71results about How to "Troubleshoot technical issues with poor performance" patented technology

The invention discloses a data synchronization method and device. The method comprises the steps of reading a catalogue index file in a first data source, thereby obtaining data catalogues of n applications contained in the catalogue index file; reading a data index file Ei contained in the data catalogue Di; resolving an update configuration file of the application Ai, thereby obtaining catalogues of target data and a type attribute of the target data in the application Ai, wherein the target data is the data on which data synchronization needs to be carried out; and converting a data structure of the data in the first data source according to the change information in the data index file Ei and the type attribute of the target data, wherein a second data source obtains the data after the data structure is converted in the first data source, thereby finishing data synchronization. According to the method and the device, the technical problem that in the prior art, the synchronization performance of the data between different databases is relatively low due to the fact that a user cannot screen the data needing to be synchronized according to the demand of the user is solved.

An embodiment of the invention discloses a noise suppression method, device and system, and solves the technical problem of poor performance of an ANC (automatic noise canceller) method due to large difference between sound field environments of a reference microphone and a main microphone at present. The noise suppression method includes the steps: S1, receiving internal noise acquired by a reference voice acquisition mechanism and voice signals containing external noise acquired by a main voice acquisition mechanism by a noise suppression device when the voice signals are inputted; S2, extracting internal signal characteristics corresponding to the internal noise; S3, acquiring external approximate characteristics corresponding to the external noise according to the internal signal characteristics and a preset mapping formula; S4, converting the external approximate characteristics into corresponding noise signal estimation values by Fourier inverse transform; S5, processing the voice signals containing internal noise estimated and acquired by noise signals in a preset noise cancellation mode to obtain denoised voice signals after noise suppression.

The invention provides a method and a device for detecting a moving object on the ground by using airborne passive radar, and electronic equipment. The method comprises the following steps of: pre-processing airborne passive radar signals to obtain a distance compression gate; acquiring training samples according to the distance gate; establishing a joint sparse matrix recovery model based on clutter sparsity according to the training samples; calculating the joint sparse matrix recovery model by using a preset algorithm to obtain a clutter space-time spectrum; and eliminating the clutter component and the noise component in the distance gate to obtain target components by using the clutter space-time spectrum. Input parameters of the preset algorithm comprise a training sample matrix, a time domain sparse dictionary, a spatial sparse dictionary, and a preset maximum number of iterations. According to the method and the device for detecting the moving object on the ground by using theairborne passive radar, and the electronic equipment, the technical problem of poor performance of the existing airborne passive radar clutter suppression method is solved. The method can effectivelysuppress the strong clutter and the noise to obtain the target components.

The invention discloses a micro-arc oxidation method for surfaces of copper, copper alloy, zinc and zinc alloy. The method includes the steps that aluminum films are manufactured on the surfaces of the copper or the copper alloy or the zinc or the zinc alloy through a magnetron sputtering method, the aluminum films are then treated through a micro-arc oxidation method, and therefore aluminum oxide ceramic films are obtained. Due to the fact that non-valve metal subjected to magnetron sputtering treatment is then subjected to micro-arc oxidation, the aluminum oxide ceramic films are obtained, abrasion resistance is improved by more than 10 times, and corrosion resistance is enhanced greatly; and the technical problem that the ceramic films cannot be manufactured for the non-valve metal through the micro-arc oxidation method or the performance of the manufactured ceramic films is quite poor is solved.

The invention discloses a method and a device for determining a language model. The method comprises the following steps: acquiring a first corpus, wherein the first corpus is a language text selectedin a preset context; Obtaining a first language model by training the first language data; adopting A first language model to filter the target text to obtain a second corpus, and acquiring a secondlanguage model by training the second corpus, wherein the target text is retrieved using a keyword set extracted from the first corpus as an index. The invention solves the technical problem that theperformance of the corpus is low because the language model in the prior art is only acquired through the strategy of piling up the corpus.

The invention discloses a distributed file system-based data reading method and device and a distributed file system-based data writing method and device. The data reading method comprises the following steps of: receiving a reading request sent by client equipment by a first storage node, belonging to a first area, in a distributed file system, wherein the reading request is used for requesting to-be-read data; judging whether a first storage position corresponding to the to-be-read data has a preset sign or not; and when the judging result is positive, reading the to-be-read data from a second storage node, wherein the second storage node is located in the second area and the second area is different from the first area. According to the data reading method and device and the data writing method and device, the technical problem that distributed file systems are bad in performance when carrying out multi-regional cluster storage in correlation techniques is solved.

