36results about How to "Rich generation method" patented technology

The invention relates to a target detection model training method and device, a computer readable storage medium and computer equipment, and the method comprises the steps: obtaining a feature map ofa sample image during training, and determining an initial detection box in the feature map according to a preset rotation angle, a preset scale and a preset target aspect ratio; adjusting the position of each initial detection frame to obtain the position information of the prediction detection frame, and adjusting the network parameters of the regression network according to the position information and the real position information in the annotation information of the sample image; predicting the prediction probability of the target corresponding to each preset category according to the target detection area determined by the position information of the prediction detection frame; and adjusting the network parameters of the classification network according to the real category information in the annotation information of the sample image and the prediction probability, and then obtaining a target detection model used for performing target detection on the image. According to the scheme provided by the invention, the target detection model can identify the rotation angle of the target in the image, and the positioned target detection frame is more accurate.

The invention relates to an integration method of multiple prediction results of power load probability density, and belongs to the technical field of power system analysis. According to the method, aplurality of probability density or quantile probability prediction models are obtained through training of a plurality of three-class regression models set by different hyper-parameters, and the output of the quantile prediction models is converted into a probability density model conforming to Gaussian distribution through Gaussian distribution assumption of loads and a least square method. A probability density prediction integration method is adopted, a probability density prediction optimal integration model is constructed based on the trained probability density prediction model and result, the weights of different probability density prediction methods are determined, and therefore the probability loss of the continuous grade of the final integration prediction model is minimum. The method is finally converted into a quadratic programming problem, the global optimal integrated weight is quickly searched by utilizing mature commercial software, the probability density short-termload prediction precision is improved, and the dispatching operation cost of the power system is further reduced.

The invention relates to a machine vision-based workpiece hole quantity detection method, which is used for detecting whether a produced workpiece hole quantity meets a defined requirement or not. Themethod comprises the steps of establishing a hole quantity grayscale image template of a workpiece; collecting a to-be-detected workpiece image, and performing initialization processing on the collected to-be-detected workpiece image; and performing binary image morphology operation on the hole quantity grayscale image template and the processed to-be-detected workpiece image, and according to anoperation result, judging whether the workpiece hole quantity meets the defined requirement or not. Compared with the prior art, the method has the advantages of automatic detection, convenient implementation, high detection efficiency and the like.

The embodiment of the disclosure discloses a method and device for generating audio, equipment and a medium. According to one specific embodiment, the method for generating the audio comprises the following steps: acquiring target voice style information and user voice audio sent by a target user; and on the basis of the target voice style information and the user voice audio, generating a targetvoice audio. The voice style of the target voice audio is a voice style indicated by the target voice style information, the tone of the target voice audio matches the tone of the user voice audio; and the text information corresponding to the target voice audio matches the text information corresponding to the user voice audio. According to the embodiment, on the basis of keeping the tone of theuser voice audio and the corresponding text information, the voice style of the user voice audio can be converted, so that the generation mode of the voice audio is enriched.

Embodiments of the present application disclose a method for generating encoded information and a method for identifying the encoded information. A specific embodiment of the method includes: adding afirst preset number of positioning images at a preset position to a target image containing encoding information to generate a first encoded image; rotating the first encoded image by a second presetnumber of angles about a center region of the first encoded image to obtain second encoded images at a second preset number of different rotation angles; based on the obtained second encoded image, generating a set of encoded images and outputting the set of encoded images. This embodiment enriches the generation mode of the encoded information, so as to help to meet the design needs of differentusers and enrich the presentation of encoded information.

The embodiment of the invention discloses a method for generating an application, terminal equipment and a computer readable medium. The method is applied to a first terminal. A specific embodiment ofthe method comprises the steps of obtaining an application material input by a user through a target host application; obtaining a pre-generated initial application; and editing the initial application according to the application material to generate a target application, the application material being used by the target application in the running process of the target application. The embodiment enriches the generation mode of the application.

The embodiment of the invention relates to an interaction method and device, electronic equipment and a storage medium, and the method comprises the steps: responding to a first game event completed by a user, and determining a first component element of a virtual item which can be obtained by the user; judging whether the user completes a second game event associated with the first game event or not; in response to completion of the second game event by the user, determining a second component element of the virtual item; generating a virtual item based on the first composition element and the second composition element, and displaying the virtual item in a picture of the virtual scene. According to the embodiment of the invention, the generation modes of the virtual props in the game are enriched, and the interactivity and interestingness of the game are improved.

The embodiment of the invention discloses a shooting method and equipment. The shooting method is applied to the terminal equipment. One specific embodiment of the shooting method comprises the following steps: shooting a shot person to obtain a face image of the shot person; obtaining a target image template from a predetermined image template set, the target image template corresponding to an identity recognition result obtained by performing image recognition on the face image; and generating a photographed image or a photographed video of the photographed person based on the target image template. According to the embodiment, different image templates can be obtained on the basis of different face images, so that shot images or shot videos of different shot persons are generated, the probability that the generated shot images or shot videos meet the requirements of the shot persons is increased, and the generation modes of the images or videos are enriched.

