121 results about "Lossy compression" patented technology

The invention provides a structured memory data processing method and system oriented to long time sequence interaction, and is applied to the technical field of natural language processing and artificial intelligence memory modeling. The method comprises the following steps: acquiring a real-time interaction data stream, partitioning the real-time interaction data stream into dialogue data blocks, loading a hierarchical attribute mode and a previous time step attribute tree instance, inputting serialized texts of the hierarchical attribute mode and the previous time step attribute tree instance into a generative model, generating a writing, rewriting, deleting or null operation instruction aiming at a leaf node path, analyzing, updating, generating a current attribute tree instance, and storing the current attribute tree instance; by means of the scheme, lossy compression and structured evolution of the infinite long dialogue stream can be achieved, memory forgetting is relieved on the premise that a context window is not expanded, and long-time-sequence information retrieval precision and storage efficiency are improved.

A method of encoding data includes determining, by an encoder, a first data instance by corrupting the data with Gaussian noise. The method also includes determining, by the encoder, information representative of one or more conditional distributions. The method additionally includes determining, by the encoder, an index of a corrupted data instance of the sequence of progressively less corrupted data instances. The index corresponds with a conditional distribution of the one or more conditional distributions which causes the corrupted data instance to have a desired bit-rate. The method further includes transmitting, from the encoder to a decoder, the first data instance and the information representative of the one or more conditional distributions to enable the decoder to recover the corrupted data instance having the desired bit-rate and use the corrupted data instance to generate output data representative of the data.

A system and methods for upsampling of decompressed genomic data after lossy compression using a neural network integrates AI-based techniques to enhance compression quality. It incorporates a novel deep-learning neural network that upsamples decompressed data to restore information lost during lossy compression, taking advantage of cross-correlations between genomic data sets.

The invention provides an electroencephalogram data transmission method combining lossless and lossy compression, belongs to the field of electroencephalogram data transmission and compression processing, and is used for solving the problems of low transmission efficiency of pure lossless compression and easy loss of key diagnosis information of pure lossy compression in related technologies. According to the method, multi-dimensional features of electroencephalogram data are extracted, a key area and a non-key area are divided, lossless compression and lossy compression processing are adopted respectively, the area division proportion is dynamically adjusted in combination with transmission bandwidth, signal features and application scenes, and corresponding data are transmitted through dual protocols. The transmission efficiency can be improved while the clinical diagnosis marker and the core characteristics are reserved, and the scene adaptability and the transmission reliability are achieved at the same time.

One example method includes at a source, at a source, performing a symbolic aggregation process on a series of raw data generated and / or collected by the source, to create a series of symbols, inputting, by the source, the series of raw data and the series of symbols to a lossy compression algorithm operating at the source, running, at the source, the lossy compression algorithm to obtain a series of raw values, and a sparse series of raw values, and transmitting, by the source to a target, the series of raw values, and the sparse series of raw values.

Methods, systems, and devices for wireless communications are described. A user equipment (UE) may receive a control message that indicates one or more parameters for lossy compression of a feedback payload. The one or more parameters may indicate at least a compression ratio between an original feedback payload size and a compressed feedback payload size. The one or more parameters may also indicate a bundle size, a quantity of bundles, a quantity of most likely outcomes to distinguish, a quantity of negative acknowledgments (NACKs) to be distinguished, or a combination thereof. The UE may monitor for one or more downlink transmissions associated with the feedback payload and transmit the feedback payload based on the one or more downlink transmissions. The feedback payload may be compressed from the original feedback payload size to the compressed feedback payload size in accordance with the one or more parameters.

A device for decoding video data includes an integrated circuit (IC) comprising a video decoder, and a memory that is external to the IC and coupled to the IC. The video decoder is configured to in a first mode, decode a first frame based on a first reference frame stored in the memory, and in a second mode, decode a second frame based on a second lossy reference frame, wherein the second lossy reference frame is generated based on decompression of a lossy compressed reference frame.