The invention discloses a task storage method and apparatus. The method comprises the steps of obtaining a to-be-stored task set; according to data quantity information corresponding to task data, determining group information corresponding to the task data from preset metadata; according to the group information and partition information corresponding to the task data, reading corresponding storage positions of the task data from a pre-configured external data file; and storing the task data in the task set to the corresponding storage positions of the task data respectively. According to the method and the apparatus, the technical problem of low system performance caused by generation of fragmentary files due to storage of the task data in a distributed database according to type information in the prior art is solved.

The embodiment of the invention provides a data storage method and device and electronic equipment, and the method comprises the steps: determining an estimated heat state of sample data according tothe historical heat state of the sample data in a historical time period; and storing the sample data in a storage block corresponding to the estimated heat state. Predicting in advance to obtain an estimated heat state of the sample data in a future time period; storing the sample data in a storage block corresponding to the estimated hot degree state in advance, wherein the larger the estimatedheat state is, the higher the sample data is; the data is pre-stored in a storage block with higher read-write speed. Therefore, the feedback speed of the storage system to the access sample data canbe improved, so that the performance of the storage system is improved, the technical problem that the performance of the storage system is poor due to the fact that a certain hysteresis exists in anexisting storage strategy based on hybrid storage is solved, and the technical effect of improving the performance of the storage system is achieved.

The invention discloses an antenna assembly which comprises an internal antenna, a RF (radio-frequency) circuit, a drawing type antenna, a switch assembly, a detection switch and a controller, wherein a first end of the switch assembly is connected with the RF circuit, a second end is connected with the internal antenna, and a third end is connected with the drawing type antenna through the detection switch; the controller is connected with the switch assembly and the detection switch respectively, controls the first end and the second end of the switch assembly to be communicated, and controls the first end and the third end of the switch assembly to be communicated when detecting a signal drawn by the drawing type antenna. The invention further discloses a control method of the antenna assembly. The antenna mode of the antenna assembly is switched between the internal antenna mode and the drawing type antenna mode according to requirements of actual scenes, and the performance of the antenna assembly is improved.

The invention provides a preventive maintenance method for fatigue cracks of steel box girders and a steel box girder, which relates to the technical field of maintenance of steel box girders and is designed to solve the problem of poor effect of existing repair methods for cracks of steel box girders. The preventive maintenance method for fatigue cracks of the steel box girder is used for curingthe welding seam of the steel box girder, comprising the following steps of: polishing the surface of the welding seam to be flush; Grinding is carried out in the direction of coating cracking on theweld to eliminate microcracks on the weld. The steel box girder is cured by the precautionary maintenance method of the fatigue crack of the steel box girder. The invention provides a preventive maintenance method for fatigue cracks of a steel box girder and the purpose that the steel box girder is used for realizing accumulative damage to a metal surface and elimination of small cracks before thecrack extends over a large area, thereby greatly improving the fatigue resistance at the welding seam of the existing steel box girder.

The invention discloses a method and device for optimizing the performance of an image processing server end. The method comprises receiving an image processing request; distributing the load balanceof a plurality of graphics processors in the depth learning inference framework through a preset application programming interface. The method solves the technical problem of poor performance of the server end. The method can improve the performance of the server end.

The invention discloses a task model acquisition method and device, a storage medium and an electronic device. The method comprises the following steps: obtaining target model parameters obtained by carrying out model training on the first training model, wherein the first training model comprises: a first encoder and a first decoder, the target model parameters comprisea first model parameter corresponding to the first encoder, the input of the first training model is a first image of a first channel of the target optical image, the output of the trained first training model is a prediction image corresponding to a second image of a second channel of the target optical image, and the graph similarity between the prediction image and the second image is greater than or equal to a target similarity threshold; initializing model parameters of a second training model by using the target model parameters to obtain an initial task model, the second training model comprising a second encoderhaving the same structure as the first encoder; and training the initial task model by using the target optical image to obtain a target task model.

The invention discloses a language model obtaining method and device. The method comprises: acquiring a first corpus and a second corpus, wherein the first corpus is a language text collected randomly, and the second corpus is a language text selected under a preset context; carrying out confusion degree calculation on the first corpus by adopting a first language model obtained by training the second corpus so as to screen out a third corpus, the confusion degree between the third corpus and the second corpus being smaller than a preset threshold value; and performing interpolation processingon a second language model obtained by training the first corpus and a third language model obtained by training the third corpus to obtain a language model to be used. The technical problem that a language model is low in performance is solved.

