140 results about "Information recovery" patented technology

The embodiment of the invention provides a low-illumination image enhancement method and system based on frequency decomposition and multi-scale fusion, and the method comprises the steps: obtaining a to-be-processed image, and decomposing the to-be-processed image into a low-frequency component feature map and a high-frequency component feature map through employing an iterative mean filter; a low-frequency tensor is generated based on the low-frequency component feature map, and a high-frequency tensor is generated based on the high-frequency component feature map. And fusing the high-frequency tensor and the low-frequency tensor to generate an enhanced feature map. According to the method, iterative mean filtering can be applied to separate frequency bands, and an image to be processed is decomposed into a high-frequency component and a low-frequency component. More accurate image enhancement is realized through different sub-networks, that is, a high-frequency enhancement network focuses on detail reconstruction and noise suppression, and a low-frequency enhancement network is responsible for recovery and transformation of illumination and color information, so that the image enhancement quality can be improved, and the problem of poor image enhancement effect is solved.

The invention discloses a covert communication system based on a generative diffusion model, and belongs to the technical field of covert communication. In order to solve the problems of insufficient concealment and poor anti-noise performance in the prior art, hidden transmission and reliable recovery of information are realized by constructing a diffusion generation and classification discrimination module. The sending end carries out noise addition and shaping on the original bit waveform by using a forward diffusion process to form a noise-like hidden signal; and a receiving end recovers a signal structure through a reverse denoising process, and accurately judges bit information through a classifier. According to the method, a Gaussian approximation hidden waveform library is constructed by adopting a signal pair training model which is consistent in statistics but distinguishable in structure, and random sampling mapping and boundary smooth splicing are combined, so that transmission signals are difficult to distinguish from noise in a frequency domain and a feature space; therefore, the avoidance capability, the anti-interference robustness and the information recovery accuracy under the low signal-to-noise ratio of the detection means are remarkably improved.

The invention relates to a recoverable visual encryption method for a human body privacy area in an image collected by an unmanned aerial vehicle. The method comprises the following steps: constructing a joint deep learning model integrating a pre-training target detection module, a self-adaptive visual encryption module and an enhanced visual decryption module; in the training stage, target detection module parameters are frozen, and an encryption and decryption module is optimized end to end through combined loss formed by a composite decryption loss function and an encryption strength loss function; during application, the target detection module accurately locates a human body sensitive area, the encryption module applies size-adaptive and strength-learnable pixelated transformation to generate a natural encryption image, and the decryption module realizes high-fidelity original information recovery through an enhanced U-Net network under an authorization condition. Compared with an existing traditional image encryption or irreversible desensitization technology, the method has the advantages that seamless integration with a mainstream target detection process is achieved, the integrity and restorability of information are guaranteed while high visual concealment is achieved, and the method has the advantages of being intelligent, high in efficiency and high in practicability.

The invention provides a database log analysis and recovery method and device, equipment, a medium and a program product, and relates to the field of data processing. Periodically generating a snapshot file of cache record data involved in a database log analysis process, and recording a snapshot identifier corresponding to the snapshot file; updating version information of a data dictionary, and binding the snapshot identifier with the version information of the data dictionary; when database log analysis is interrupted, matching corresponding target version information from the data dictionary according to a target snapshot identifier; recovering the state of the cache record data based on the target version information, and determining the initial position of database log analysis; according to the method, the snapshot file of the cache record data is periodically generated, and the snapshot identifier and the data dictionary version information are bound, so that the cache data state and the analysis starting position are quickly recovered based on the latest snapshot after the database log analysis is interrupted, and the fault tolerance and the recovery efficiency of log analysis are improved.