Lossy methods and hardware for compressing data and the corresponding decompression methods and hardware are described. The lossy compression method comprises dividing a block of pixels into a number of sub-blocks and then analysing, for each sub-block, and selecting one of a candidate set of lossy compression modes. The analysis may, for example, be based on the alpha values for the pixels in the sub-block. In various examples, the candidate set of lossy compression modes comprises at least one mode that uses a fixed alpha channel value for all pixels in the sub-block and one or more modes that encode a variable alpha channel value.

Deep learning data compression using multiple hardware accelerator architectures is provided herein. A system includes a computing device and first and second hardware accelerators coupled thereto. The first and second hardware accelerators may be of different types, such as a tensor streaming processor and a field programmable gate array. The first and second hardware accelerators may be directly connected to one another, such as by a chip-to-chip connection. The first and second accelerators may implement different stages of a data pipeline, such as lossless and lossy compression stages of a learned image compression.

This invention provides an end-to-end voice encryption method, apparatus, and Bluetooth headset suitable for multi-hop lossy channels. The method includes front-end noise reduction, framing, and segmentation of the acquired voice at the transmitting end. Based on the session mode, a group public key or a point-to-point private key is automatically selected as the master key from a pre-configured key set. Then, multiple time-domain segments within each frame are scrambled, subjected to segment-by-segment Fast Fourier Transform, and frequency-domain encryption based on subkeys, according to the master key. The encrypted voice is then generated through inverse transformation and overlapping weighted smoothing concatenation, and transmitted through a multi-hop voice communication channel containing multi-level lossy encoding and decoding. At the receiving end, the reverse process is performed: framing and segmentation, transformation, inverse frequency-domain encryption, and inverse time-domain scrambling. Combined with U-shaped neural network noise reduction at both the front-end and back-end, the voice is restored to intelligible speech. This method enables flexible key management and high-quality secure voice transmission in scenarios where group calls and point-to-point private calls coexist and undergo multiple lossy compression operations.

The embodiment of the application discloses a lossy compression method, a decompression method and equipment for time sequence floating point data, which are used for improving the data compression rate and the decompression speed of the time sequence floating point data. The method comprises the following steps: acquiring a floating point array, the floating point array comprising a plurality of floating point data arranged in time sequence; sequentially determining other floating point data in the floating point array as target floating point data, and determining a previous floating point data of the target floating point data from the floating point array; determining a difference between the target floating point data and the previous floating point data as a difference data of the target floating point data; performing quantization processing on the difference data based on a preset error boundary to obtain quantized difference data corresponding to the target floating point data, wherein a difference between lossy data corresponding to the target floating point data and the target floating point data is not greater than the preset error boundary, and the lossy data corresponding to the target floating point data is obtained by performing inverse quantization processing on the quantized difference data corresponding to the target floating point data; and performing compression processing on each quantized difference data.

A data storage system (90) that stores data that is lossy compressed includes a lossy compression device (9). The lossy compression device (9) includes a smoothness decision unit (18) that decides smoothness according to the rarity of an event indicated by subject data, as subject smoothness, and a data smoothing unit (22) that generates smoothed subject data by smoothing the subject data with the subject smoothness.

A system and method for complex-valued radar image compression integrates AI-based techniques to enhance compression quality. It incorporates a novel AI deblocking network composed of convolutional layers for feature extraction and a channel-wise transformer with attention to capture complex inter-channel dependencies. The convolutional layers extract multi-dimensional features from the complex-valued radar image, while the channel-wise transformer learns global inter-channel relationships. This hybrid approach addresses both local and global features, mitigating compression artifacts and improving image quality. The model's outputs enable effective complex-valued radar image reconstruction, achieving advanced compression while preserving crucial information for accurate analysis.

