9results about How to "Improve reading speed" patented technology

This invention relates to the field of image restoration technology, specifically disclosing an image restoration method and device based on discrete wavelet transform. An image restoration network, called Multi-Wavelet Attention Memory Neural Network (MWA-MNN), is designed, consisting of three sub-networks. These sub-networks include Discrete Wavelet Attention Blocks (DWABs) and Coordinate Channel Attention Blocks (CCABs) to extract spatial and channel feature information, effectively separating degraded image layers such as rain streaks from the image to obtain a clean image. The invention leverages the expressive power of wavelet transform for high-frequency information, thereby improving image restoration performance. The sub-networks also include Supervised Selective Kernel Blocks (SSKBs) for filtering effectively extracted features in each sub-network. This invention also proposes a memristor-based cross-array MWA-MNN circuit implementation scheme. Memristors are a novel type of non-volatile memory with faster read speeds and lower power consumption. Experimental results show that this method has good performance and versatility in various image restoration tasks.

This application discloses a method, apparatus, storage medium, and computer device for rendering the interface of a mini-program. The method includes: responding to a rendering request, determining a key of first rendering data in the local storage space of the mini-program, wherein the first rendering data is determined according to the interface content of the mini-program; reading the first rendering data from a target object according to the key of the first rendering data, wherein the target object is configured in the runtime memory of the mini-program, target object data stored in the target object is obtained from the local storage space of the mini-program, and the target object data includes the first rendering data; and rendering the interface according to the first rendering data. This accelerates data reading speed, reduces rendering blocking, improves the smoothness of mini-program operation, effectively improves the display efficiency of the user interface, and ensures greater consistency and accuracy of the read data, avoiding rendering errors caused by untimely updates of local storage.

The application discloses a data desensitization method, device and system, electronic equipment and a storage medium. The method comprises the following steps: in the case that a row event notification information sent by a disk is received, reading corresponding row data from the disk through a reading thread; and according to subscription information, writing the read row data to a data subscriber through a writing thread, so that the data subscriber performs data desensitization processing on the row data. Through the reading and writing mode of the reading thread and the writing thread, the reading and the writing are completed by different threads, the reading and the writing are separated, and the data reading speed is improved.

This invention relates to the field of data processing technology, specifically to a blockchain-based method and apparatus for storing civil engineering test data. The method includes: acquiring civil engineering test data stored on the blockchain and real-time collected civil engineering test data, and preprocessing it to obtain analysis data; determining the importance, access frequency, and final repetition rate of character data based on the analysis data, and obtaining the encoding priority of the character data, adjusting the Huffman coding table; compressing all civil engineering test data according to the adjusted Huffman coding table, and re-storing it in the blockchain; and comprehensively adjusting the Huffman coding table based on the obtained encoding priority of each character data, compressing the data, and re-storing it in the blockchain. This method reduces the amount of data while maintaining access speed, thereby improving the reading speed for high-access data during subsequent data access.

This application relates to a data reading method, apparatus, device, and storage medium. The data reading method includes: after electrically connecting to a storage device, scanning multiple files in the storage device; detecting whether the file type of a scanned file is appearing for the first time; if so, storing the scanned file in a database; otherwise, storing the scanned file in a preset file set in memory; whenever the number of files in the preset file set reaches a preset number, transferring the files in the preset file set to the database and displaying a message indicating that the file list is pending update; after determining that the scanning of all files in the storage device is complete, adding a flag bit to the storage device, wherein the flag bit is used to indicate that the scan has ended. The data reading method provided in this application can improve the problem of slow data reading speed in the prior art.

The application provides a low-cost and high-reliability starting loading redundancy backup method and system for a satellite-borne information system, computer equipment and a storage medium. After the satellite-borne information system is powered on, a three-out-of-two anti-fuse FPGA loads a first-level boot program for each non-FPGA processor from an SPI Flash. When the detection of the first-level boot program loading of a certain non-FPGA processor fails, the first-level boot program is scanned and repaired. After the first-level boot program of the main control CPU is successfully started and loaded, the operating system image file of the main control CPU is read from a solid state disk array in a three-out-of-two manner. If the reading fails, the operating system image file is repaired. After the reading is successful, the operating system image file is loaded to the main control CPU to complete the starting of the main control CPU. After the operating system of the main control CPU is started, the main control CPU management software is run to create starting loading threads for the remaining non-FPGA processors and FPGA processors, respectively, and to completely start and load each processor.

The embodiment of the application provides a data backup method and a computing device, the method comprises the following steps: in response to a backup request for target data, acquiring bitmap information of the target data; according to the bitmap information, pre-reading at least part of data blocks in the target data from a storage medium to a cache; in response to a read request for a target data block in the target data, querying whether the target data block exists in the cache; if the target data block exists in the cache, reading the target data block from the cache. The data backup method and the computing device provided by the embodiment of the application can convert a data read request into a memory access. Since the storage medium has a high access delay, compared with a mode of reading a data block from the storage medium, the mode of reading a data block from the cache improves the data read speed, thereby improving the data backup efficiency.

The invention belongs to the field of high-performance storage systems and computer system structures, and particularly discloses an LSM tree reading optimization design method based on a partition ordered hash table, which is characterized in that a hash table is maintained in a memory, the table is used for caching a key read from a disk recently, and a file number and a file size corresponding to the key are recorded; when a user initiates a query request, if a target key is not hit in both the memtable and the immutable memtable, a traditional LSM tree mechanism can directly enter a disk for layer-by-layer query, but in the method, the table can be queried in a memory preferentially. According to the method, the key and the file number and the file size of the corresponding stable are maintained in real time by utilizing the hash table of the partition; whenever a new stable is generated, the system checks whether a key in the new stable is overlapped with an existing key in the hash table, and if the key in the new stable is overlapped with the existing key in the hash table, the file number and the file size of the corresponding key in the hash table are updated. In this way, the target stable can be directly positioned during query, layer-by-layer traversal is not needed, and therefore the reading speed is effectively increased.

The present application relates to a kind of equipment time series data storage model construction method and system suitable for coal mine multi-disaster fusion analysis, belong to coal mine safety disaster analysis technical field, to solve the problem of data effectiveness, history deficiency in prior art.The method first abstracts equipment time series data into point type data mode and line type data mode, and constructs the data storage conceptual model including monitoring time series data model, alarm time series data model, personnel positioning monitoring data model, monitoring time series slice data model and alarm time series slice data model.Database system using NoSQL technology, design the database table into conceptual model, and create the corresponding wide table structure, take column-based storage mode, set the compression strategy of column, realize long-term online storage and efficient reading of data.The equipment time series data storage model constructed by the present application can effectively reduce the complexity of data access, reduce data load and operation amount.