The invention belongs to the technical field of wireless communication, and particularly relates to a communication sensing integrated system based on non-orthogonal AFDM (Alternating Frequency Division Multiplexing). According to the system, a bandwidth compression factor is introduced into an AFDM framework to construct a non-orthogonal waveform, and controllable overlapping between chirp subcarriers is realized, so that the spectrum utilization rate is improved, and the robustness of the system under a high-speed dynamic channel is enhanced. And a receiving end realizes reliable information recovery through maximum likelihood detection, and processes echo signals in combination with a two-dimensional parameter estimation algorithm so as to realize high-precision sensing of the target distance and speed. The system has good anti-Doppler capability and perception precision performance, and is suitable for communication perception fusion application in a high-speed mobile scene.

The embodiment of the invention provides a software content layout data processing method and device, and the method comprises the steps: receiving template data, annotation data and configuration data of a preset software template, carrying out the structural analysis of the template data according to a template analysis engine, determining a structural analysis result, initializing the configuration data according to a rule engine, and carrying out the structural analysis of the configuration data. And determining a rule engine instance, based on the structure analysis result and the rule engine instance, performing standardized preprocessing on the template data through a template preprocessing engine, determining a standardized template structure, and replacing the template content of the standardized template structure according to the annotation data. According to the method and the device for replacing the small red book template, the template is replaced, the labeling information recovery operation, the spacing deviation recovery operation, the exceeding detection operation and the layout adjustment operation are sequentially executed on the basis of the replaced template, and a final template structure is obtained for a user to use, so that the small red book template typesetting efficiency and quality can be improved.

The invention discloses a depth guidance image defogging method and system for deep ground medical rescue, and aims to improve the visibility of medical rescue monitoring images and the accuracy and real-time performance of structural information recovery in deep ground environments such as mines. By introducing a scale collaborative attention module, collaborative learning of multi-scale features is realized, scale confusion is relieved, and the cross-scale feature collaboration capability is enhanced; and meanwhile, an error perception guide module is adopted to fuse local feature difference and global dependence, and mutual promotion optimization is performed on defogging and depth estimation tasks, so that more accurate image restoration is realized in a region with a complicated structure or non-uniform fog distribution. According to the method, close collaborative optimization and performance improvement between image defogging and depth estimation tasks can be realized, and the problems of scale confusion and limited scale collaborative capability in multi-scale feature learning can be solved to a certain extent.

The invention provides a deep and large fracture depicting method based on low-frequency electromagnetic information recovery enhancement, and the method comprises the steps: S1, collecting electromagnetic data, and carrying out the preprocessing of the electromagnetic data, and obtaining the preprocessed electromagnetic data; s2, obtaining frequency domain apparent resistivity and phase according to the preprocessed electromagnetic data; s3, establishing a shallow stratum electrical model; s4, acquiring a depth domain apparent resistivity profile according to the shallow stratum electrical model; and S5, depicting deep and large fracture distribution based on the depth domain apparent resistivity profile. According to the invention, enhancement of low-frequency electromagnetic signals is realized, the inversion precision of electromagnetic data is improved, low-frequency geological information is effectively recovered, and deep and large fractures can be effectively described.

The present disclosure provides an encryption and decryption system and method. The encryption and decryption system includes a transmitting device and a receiving device. The transmitting device is configured to store a plurality of original images. The receiving device is connected to the transmitting device and configured to store a corresponding table. The transmitting device generates an encrypted string, selects a representative transformed image from the plurality of transformed images according to the encrypted string, transmits the representative transformed image to the receiving device, and does not transmit the encrypted string to the receiving device. The receiving device identifies a first original image number and a second original image number from the representative transformed image, and looks up the corresponding table according to the first original image number and the second original image number to generate the encrypted string. Thus, various network attacks can be prevented, resource and time consumption can be greatly reduced, the problem of excessive distortion or low capacity of embedded data can be solved, and the information recovery capability can be greatly improved.