The invention discloses a distributed data acquisition method and device. The method comprises the following steps of acquiring generated message data, and sending the acquired message data to a preset front message queue; utilizing a preset load balancer to distribute the message data in the front message queue to each preset message middleware according to a preset data distribution mode; and analyzing sub-message data in each message middleware, and storing the analyzed sub-message data into a preset database which is subjected to read-write separation processing. Compared with the prior art, the transmission pressure of the message data can be decomposed, the database is subjected to read-write separation processing in advance, and the data processing performance is high, so that a problem of poor performance of the whole system due to the fact that an automatic monitoring system of rail transit signal equipment has middleware blockage and delayed data processing when the acquireddata is larger in the prior art is solved.

The invention discloses a product object acquisition method and device. The method comprises the steps that product request information submitted by a current account is received, and at least one piece of product constraint information saved in advance is called according to the product request information, wherein a constraint model of a tree structure is adopted to save the product constraint information, and each piece of the product constraint information comprises a constraint name, at least one constraint condition and a constraint result; at least one piece of product constraint information is traversed, and if a constraint condition matched with product parameters contained in the request information and a constraint result matched with parameter data are obtained through query in any one piece of product constraint information, it is determined that a requested product object is verified successfully; and under the condition that the product object is verified successfully, the product object is submitted to a processing system. Through the product object acquisition method and device, the technical problem that in the relevant art, system performance is low because a user constantly calls a system background to verify constraint conditions when performing product configuration selection is solved.

The invention discloses a power distribution network adaptation evaluation processing method and device. The method comprises the following steps: acquiring data transmission information of a backbonelayer in a communication system of a power distribution network; obtaining data transmission information of an access layer of the communication system, wherein the data transmission information of the backbone layer and the data transmission information of the access layer both comprise data transmission flow and data transmission delay requirements; and matching in a network model according tothe data transmission information of the backbone layer and the access layer, and selecting a network type matched with the data transmission information. According to the application, the technical problem that the communication networking performance is poor due to the fact that various distribution network communication networking rules exist in each power supply area and the phenomenon that the network transmission information is not matched with the communication networking type exists is solved.

The invention discloses a device and method for correcting a multi-cavity special-shaped double-layer thin-wall section pipe through rubber and plastic soft dies. The correcting device comprises an upper bottom plate, a lower bottom plate, a circumferential bulging die, a fixing assembly, a pressure suction assembly and a rubber sealing plate. The circumferential bulging die is used for containingthe to-be-machined multi-cavity special-shaped double-layer thin-wall section pipe and can be attached to the formed multi-cavity special-shaped double-layer thin-wall section pipe. The pressure suction assembly comprises hydraulic cylinders and rubber soft dies. The multiple hydraulic cylinders are arranged on one side of the circumferential bulging die in a surrounding mode and provide deformation restoring force for the circumferential bulging die. According to the shape of the to-be-machined multi-cavity special-shaped double-layer thin-wall section pipe, the multiple rubber soft dies aresequentially placed in the circumferential bulging die, and the to-be-machined multi-cavity special-shaped double-layer thin-wall section pipe is extruded after pressing and expansion. The multi-cavity special-shaped double-layer thin-wall section pipe is corrected through the device and method, the defects such as recesses and protrusions on the surface of the thin-wall section pipe can be overcome, manual correction is avoided, the correction time is shortened, and the correction efficiency, quality and precision are improved.

The invention discloses a binocular camera distortion correction system and method based on an FPGA, and an intelligent terminal. The system comprises an image caching module which is used for receiving a left image and a right image collected by a binocular camera, and caching the data of the left image and the data of the right image; the distortion correction module is used for reading the correction table data and decomposing the correction table data into storage addresses; and reading and caching the data of the left image and the data of the right image, reading an image data matrix according to a storage address of table data decomposition, and substituting the image data matrix into a bilinear interpolation module to perform bilinear interpolation calculation, thereby outputting aleft image correction image and a right image correction image. The technical problem that in the prior art, a camera distortion system occupies a large amount of memory and system resources, and consequently the overall performance of the system is poor is effectively solved.