The application discloses a kind of based on information redundancy's call chain data compression method and system, method includes: for the call chain data collected, similar data is extracted using KMeans algorithm, the data point in the call chain data is grouped, and each cluster group is obtained;Wherein, the data similarity in the same cluster group is high, and the data similarity between different cluster groups is low;Using the offline tail sampling strategy based on hierarchical clustering, the cluster group is sampled and handled, and a structured file is output;The structured file is losslessly compressed using a lossless compression method, and the compression of the call chain data is completed.The application is based on the characteristics of call chain data, combined with the sampling idea of lossy compression, greatly reduces the size and storage burden of the compressed file on the basis of maintaining data analysis value, improves efficiency and real-time performance, can guarantee data integrity and strengthen generalization ability, and can be widely applied in computer technology field.

A device for decoding video data includes an integrated circuit (IC) comprising a video decoder, and a memory that is external to the IC and coupled to the IC. The video decoder is configured to in a first mode, decode a first frame based on a first reference frame stored in the memory, and in a second mode, decode a second frame based on a second lossy reference frame, wherein the second lossy reference frame is generated based on decompression of a lossy compressed reference frame.

The invention relates to the technical field of image compression and artificial intelligence, and discloses a full-view pathological image self-adaptive differential compression method based on AI semantic importance and application of the full-view pathological image self-adaptive differential compression method. In order to solve the problems of high storage cost and loading lagging caused by a huge full-view pathological image file, the method comprises the following steps: acquiring a low-resolution thumbnail of the full-view pathological image file, inputting a pre-trained semantic evaluation model, and generating a two-dimensional probability mask comprising an extremely low importance level, a medium importance level and a high importance level; mapping the coordinates of the image to a high-resolution image and dividing the image into grid blocks; color mean values are extracted from the blocks with extremely low importance to perform bypass replacement, lossy and lossless compression is performed on the blocks with medium importance and the blocks with high importance respectively, and smooth transition areas are generated at junctions; and finally generating a compressed file embedded with the index dictionary. According to the invention, the storage occupation and transmission delay of the pathological data are effectively reduced, and the diagnosis fidelity of the core focus area is guaranteed.

A system and method for lossy compression and recovery of data are described. The original data is first truncated. The truncated data is then compressed. The compressed truncated data can then be stored and / or transmitted efficiently using fewer bits. To recover the data, the compressed data is then decompressed and the recovered bits are linked. The recovered bits are selected to compensate for the statistical bias introduced by the truncation.

In one implementation, a shifted-window transformer with sorted grouping is used for point cloud compression. The points are first sorted according to a specified order and then the sorted points are grouped into windows with an equal number of points. All resulting windows are operated upon by self-attention to obtain initial attention features, followed by shifting the windows and another self-attention. We utilize the proposed transformer for lossless compression of point clouds via the octree representation and lossy compression via feature coding. Downsampling can be used to obtain features at different resolutions, which can be combined in a multi-scale solution. In addition, in a hybrid approach, the features can be divided into two parts: one for the window attention and the other for convolutions. The same sorting strategy can be used throughout the proposed architecture, or the network can switch between different sorting strategies every few attention layers.

In one implementation, a shifted-window transformer with sorted grouping is used for point cloud compression. The points are first sorted according to a specified order and then the sorted points are grouped into windows with an equal number of points. All resulting windows are operated upon by self-attention to obtain initial attention features, followed by shifting the windows and another self-attention. We utilize the proposed transformer for lossless compression of point clouds via the octree representation and lossy compression via feature coding. Downsampling can be used to obtain features at different resolutions, which can be combined in a multi-scale solution. In addition, in a hybrid approach, the features can be divided into two parts: one for the window attention and the other for convolutions. The same sorting strategy can be used throughout the proposed architecture, or the network can switch between different sorting strategies every few attention layers.

A data storage system (90) that stores lossily compressed data includes a lossy compression device (9). The lossy compression device (9) includes a smoothing specification unit (18) that specifies a smoothing according to the rarity of an event specified by subject data as subject smoothing, and a data smoothing unit (22) that produces smoothed subject data by smoothing the subject data with the subject smoothing.