The invention discloses an information recommendation method and device, a storage medium and a server, and belongs to the technical field of Internet of Vehicles. The method comprises the following steps: receiving a journey navigation request sent by a vehicle, wherein the journey navigation request comprises a destination; determining at least one key passing point on the driving route from which the vehicle drives to the destination; acquiring environmental data of each key passing point; determining a target insurance type matched with the driving route according to the acquired environment data; and recommending insurance policy data corresponding to the target insurance type to the vehicle. According to the embodiment of the invention, insurance type recommendation is carried out for the user according to the environment data in the to-be-traveled travel; insurance type recommendation is carried out for the user in a personalized mode according to the environment data in the journey in the vehicle-mounted environment, the user can be helped to generate a more adaptive vehicle insurance policy, the driving safety of the user is guaranteed more comprehensively according to theexpectation of the user, the applicability is high, intelligence is achieved, and the generation mode of the insurance policy is enriched.

The invention relates to a detecting method for workpiece shape based on machine vision. The detecting method is used for detecting whether the shape of a produced workpiece has defects or not. The method comprises the following steps: establishing an edge detection image template of the workpiece; acquiring an image of a to-be-detected workpiece and carrying out initialization processing on the image of the to-be-detected workpiece; carrying out binary image morphological operation on the edge detection image template and the processed image of the to-be-detected workpiece; judging whether the shape of the workpiece has the defects or not according to an operation result. Compared with the prior art, the detecting method disclosed by the invention has the advantages of capability of realizing automatic detection, convenience in realization, high detection efficiency and the like.

The embodiment of the invention discloses a method and device for generating an image. A specific embodiment of the method comprises the following steps of: obtaining an initial image input by a user,and the user selects image type information from a predetermined image type information set, wherein the image type information in the image type information set is in one-to-one correspondence withimage generation models in a pre-trained image generation model set, and the image generation models are used for generating target images of image types represented by the corresponding image type information; determining an image generation model corresponding to the acquired image type information from the image generation model set; and inputting the initial image into the determined image generation model to generate a target image. According to the embodiment, image generation modes are enriched, and different types of images can be generated according to different requirements of users.

The invention relates to the field of big data information query, in particular to a user-defined report generation method and device, equipment and a storage medium. The method comprises the steps that basic configuration information of a report to be generated is obtained, report options are created based on the basic configuration information, and the basic configuration information comprises a report name and a data obtaining mode; obtaining query information of a to-be-generated report, and adding the query information to the report options based on the query information; in response to the report name triggered by the user, sending a report query request based on the data acquisition mode and the query condition in the report option; and taking a query result returned by the report query request as the content of the report to be generated so as to generate the report. According to the scheme, the user-defined report is generated according to customer needs, the report is not limited to a specific type any more, report types and generation modes are enriched, and good universality is achieved.

The embodiment of the invention discloses a method and system for generating an application, a server and a computer readable medium. The method is applied to a server. A specific embodiment of the method comprises the steps of sending a pre-generated initial application to a first terminal; obtaining a first type of material input by a first user for the initial application through a target hostapplication running in the first terminal from the first terminal; importing the first type of materials into a page of an initial application to generate an intermediate application; sending the intermediate application to the second terminal; obtaining a second type of material input by a second user for the intermediate application through a target host application running in the second terminal from the second terminal; and importing the second type of materials into a page to be imported into the second type of materials included in the intermediate application to generate a target application. According to the embodiment, the steps of coding, importing the first material, importing the second material and the like in the application generation process are relatively and independentlycarried out, and the application generation modes are enriched.

The embodiment of the application discloses a method and a device used for generating information. A specific embodiment of the method includes: acquiring current input information and a current imagecorresponding to the current input information; determining whether the current image and a pre-acquired target history image indicate the same user; and in response to a case where the current imageand the target history image indicate the same user, generating the replying information of the current input information on the basis of the current input information and history input information,wherein the history input information is pre-acquired information corresponding to the target history image. The embodiment enriches generation manners of the information.

The invention discloses a method and device for generating group announcement information. A specific embodiment of the method comprises the steps of obtaining a chat message corresponding to a target user; determining a target message from the chat messages according to a received selection operation; and generating group announcement information of a target group corresponding to the target user according to the target message. According to the method for generating the group announcement information of the target group corresponding to the target user based on the chat message of the target user, the group announcement information can be flexibly generated, and the generation mode of the group announcement information is enriched.

The invention relates to a method for generating and editing a GOOSE message sequence. The method is characterized by comprising the following steps: (1) generating the message sequence: storing the message sequence by using a composite link list, wherein each frame of message is a node (pFNode) in the composite link list (SList), and editing the message sequence by means of pFNode replication, and insertion and deletion operations of the composite link list, wherein the pFNode comprises: a context (context) of a message structure, a pointer (pnext) appointing to the next pFNode, a time interval (ts) with the next frame of message and a termination condition (condition). By adoption of the method provided by the invention, the problem relay protection personnel cannot conveniently and efficiently edit the message sequence when performing a communication fault tolerance test on a relay protection device is solved.

The embodiment of the invention discloses a method and device for generating a video, electronic equipment and a medium. A specific embodiment of the method comprises the steps that a background image of a to-be-monitored area is acquired, the to-be-monitored area comprises a preset alarm area, and the background image comprises an image without a preset intruding object in the to-be-monitored area; acquiring an image of a preset intruding object; at least two fusion images are generated, and the fusion images are used for presenting different positions of the preset intruder in the to-be-monitored area; and converting the at least two fused images into a video. According to the implementation mode, the video data set is flexibly constructed according to needs, the video generation mode is enriched, and a basis is provided for efficiently constructing the test video in the safety field at low cost.