature dimension values are traversed and calculated; until a target node corresponding to the multiple text vectors is traversed, class probabilities corresponding to the multiple text vectors are calculated according to the target node, and a class result corresponding to the medical record information is obtained according to the class probabilities and is pushed to a terminal.

Abstract: This application discloses a data access authority management method, apparatus, terminal device and storage medium. The data access authority management method comprises obtaining report metadata in Tableau, the report metadata comprises report ID; creating folder data in Portal platform, the folder data comprises at least one folder, and the folder comprises folder ID; creating a correlation relationship between the report ID and the folder ID in Portal platform; obtaining user class authority configuration request entered by user, the user class authority configuration request comprises user class ID and target folder ID; performing user class authority configuration based on the user class authority configuration request in Portal platform, so as to enable the user class corresponding to the user class ID to have access authority to access report metadata which is corresponded to the report ID corresponding to the target folder ID.

Abstract: The present disclosure publishes a vulnerable plaque identification method. The method includes: receiving an angiocarpy image sent by a terminal device; transforming the angiocarpy image in an original Cartesian coordinate system into a polar coordinate system to form a polarization image; constructing a faster RCNN architecture and accomplishing a training; inputting the polarization image into the faster RCNN architecture accomplished the training to identify the polarization image, and outputting the image with the marked vulnerable plaques; feed backing the image with the marked vulnerable plaques to the terminal device. The present disclosure also publishes an application server and a computer readable medium. The vulnerable plaque identification method, the application server, and the computer readable medium can quickly and correctly conform the position of the vulnerable plaque.

Abstract: Disclosed are a speech recognition method, apparatus, computer device and storage medium. The method includes: performing a framing and an acoustic feature extraction of a speech-information-to-be-tested according to a default rule to obtain a frame-level speech feature sequence; dividing the frame-level speech feature sequence into n blocks sequentially; inputting all blocks into a preset bidirectional LSTM-RNN model parallelly to obtain an output result of the corresponding neuron in an output layer of the preset bidirectional LSTM-RNN model corresponding to the forward recognition result and backward recognition result of each block to obtain a speech recognition result of the speech-information-to-be-tested. The present application can improve the speech recognition effect significantly and reduce the time delay of the speech decoding effectively.

Abstract: An image processing method and an electronic device are provided, the method extracts a first object mask of a texture image and a second object mask of a to-be-optimized image. An image recognition model is used to obtain a first content matrix, a first texture matrix, a second content matrix, a second texture matrix, a first mask matrix, and a second mask matrix. A total loss of the to-be-optimized image is determined, and the total loss is minimized by adjusting a value of each pixel of the to-be-optimized image, thereby an optimized image is obtained. By utilizing the image processing method, quality of final image is improved.

Abstract: A method, a device, a terminal and a storage medium for acquiring slant value of slant image are provided. The method for acquiring slant value of a slant image comprises: analyzing the slant image and acquiring coordinate information of a plurality of boundary lines of the slant image; acquiring first slant values of the boundary lines by analysing and calculating the coordinate information; acquiring a correction value; calculating difference values between the first slant values and the correction value respectively; determining the first slant value corresponding to the minimum difference value as the slant value of the slant image. The technical solution of the present disclosure can uniquely determine a slant value of slant image.

Abstract: A method of querying a status of an electronic insurance policy includes: receiving an inputted policy number; querying an initial status and an operation flow set of the electronic insurance policy corresponding to the policy number when the policy number is a valid policy number; obtaining a mapped status subset according to a status total set to which the initial status belongs and a status that each operation flow in the operation flow set is mapped in the status total set; and displaying the initial status of the electronic insurance policy and the obtained status subset.

Abstract: A service line-based predication method and device, a storage medium and a terminal are provided. The method includes: when service predication is performed on a specified service line, acquiring a predication model corresponding to this specified service line, and input dimensions and output dimensions of this predication; acquiring predication data satisfying the input dimensions from a data warehouse; performing trend analysis on the predication data adopting Monte Carlo simulation and geometric Brownian motion through the predication model to obtain the predication values of the output dimensions; and calculating total task amount and manpower quantity required to be input of the specified service line within a specified period of time according to the predication values. The predication model is divided into an incoming call predication model and a calling predication model according to service types.

Abstract: A method of scanning website vulnerability comprising: reading a vulnerability scan task in a scan task pool; finding a website corresponding to the vulnerability scan task, acquiring access data of the website, and obtaining a popularity coefficient of the website according to the access data; acquiring historical vulnerability scan data and a vulnerability risk level table, and obtaining a security risk coefficient of the vulnerability scan task according to the historical vulnerability scan data and the vulnerability risk level table; acquiring update time data of the vulnerability scan task, and calculating a time coefficient of the vulnerability scan task according to the update time data; inputting the popularity coefficient, the security risk coefficient, and the time coefficient into a preset priority evaluation model for processing, and obtaining an execution priority weight of the vulnerability scan task; and executing vulnerability scan tasks in the scan task pool in descending order according to t

Abstract: A method of storing data, a device, a computer apparatus, and a storage medium relate to data processing technical field. The method of storing data comprising: acquiring call data obtained in a unit time period; preprocessing the call data based on a preset processing rule according to information carried by the call data; and storing a pre-processed call data, a start time and an end time of the unit time period simultaneously.

Abstract: A method for segmenting a list includes: acquiring user information and a list type of a to-be-segmented list through a list segmenting assembly; reading configuration information from a memory database, acquiring selectable dimensions from this configuration information according to the list type and the user information, and displaying the selectable dimensions; acquiring segmentation dimensions selected by a user and segmentation ranges corresponding to the segmentation dimensions through the list segmenting assembly, transmitting the list type, the segmentation dimensions selected by the user and corresponding segmentation ranges thereof to a big data platform to segment the to-be-segmented list according to the segmentation dimensions and the corresponding segmentation ranges thereof through the big data platform to obtain table results; and acquiring the table results through the list segmenting assembly, and generating follow-up tasks according to the table results and issuing to agents.

Abstract: An electronic device for identity verification includes a memory and a processor; the system of identity verification is stored in the memory, and executed by the processor to implement: after receiving current voice data of a target user, carrying out framing processing on the current voice data according to preset framing parameters to obtain multiple voice frames; extracting preset types of acoustic features in all the voice frames by using a predetermined filter, and generating multiple observed feature units corresponding to the current voice data according to the extracted acoustic features; pairwise coupling all the observed feature units with pre-stored observed feature units respectively to obtain multiple groups of coupled observed feature units; inputting the multiple groups of coupled observed feature units into a preset type of identity verification model generated by pre-training to carry out the identity verification on the target user.