13results about How to "Guaranteed reconstruction quality" patented technology

The invention provides a channel environment adaptive sampling-semantic-channel coding joint optimization method and system, and relates to the technical field of semantic communication, a sending end performs semantic sampling on an image needing to be transmitted to obtain a semantic sampling result; the sending end inputs the semantic sampling result into an encoder to obtain an image feature vector needing to be transmitted; a receiving end receives the image feature vector influenced by the physical channel noise and preprocesses the image feature vector; the receiving end inputs the preprocessing result into a decoder to obtain an image feature vector needing semantic reconstruction; and the receiving end performs semantic reconstruction according to the image feature vector needing semantic reconstruction to obtain a reconstructed image. According to the method provided by the invention, the characteristics of an information source and a channel are combined, image transmission in a variable signal-to-noise ratio environment can be supported while semantic sampling and reconstruction are considered, and the reconstruction quality of the image is ensured.

The invention relates to a cloth flaw detection method based on high-frequency feature constraint and edge entropy minimization, and the method comprises the steps: 1, collecting a flawless and abnormal cloth training picture data set, carrying out the initialization of hyper-parameters, and carrying out the initialization and fine adjustment of model parameters; 2, randomly sampling an image sample x, obtaining vector quantized image features through a feature encoder, generating flaw noise through an anomaly generator, adding the flaw noise into original features of an image, respectively inputting features with anomalies into an image reconstruction decoder and an image restoration decoder, and carrying out image reconstruction; and reconstructing an abnormal image and an abnormal image respectively, and the like. The detection method has the advantages that the high-frequency feature constraint module and the edge entropy minimization module are introduced, so that the cloth defect anomaly detection accuracy is remarkably improved; and the high-frequency feature constraint module converts the real image and the reconstructed image into frequency domain representation by using discrete Fourier transform.

The invention discloses an X-ray thermoacoustic imaging method based on a multi-dimensional information fusion network, and relates to the technical field of artificial intelligence medical imaging. The method can adapt to rapid reconstruction of high-quality X-ray thermoacoustic images under different sampling conditions, effectively extract and utilize information of the signals and the images, guarantee the reconstruction quality through complementation of feature information in the signals and the images, reduce the time cost overhead, avoid the defects of a traditional X-ray thermoacoustic imaging method, and improve the imaging quality of the X-ray thermoacoustic images. And the development process of X-ray thermoacoustic imaging is accelerated.

The invention discloses an outdoor scene-oriented efficient multi-view three-dimensional reconstruction method and system, and belongs to the field of multi-view three-dimensional reconstruction, and the method specifically comprises the steps: preparing a multi-view three-dimensional reconstruction data set for model training; constructing a multi-view three-dimensional reconstruction network comprising an adaptive feature extraction converter module based on big kernel attention, a cost regularization module and a depth estimation module; performing training and fine tuning on the network model to obtain a final model; building a multi-view three-dimensional reconstruction system and equipment; according to the method, the problems of poor accuracy and low integrity of a reconstruction result when an existing multi-view three-dimensional reconstruction method faces an outdoor large-scale scene can be solved, and the training reasoning time and the calculation cost of multi-view three-dimensional reconstruction can be reduced; and an important foundation is laid for application of outdoor large-scale scene image data in the field of multi-view three-dimensional reconstruction and development of a three-dimensional reconstruction technology.

The invention provides a video coding method and device and a video decoding method and device, and relates to the technical field of video coding and decoding. Comprises: obtaining a student model; the student model is a model obtained by guiding a first network model to train through a teacher model, the teacher model is a model obtained by training a second network model, and when the same video is coded based on the first network model and the second network model respectively, the first network model and the second network model are selected; the code rate of the coding result of the first network model is smaller than that of the coding result of the second network model; and obtaining a coding result of the video to be coded according to the student model. According to some embodiments of the invention, in a video coding scheme for carrying out video coding by using a deep learning network model, the video reconstruction quality is ensured, and the code rate of a video coding result is reduced at the same time.

The application discloses a real-time leakage positioning monitoring system and method based on electrical resistance tomography, relates to the technical field of process tomography and industrial safety monitoring, and solves the technical problem that it is difficult to intelligently identify, three-dimensionally position and quantitatively evaluate leakage events; the system comprises a sensor array, a multi-channel data acquisition and excitation module, a real-time data processing and image reconstruction module, a leakage intelligent identification and positioning module, a man-machine interaction and data management module; the core lies in that a multi-layer nested annular array and a sparse measurement strategy are adopted to improve data acquisition efficiency, a physical information neural network is introduced to reconstruct an image, a measurement topological coding matrix and a physical residual correction module are used to reconstruct the image, and finally, a time-space dual-domain leakage probability field model is constructed, a virtual sensor network and an improved D-S evidence reasoning framework are combined to intelligently identify and position a leakage diffusion mode.

The embodiment of the invention belongs to the technical field of voice processing, is applied to a voice signal reconstruction scene, and relates to a voice signal reconstruction method, device and equipment and a storage medium thereof. Performing down-sampling processing on different sub-band features by adopting an independent encoder to obtain a low-dimensional feature vector of each sub-band feature; performing quantization processing on the low-dimensional feature vector corresponding to each sub-band feature to obtain a discrete numerical result; and inputting discrete numerical results corresponding to all the sub-band features into a decoder, and performing up-sampling recovery processing to obtain a reconstructed voice waveform corresponding to the original voice signal. According to the method, each sub-band feature is firstly subjected to down-sampling processing, then up-sampling recovery is carried out in combination with all down-sampling quantization results, and audio information in each sub-band feature is fully utilized, so that the high-frequency part of the voice is reconstructed to be more detailed, and the audio synthesis quality of intelligent voice customer service in the financial field is ensured.