The invention discloses an image processing method, an image processing device and image processing equipment based on palette prediction, and aims to solve the problem of inconsistent coding and decoding caused by lack of prediction information construction in a decoding process of a palette prediction mode if the palette prediction mode is adopted during loop filtering coding based on a low-complexity neural network. Meanwhile, compression performance can be improved to a certain extent, the embodiment of the invention provides a solution, existing coding information or information which can be obtained by a coding end and a decoding end is used as prediction information of loop filtering of the low-complexity neural network, the prediction information is input into a loop filtering model of the low-complexity neural network, and the compression performance of the low-complexity neural network is improved. According to the method, for a block with the optimal prediction mode being a palette prediction mode, reconstruction information output after the loop filtering model of the low-complexity neural network is processed is recovered into reconstruction information stored before the loop filtering model of the low-complexity neural network is processed, so that the coding and decoding consistency can also be kept, or for a block with the optimal prediction mode being a palette prediction mode, reconstruction information output after the loop filtering model of the low-complexity neural network is processed is recovered into reconstruction information stored before the loop filtering model of the low-complexity neural network is processed.

The present application relates to a kind of information recovery methods based on scattering system, belong to spectral data processing technical field.The method is: the scattering characteristic of scattering system is calculated;Using the scattering characteristic calculated, establish scattering signal self-excitation amplification mechanism;On the scattering signal self-excitation amplification mechanism, incident light field time domain, space domain, frequency domain coding and trend decoding, wave analysis, to nearly completely recover scattering information;Using negative correlation statistical operation obtains the accurate scattering information of high noise scattering system.The present application is amplified by the scattering signal of special self-excitation method, based on light field composite coding and scattering information trend analysis, accurately judges input field information;By introducing negative correlation operation, can nearly completely recover incident field information, realize the nearly complete evaluation of the scattering characteristic of extremely high noise system.The present application promotes the progress of imaging technology not limited by the thickness of scattering medium, further promotes the development of realizing human clinical high-resolution optical reflection, transmission imaging.

The invention relates to the technical field of artificial intelligence and big data, and discloses a vector compression reconstruction method for long text semantic retrieval, which comprises the following steps: firstly, segmenting a long text T into n text segments S1, S2,..., Sn according to semantic similarity or a fixed length; secondly, for each text segment Si, generating a local semantic vector li by using a BERT pre-trained language model, and converting the input text segment Si into a high-dimensional vector by the BERT model; then global vector generation is carried out, specifically, weighted average is carried out on the local vectors li, l2,..., ln of all the segments through a weighted average method, and a global semantic vector g is generated; carrying out explicit compression and key information storage; performing staged hybrid semantic retrieval; cross-segment semantic reconstruction and reasoning are carried out, and context information recovery is carried out; performing dynamic semantic reconstruction; and generating answers based on the reconstructed context. According to the method, deeper reasoning can be carried out, and more accurate and more comprehensive answers are generated.

The invention discloses a target extraction method and device for linear cultural heritage ruins and relics. The method comprises the following steps: acquiring a remote sensing image of a linear cultural heritage; inputting the remote sensing image into a network feature extractor for multi-level feature extraction to obtain shallow layer features, middle layer features and high layer features; performing target semantic enhancement on the deep features to obtain enhanced deep features; carrying out attention and fusion operation on the shallow-layer features and the middle-layer features to obtain shallow-layer fusion features and middle-layer fusion features; performing gating attention fusion and information recovery operation on the enhanced deep-layer features, the shallow-layer fusion features and the middle-layer fusion features to obtain features after information recovery; performing linear feature reconstruction on the features after information recovery to obtain reconstructed features; and inputting the reconstruction features into a segmentation head and an auxiliary divider to obtain a prediction result. The method can achieve the automatic extraction and precise recognition of the large-scale linear cultural heritage.