Three-dimensional sensing data is transferred efficiently by applying compressed sensing using dictionary learning. Conventional image-based lossy compression improves a feature according to which power is concentrated on low-frequency components and deterioration of three-dimensional information increases. By changing the decimation rates for the depth information and the color information and using a dictionary vector created from the other restoration result for restoration, the original result is reconstructed using a small data volume.

Systems and methods related to processing medical image data. Medical image data of a patient's breast tissue is received from a medical imaging device. The medical image data includes a plurality of X-ray projection images of the patient's breast tissue from a plurality of perspectives. The medical image data is encoded using lossy compression. The medical image data is decoded to yield a transmission image. The transmission image is encoded using lossless compression to yield a first compressed image. The first compressed image is transmitted to a first viewing device. The first viewing device is configured to decode the first compressed image to yield the transmission image and view the transmission image using the first viewing device. Viewing the transmission image includes viewing at least one of the plurality of X-ray projection images of the patient's breast or a composite image of the plurality of X-ray projection images.

The present application discloses a hyperspectral image lossy compression method based on hierarchical prediction from the fact that hyperspectral image has both spatial correlation and spectral correlation, and JPEG2000 only eliminates the spatial redundancy of hyperspectral image. The present application adopts two-dimensional wavelet transform to remove spatial redundancy, and utilizes the spectral correlation of hyperspectral image to generate high-quality SI at the decoding end to reduce the code rate required by channel coding. The hyperspectral image is divided into GOP area and non_GOP area, and the WZ waveband in the GOP area adopts hierarchical prediction structure to generate SI; the WZ waveband in the non_GOP area will be specially processed, and only one decoded waveband in the front direction is used to generate SI. The hyperspectral image lossy compression method based on hierarchical prediction disclosed by the present application can obtain better rate-distortion performance in the full code rate segment compared with JPEG2000.

Method for providing validation data (23) in a data storage system (12), wherein the validation data (23) are intended for the validation of driver assistance systems (4); wherein the data storage system (12) is configured to provide the validation data (23) for the validation of at least one driver assistance system (4); and wherein the validation data (23) comprise at least some data (3) recorded by sensors (2) in vehicles during test drives; comprising the following steps: a) receiving the validation data; b) performing compression (6) of the validation data to produce compressed validation data (11) using a lossy compression method; c) generating synthesized validation data (20) based on the compressed validation data (11) produced in step b);andd) Performing a comparison of the validation data with the synthesized validation data (20), wherein a check is carried out to see whether a deviation between the synthesized validation data (20) and the validation data received in step a) is within a specified tolerance range, wherein the tolerance range is chosen such that deviations are detected which are within a signal noise caused by properties of at least one sensor (2) with which validation data were recorded, wherein the tolerance range is further chosen such that the distance between synthesized validation data (20) and the recorded reality is not greater than the distance between the validation data received in step a) and the same reality;ande) Storing the compressed validation data (11) in the data storage system (12) if, in step d), it was determined that the deviation is within the specified tolerance range.;

Post training compression of 3DGS data is agnostic to training in a traditional signal compression perspective. Gaussian parameters are treated as signals. Pre-processing and transform coding techniques are used to compress the signals effectively. Firstly, lossless / lossy compression is performed on 3DGS geometry (positions, scales, rotations) using a point cloud coding-based (e.g., G-PCC, GeS) framework. Positions are compressed using occupancy tree coding. Scales and rotations are encoded as attributes using transform coding. The widely used block-based graph Fourier transform (GFT) is used to compress the attributes (base colors, spherical harmonic coefficients and opacities). In addition, a graph construction strategy is used for 3DGS data that computes the edge weights based on similarity (or dissimilarity) between the 3D Gaussian distributions using KL-divergence. Alternatively, positions can be encoded using occupancy tree (e.g., G-PCC, GeS) or AI-based PCC methods, and any subset of Gaussian parameters or the transformed coefficients of Gaussian parameters can be mapped into 2D frames and encoded by video coders.