A pulse camera-oriented high-efficiency and high-quality image reconstruction method and system belong to the technical field of high-speed imaging, and comprise the following steps: obtaining explicit pulse representation by an explicit pulse representation method based on internal statistics, obtaining implicit pulse representation by an implicit pulse representation method based on a neural network, and obtaining an implicit pulse representation by an implicit pulse representation method based on neural network; the pulse embedding features with more characterization force are obtained after the pulse embedding features are mixed; setting the pulse embedding features as bottom features of a feature pyramid to construct a multi-layer feature pyramid, and sequentially extracting multi-granularity feature representations from a plurality of continuous pulse substreams; symmetrically capturing interdependence between the intermediate features and the time context features by using a multi-head cross attention mechanism of a top pyramid, and obtaining interaction features; motion-intensity interaction is carried out progressively from coarse granularity to fine granularity, multiple groups of motion fields are estimated in a bottom pyramid, and joint refinement of space-time complementary information is realized. According to the invention, the precision of image reconstruction of the pulse camera is improved, and the demand of the model for computing resources is reduced.

The application belongs to the technical field of compressed sensing reconstruction, and particularly relates to a compressed sensing reconstruction method and system based on adaptive fragmentation and dynamic resource management. Original signal data of a target scene is acquired through sparse sampling; when the total length of the original signal data is greater than a preset fragmentation length parameter, the original signal data is divided into a plurality of fragmented data to form a signal fragmented set; the memory usage of the system is monitored in real time; when the memory is greater than a preset storage threshold, dictionary elements are generated and the generated results are cached; when the memory is less than or equal to the preset storage threshold, the dictionary elements are directly used after being generated, and the storage space occupied by the dictionary elements is released after being used up; then, each fragmented data is reconstructed, and a complete reconstructed signal is obtained through linear splicing and fusion. The application is suitable for processing sparse reconstruction tasks of large-scale signal, image and video data, and can significantly reduce the demand for computing resources while ensuring the reconstruction quality.

The embodiment of the invention discloses a scene reconstruction method and device, equipment and a storage medium, and the method comprises the steps: carrying out the scene reconstruction through a preset scene reconstruction mode if a current moment is a first three-dimensional scene relative to a target scene, and obtaining a current three-dimensional scene; otherwise, acquiring a plurality of first image frames and second image frames formed by acquiring the target scene at different visual angles, and acquiring a previous three-dimensional scene reconstructed based on the target scene; according to the first image frames and the second image frames, performing pose adjustment on Gaussian points in the previous three-dimensional scene to obtain a deformed three-dimensional scene; performing Gaussian point addition and deletion and Gaussian point feature optimization processing on the deformed three-dimensional scene according to each second image frame to obtain a processed optimized three-dimensional scene; and determining the optimized three-dimensional scene as a current three-dimensional scene at the current moment. According to the method, the current three-dimensional scene with higher relevance with the previous three-dimensional scene can be obtained in the time domain, so that the smoothness and continuity from the previous three-dimensional scene to the current three-dimensional scene in the time domain are ensured.

This invention discloses a deep learning-based method for super-resolution processing of airborne optoelectronic video images. It constructs a high-order dynamic degradation model that integrates basic image degradation operations with the physical degradation mechanism of airborne optoelectronics, simulating the complex degradation process of the real imaging chain through multiple rounds of degradation loops. The dynamic degradation model generates a training dataset of low-resolution and high-resolution image pairs. A super-resolution reconstruction network is constructed based on a generative adversarial network framework. The generator uses the RRDB module for deep feature extraction and reconstructs high-resolution images through multi-stage upsampling. The discriminator adopts a U-Net structure with spectral normalization and introduces semantic consistency constraints. A two-stage course learning is performed using a domain-specific training dataset. This method can accurately simulate the complex degradation process of airborne optoelectronic imaging, improve the model's generalization ability, enhance the fidelity of tactical target details, and enable real-time inference on airborne platforms.

The invention discloses an image super-resolution reconstruction method based on a geometric transformation enhanced state space model, and belongs to the technical field of image processing and computer vision. The method solves the problem of poor reconstruction quality caused by insufficient long-range dependence modeling capability, weak geometric transformation robustness and insufficient local texture detail recovery of the existing method. The method comprises the following steps: constructing a high-resolution and low-resolution image pair data set and constructing a reconstruction network; training the network by using the training set; and inputting a to-be-processed low-resolution image into the trained network to obtain a high-resolution image. The network comprises a shallow feature extraction module, a plurality of residual state space groups connected in series and a reconstruction module. Each residual state space group comprises a plurality of visual state space layers, and each visual state space layer sequentially performs geometric transformation enhancement, double-branch parallel processing, feature fusion, double residual learning and inverse geometric transformation on input. The method is used for converting a low-resolution image into a high-resolution image.

The invention provides an auxiliary stream anti-error code network transmission method for a video codec based on deep learning, and relates to the technical field of video transmission, a decoding end performs coding and decoding by using an error code stream, and in the coding and decoding process, a decoding result corresponding to the error code stream is set to be all zero to obtain a slightly distorted decoding reconstruction frame APn-1; and the decoding end takes the decoded reconstructed frame APn-1 as a reference image, refreshes a decoding buffer of the decoding end, and performs coding and decoding under the condition that all reference contents except the reference image are set as None, so as to obtain a correctly decoded reconstructed frame APn. The method is a low-error network transmission method, the transmission bandwidth requirement is reduced while the video reconstruction quality is ensured, and the error code resistance robustness of the video codec based on deep learning is enhanced.