The invention provides an OCTA retinal vessel segmentation method for improving U-Net based on adjacent feature extraction, and belongs to the field of medical image processing. According to the method, on the basis of a U-Net network, feature maps of adjacent layers are combined through a Proximal Feature Merge (PFM) module, and the capability of segmenting small blood vessels is remarkably improved; meanwhile, a shallow feature complementation module (SFC) is introduced, feature maps of different levels of an encoder are combined at the deepest level, and connectivity of segmented blood vessels is effectively guaranteed through a series of information recovery; in order to adapt to small blood vessels and refined edge information, a BottleUnext module is further added in the method, the resolution of the feature map is gradually amplified, then complex features are extracted, and finally the resolution of the feature map is recovered. Experiments show that compared with an existing method, the method shows more advanced performance in an OCTA retinal blood vessel task, and accurate segmentation of the retinal blood vessel is achieved.

The present application provides a power control device that can suppress the power consumption reduction effect obtained by leveling execution from being weakened by suppressing the power consumption amount after leveling execution. The power control device includes a communication unit and a control unit. The control unit stores information of a leveling instruction and information of a recovery mitigation request. The information of the leveling instruction includes information of an execution time and information of a first allowable average power in the execution time. The information of the recovery mitigation request includes information of a recovery period and information of a second allowable average power in the recovery period. The control unit controls the driving of a power load device in such a manner that the average consumption power of a facility in the execution time becomes lower than the first allowable average power, and controls the driving of the power load device in such a manner that the average consumption power of the facility in the recovery period becomes lower than the second allowable average power. The second allowable average power is greater than the first allowable average power.

This invention provides a gaming machine that partially unifies the backup and recovery process, which restores predetermined information to its pre-game state, with the backup process, which backs up the predetermined information, thereby reducing the pressure on the program area capacity of the main ROM. [Solution] In the game start state control process of the main process, a backup of various information is acquired, and further, in the game end state control process of the main process, a process is provided to restore the backed-up various information according to predetermined conditions. In this process, when backing up and when restoring the various information, the values ​​of the registers that store the addresses of the respective data storage areas are swapped, thereby making some of the programs common.

This invention provides a method for constructing a power distribution system fault recovery agent based on large-scale and small-scale model collaboration. The method includes: retrieving large-scale model prompts and context from a pre-built knowledge base based on received fault information recovery statements; decomposing the fault information recovery statements into several sub-tasks according to a task decomposition prompt template, and calling the corresponding small model to execute each sub-task; generating a first recovery strategy based on the large-scale model prompts, the large-scale model context, and the fault information recovery statements; generating a second recovery strategy based on the output results of all small models; and using the first and second recovery strategies as the output fault recovery strategies to perform fault recovery on the power distribution system. This invention, by employing a large-scale and small-scale model collaboration method, effectively leverages the excellent text semantic understanding and extraction capabilities of large language models, overcoming the difficulty of extracting and utilizing the large amount of prior knowledge in text form from power systems for intelligent agents.

The application provides a heat-aware storage engine performance optimization method, which comprises the following steps: determining a target database, initializing a maximum heap in the memory of the target database, and maintaining the file heat of the system; reading file heat information from the log of the target database, and recovering the file heat identification mechanism based on the maximum heap; writing the file heat information into a Memtable of an LSM-tree memory component by a storage engine, and making the data of the Memtable fall to the disk and generate a CuckooTable file; establishing a Bloom filter for the CuckooTable file on an NVM; recording the corresponding relationship between the Bloom filter and the file in an index queue; and updating the index queue when the number of files in the LSM-tree reaches a preset threshold. The method provided by the application can be used to optimize the performance of a storage engine designed with a LSM-tree structure.

This invention provides a gaming machine that partially unifies the backup and recovery process, which restores predetermined information to its pre-game state, with the backup process, which backs up the predetermined information, thereby reducing the pressure on the program area capacity of the main ROM. [Solution] In the game start state control process of the main process, a backup of various information is acquired, and further, in the game end state control process of the main process, a process is provided to restore the backed-up various information according to predetermined conditions. In this process, when backing up and when restoring the various information, the values ​​of the registers that store the addresses of the respective data storage areas are swapped, thereby making some of the programs common.