The invention discloses an image processing method and electronic equipment. The method comprises a step of collecting and obtaining a first image comprising a first facial object through an image collection unit in the electronic equipment, a step of obtaining the first pixel ratio of the first facial object in the first image, a step of determining a first filtering parameter according to the first pixel ratio, a step of carrying out one-dimensional filtering on the first image in the first direction of an image coordinate system, and obtaining a first filtering image, and a step of carrying out one-dimensional filtering on the first filtering image in a second reaction which is orthogonal to the first direction in the image coordinate system, and obtaining a second image.

The application discloses an in-vehicle voice processing method and device. According to the in-vehicle voice processing method, in-vehicle voice processing is executed through a first terminal and asecond terminal, the first terminal is used for controlling the second terminal to execute voice processing through in-vehicle voice application mounted in advance, and the second terminal is used forreceiving first voice content, sent by the in-vehicle voice application, in a vehicle. The in-vehicle voice processing method comprises the following steps: judging whether the first terminal is connected with the second terminal through a wireless communication protocol or not; if yes, playing the first voice content on the second terminal and monitoring second voice content on the first terminal; and continuing to pick up the monitored second voice content on the first terminal until the in-vehicle voice application operates on the background of the first terminal. Through the adoption of the in-vehicle voice processing method and device, the technical problem that the in-vehicle voice processing method is relatively poor in effect is solved, and the in-vehicle voice processing method and device can be used for aided navigation in the vehicle driving process.

The invention provides a video redirecting method and a video redirecting device, and relates to the field of information technologies. The method comprises the steps of intercepting an audio-video to be played before a video player in a server decodes and plays the audio-video to be played, wherein the audio-video to be played is the video to be played under the request from the client to the server via a remote desktop; processing the video player, thus allowing the processed video player to not generate alarm information after the audio-video to be played is intercepted; and after the video player is processed, sending the audio-video to be played to the client, wherein after the client receives the audio-video to be played, the audio-video to be played is combined with an image of the remote desktop, and a combined result is displayed on a display interface of the client. According to the method and the device, the technical problem of poor performance when the server transmits audio-video data to the client via the remote desktop in the prior art is alleviated.

The invention provides an antenna device and electronic equipment. The antenna device comprises an antenna module and an antenna cover; the antenna module is used for transmitting and receiving a first radio frequency signal of a first preset frequency band towards a first preset direction range and is also used for transmitting and receiving a second radio frequency signal of a second preset frequency band towards a second preset direction range, and the first preset frequency band is smaller than the second preset frequency band; an overlapping area exists between the first preset direction range and the second preset direction range; a radome and the antenna module are arranged at intervals; the radome comprises a substrate and a resonance structure borne on the substrate; at least part of the resonance structure is located in the overlapping area; the resonance structure has an in-phase reflection characteristic at least for the first radio frequency signal and an in-phase reflection characteristic for the second radio frequency signal. The antenna device provided by the invention has relatively high gain; when the antenna device is applied to the electronic equipment, the electronic equipment has relatively good communication performance.

The invention discloses a face aging processing method based on multiple emotions. The method comprises the following steps: acquiring an original picture and a target picture corresponding to the original picture; respectively inputting the obtained original picture and the target picture into encoders Ee and EI of the trained first network model so as to respectively obtain key features of the original picture and key features of the target picture; inputting the obtained key features of the original picture and the key features of the target picture into a generator GE of a trained first generative network to obtain a color picture; and inputting the color picture into a generator GA of a trained second network model to obtain an aged color picture. According to the method and system, multiple losses are utilized to ensure that the CD-Net learns a real image under the condition that other information is not lost, vivid facial aging images under different emotions can be generated, and the emotional characteristics are specific and not random.

The invention relates to the technical field of wool pre-carding, and discloses a wool pre-carding device for a wool carding machine. The device comprises a pre-carding base, wherein a storage grooveis formed in the pre-carding base, a pre-carding cylinder is arranged right above the storage groove, the left end face of the pre-carding cylinder is connected with an output shaft of a first pre-carding motor, first pre-carding vertebral teeth are evenly distributed on the side wall in a pre-carding cylinder cavity, first pre-carding needles and second pre-carding needles are evenly distributedon the left end face and right end face in the pre-carding cylinder cavity, gaps are formed among the second pre-carding needles, the first pre-carding needles and second pre-carding vertebral teeth evenly distributed on the outer side wall of a pre-carding shaft, the right side face of the pre-carding shaft penetrates through the right end face of the pre-carding cylinder, extends to the outer portion of the pre-carding cylinder and is connected with an output shaft of a second pre-carding motor, and a water jet of which a water outlet end is provided with blowholes is arranged between the pre-carding cylinder and first blower fan blades. The wool pre-carding device for the wool carding machine solves the problem of low efficiency and poor effect of opening and untying tangled wool of anexisting wool pre-carding device.