The invention discloses an uplink sensing iterative optimization method based on a cellular-free system, which is used for solving the problems that the sensing precision depends on the communication demodulation quality and the communication performance is limited by the sensing result in an ISAC system, and is suitable for future 6G scenes of large-scale terminal connection and high-precision sensing cooperation. The method comprises the following steps: constructing a multi-signal collaborative sensing module, fusing pilot frequency initial sensing and data signal enhanced sensing, and extracting target parameters; constructing a dynamic channel reconstruction module, and mapping a sensing result into channel state information required by communication; and constructing a robust data demodulation module, recovering communication data by using channel information, and feeding back a demodulation error to correct a sensing result. The above modules form a closed-loop feedback mechanism, and the sensing precision and the communication reliability are gradually improved in the iteration process. Compared with the prior art, the method can effectively reduce the bit error rate and improve the sensing positioning precision, has the advantages of low cost, high accuracy and the like, and has wide practical application value and engineering popularization prospect.

A method performed by a network control relay (NCR) including a network control relay (NCR) mobile terminal (NCR-MT) and an NCR forwarding (NCR-FWD) in a wireless communication system is provided. The method comprises: receiving, by an NCR-MT, side control information from a base station; detecting a beam failure of the NCR-MT; stopping forwarding by the NCR-FWD based on detecting a beam failure; identifying the completion condition of beam failure recovery; based on completion of the beam failure recovery, the forwarding is resumed using the side control information received prior to detecting the beam failure.

The invention provides a hyperspectral image reconstruction method and system based on flow matching prior assistance, and the method comprises the steps: generating a low-dimensional condition vector based on compressed two-dimensional test data, and enabling the low-dimensional condition vector to be used for constraining a flow matching model process, and obtaining a hidden space vector; setting an initial iteration frequency and an initial signal vector, and determining a sensing matrix and measurement data; and based on the initial signal vector, the sensing matrix and the measurement data, performing loop iteration of N stages until a final reconstructed hyperspectral image is obtained, wherein the k-stage iteration specifically comprises two processes of generalized alternating projection and a de-noising device. And after N stages of iteration are completed, taking the hyperspectral image reconstruction estimation value of the Nth stage as a final reconstructed hyperspectral image. According to the method, the continuous probability modeling capability of stream matching is utilized, the spatial-spectral information recovery capability of the hyperspectral image is enhanced, and the problems of spectral distortion and spatial blur in a complex degraded scene in a traditional method are avoided.

A method for activating a connection, a device, a system, and a storage medium, the method including: a terminal device in an inactive state receives recovery configuration indication information, the recovery configuration indication information indicates that the terminal device recovers, does not recover, or partially recovers a radio resource control connection, and confirms, according to the recovery configuration indication information, whether to recover the radio resource control connection. By recovering or partially recovering the radio resource control connection according to the recovery configuration indication information, the embodiments can quickly activate the radio resource control connection between the terminal device in the inactive state and a network device, the data transmission delay is reduced, thereby complying with low-latency requirements of high-reliability and low-latency communication.

The application belongs to the field of computer vision, and provides a low-resolution pedestrian re-identification algorithm based on magnification learning, and proposes a new non-end-to-end magnification adaptive fusion and discriminative feature regression network model, which learns to restore different magnification low-resolution pedestrian images, and can extract pedestrian discriminative information for image retrieval. Specifically, a multi-task super-resolution identification learning network is designed, which includes a magnification adaptive super-resolution sub-network and a distillation identification sub-network. The magnification adaptive super-resolution sub-network can adaptively fuse contour information and detail information, and can restore pixel information for low-resolution images of different magnifications, reducing noise interference in pedestrian images. The distillation identification sub-network has an identification loss and a feature structure loss to guide the network to extract pedestrian discriminative features. A large number of experiments show that the RAMN model proposed by the application has reached the latest level in accuracy and efficiency.

Example embodiments of the present disclosure relate to monitoring a data processing model for channel information recovery. According to embodiments of the present disclosure, a solution for monitoring a data processing model for channel information is proposed. Two parallel measures by the data processing model for both the uplink and the downlink channels are used to determine whether the data processing model is available. In this way, it does not require extra feedbacks, which significantly reduces the overhead in the communication system and the computation efforts.

The invention relates to the technical field of data recovery, in particular to a network disconnection data reconnection recovery method and system, and the method comprises the steps: obtaining integrated form information and user network information, determining whether a network applied by a user has disconnection abnormity or not according to the user network information, and if yes, generating and sending network abnormity information, and performing form analysis processing on the form information input by the user and the form information not input by the user to obtain a first data form and a second data form. After a submission instruction triggered by a user is detected, first form data and second form data are collected, the first form data are input into the first data form, the second form data are input into the second data form, and a first reference form and a second reference form are obtained. And after the network information of the user is recovered to be normal, performing form perfection on the form information input by the user and the form information not input by the user. According to the invention, the accuracy and integrity of data recovery processing are improved.

The invention provides a communication method and device. In the method, a second device sends first indication information to a first device, the first indication information is used for indicating an AI model attribute of the second device, and an AI model of the second device can be deployed on a terminal side and can also be deployed on a network side. Through the method, the AI model attributes of the terminal side and the network side tend to be consistent, and the robustness of channel information recovery performance can be improved.

The embodiment of the invention discloses an image super-resolution reconstruction method and device, equipment, a medium and a product. The method comprises the following steps: determining a first target image and a second target image; wherein the first target image is a low-resolution image, and the second target image is a high-resolution image; performing feature fusion processing on the shallow feature information corresponding to the first target image and the shallow feature information corresponding to the second target image to obtain first specified feature information corresponding to the first target image and second specified feature information corresponding to the second target image; performing high-frequency information recovery processing on the first specified feature information and the second specified feature information, and performing feature fusion on the feature information after the high-frequency information recovery processing and shallow feature information to obtain target feature information; and performing up-sampling processing on the target feature information to obtain a high-resolution image corresponding to the first target image.

The invention discloses a decoupling cooperative enhancement method for an underwater image degradation factor, and belongs to the technical field of underwater image enhancement. Aiming at the problem that an underwater imaging environment is complex and changeable, and light propagation is influenced by wavelength selective absorption and suspended particle scattering to cause image degradation, a degradation factor is approximately decomposed into relatively independent scattering degradation and color cast degradation, and a binarization strategy is introduced to compress the storage and calculation amount of data, so that the image degradation is realized. And global information is compensated in back propagation through decoupling learning, and data compression precision loss is recovered. According to the decoupling collaborative optimization method, scattering and color problems are distributed to a physical compensation module and a spectrum compensation module, learning information is isolated through a gating mechanism, and degeneration decoupling is achieved; the loss function is combined with the subtasks and the global loss to compensate the binarization precision loss of the submodules, and finally, the modules are cascaded to process and recover the degraded image.

The invention provides a broadband response distributed sound wave sensing method and device, and relates to the technical field of optical fiber sensing. Single-frequency continuous light beams are split, double-sideband frequency sweep modulation is performed on a first path to generate double-sideband frequency sweep light, short-pulse-width chirped pulse light and long-pulse-width chirped pulse light are obtained through frequency band selection and pulse interception, and the short-pulse-width chirped pulse light and the long-pulse-width chirped pulse light are synthesized and injected into a sensing optical fiber; backward Rayleigh scattering light signals are collected, long-pulse-width and short-pulse-width Rayleigh scattering light signals are separated through optical filtering, and spontaneous emission noise is suppressed and amplified; coherent beat frequency detection and phase demodulation are carried out on the long-pulse-width Rayleigh scattering light signal, spectrum correlation processing is carried out on the short-pulse-width Rayleigh scattering light signal to obtain time offset, disturbance information recovery is achieved, and a sound wave measurement result distributed along the sensing optical fiber is output.