The invention relates to the technical field of coatings, in particular to a preparation method and a preparation system of water-based acrylic finishing paint. Comprising a reaction kettle, a stirring device and a feeding device, the stirring device comprises a stirring rod, an adjusting supporting rod, a sliding supporting rod, a stirring vertical rod, an adjusting assembly and a driving assembly, the adjusting supporting rod is rotationally connected with the stirring rod through the adjusting assembly, the sliding supporting rod is arranged on the adjusting supporting rod and slides with the adjusting supporting rod, and the stirring vertical rod is arranged on the sliding supporting rod; by pulling the sliding support rod, the position of the stirring vertical rod is moved left and right, the angle of the adjusting support rod is further adjusted, so that the height of the stirring vertical rod is adjusted, the stirring rod is further rotated, and raw materials in the reaction kettle can be stirred at different positions and depths by the stirring vertical rod; furthermore, the raw materials are fully stirred, so that the reaction among the raw materials is thorough, and the performance of the finishing paint is good.

The embodiment of the invention discloses an audio zooming method and device, folding screen equipment and a storage medium. The method comprises the following steps: loading a preset audio zoom parameter in the folding screen equipment; receiving an audio zooming request sent by a user; collecting a folding included angle of the folding screen equipment; and calling an audio parameter matched with the folding included angle in the audio zooming parameters, and carrying out audio zooming in the folding screen equipment according to the audio parameter. According to the method and the device, the parameters for audio zooming are loaded in the folding screen equipment in advance, when the user requests for audio zooming, the folding included angle of the folding screen equipment is directly collected, and then the parameters matched with the folding included angle are collected from the loaded parameters through the folding included angle, so that the automatic audio zooming function of the folding screen is realized, and the user experience is improved. The technical problem that in the prior art, when a folding screen mobile phone is in a multi-angle overturning state, the audio zooming performance is poor is solved.

The invention discloses a statement identification method and device, a storage medium and a processor. The method comprises the steps of obtaining a to-be-processed statement; inputting the to-be-processed statement into a target encoder to acquire semantic representation of the to-be-processed statement through processing,wherein the target encoder is an encoder shared by a grammar error detection task and a grammar error correction task; and inputting the semantic representation of the to-be-processed statement into the processing model to obtain a processing result, with the processing result comprising one of a grammar error detection result and a grammar error correction result. The technical problem of poor task processing performance of the system due to the fact that different annotation data of different tasks are difficult to directly mix and use is solved.

The invention relates to a preparation method and a use method of a Li4Mn5O12 nanosheet material, which are used for solving the technical problem of poor performance of an existing manganese oxide-based material serving as a supercapacitor electrode material. According to the technical scheme, the method comprises the steps of arranging a carbon cloth on which Mn3O4 nanosheets grow in 2M Li2SO4 electrolyte, carrying out an electrochemical oxidation reaction under a three-electrode system, and growing the Li4Mn5O12 nanosheets on the carbon cloth. According to the present invention, an asymmetric pseudocapacitor is prepared by taking a Li4Mn5O12 nanosheet electrode as a positive electrode and the commercial activated carbon as a negative electrode. An electrochemical performance test is carried out in the 2M Li2SO4 electrolyte, the voltage window of the asymmetric pseudocapacitor is as high as 2.2 V, the energy density is as high as 97.1 Wh kg <-1> when the power density is 366.7 W kg <-1>, and the specific capacitance retention rate of the asymmetric pseudocapacitor at 12000 cycles and the scanning rate of 200 mV s <-1> is 97.5%.

The invention discloses a distributed data backup method, device and system. The method comprises the steps in which, a main server receives an operation request to call a client and generates operation data based on the operation request; the main server saves the operation data to a local main storage and generates a synchronization instruction; and the main server synchronizes the operation data into a sub storage of a sub server based on the synchronization instruction, after the operation data is synchronized to the sub server, and the main server gives tacit consent to that the synchronization operation is successful and thus does not need to wait for the return of feedback information on the successful synchronization from the sub server. The invention solves the technical problem of poor system performance as a disaster recovery scheme provided by a distributed system in the prior art requires the system to maintain consistency.