The present application relates to video image communication, in particular to a kind of highway traffic safety monitoring video processing device and system, comprising: by video acquisition module, obtain road traffic video image data, utilize video compression preprocessing module, by analyzing the importance degree of each part area in video image and the important degree of loss compression threshold value obtained in video image data LBP value and transmittance are compressed before the preprocessing of video image data, and by video encoding compression module, video image communication module, video decompression module, video storage module, alarm module and visualization module realize the efficient compression and transmission data of road traffic video image and accurate analysis and early warning.The present application can protect important data while preprocessing video image to improve the compression efficiency of video image, ensure the reliability of subsequent video data analysis result, obtain accurate and reliable warning signal.

Method for storing validation data (23) in a data storage system (12), wherein the validation data (23) are intended for validating systems for processing large amounts of data; wherein the data storage system (12) is configured to provide the validation data (23) for validating at least one system for processing large amounts of data; and wherein the validation data (23) comprise at least some data recorded by sensors (2); comprising the following steps: a) receiving the validation data; b) determining (7) at least one statistical property of the received validation data; c) performing compression (6) of the validation data to produce compressed validation data (11) using a lossy compression method; d) determining at least one statistical comparison property;e) Performing a comparison of the at least one statistical property with the at least one statistical comparison property, wherein the comparison checks whether a deviation between synthesized validation data (20) producible on the basis of the compressed validation data and the validation data received in step a) lies within a predetermined tolerance range, wherein the tolerance range is chosen such that deviations are detected which lie within a signal noise caused by properties of at least one sensor (2) with which validation data were recorded, wherein the tolerance range is further chosen such that the distance between validation data (20) synthesized on the basis of the compressed validation data (11) and the recorded reality is not greater than the distance between the validation data received in step a) from the same reality;and f) Storing the compressed validation data (11) in the data storage system (12) if, in step e), it was determined that the deviation is within the specified tolerance range.;

The application discloses a lossy compression detection method and terminal fusing high-frequency space domain and DCT domain, and the method comprises the steps of obtaining a first JPEG compressed image data set, training a WebP lossy compression detection network according to the first JPEG compressed image data set to obtain a trained WebP lossy compression detection network, obtaining a second JPEG compressed image data set, performing WebP lossy compression trace detection on the second JPEG compressed image data set according to the trained WebP lossy compression detection network, and outputting a detection result; the method for extracting features of a JPEG image by fusing high-frequency space domain and DCT domain provided by the application solves the problem that the effect of the lossy compression detection method for JPEG double compression in the prior art is poor and even fails in some cases.

The invention discloses a hyperspectral image compression network and compression method based on multi-scale spectrum and spatial feature enhancement, and mainly solves the problems of insufficient hyperspectral image spectrum modeling, insufficient spatial feature extraction and high network complexity in the prior art. The network comprises a main encoder, a main decoder, a super-prior encoder, a super-prior decoder and an entropy model. The main encoder comprises a spectral attention gating data unit, a convolution unit and a multi-scale spatial adaptive feature attention enhancement unit, and is used for converting an input image into potential representation and removing spatial and spectral redundancy; the main decoder and the main encoder are symmetrical in structure; the super-prior encoder comprises a convolution layer and an activation layer and is used for extracting auxiliary information from the output of the main encoder; the super-prior decoder and the super-prior encoder are symmetrical in structure; the entropy model is Gaussian distribution based on output parameters of a super-prior decoder. After the network is trained, lossy compression of a hyperspectral image can be realized. The method reduces the network complexity and spectral distortion, improves the reconstruction quality of complex ground feature details, and is suitable for earth observation, meteorological monitoring and the like.

A service compresses a data set in accordance with a data compression rate, resulting in generation of compressed data having a first data quality. The service decompresses the compressed data, resulting in generation of decompressed data. The service filters, from the decompressed data, data that is identified as belonging to a selected class of data. The service causes an ML classifier to perform a classification operation on the data that is identified as belonging to the selected class of data. The service determines a classification accuracy of the classification operation performed by the ML classifier. The service determines whether the classification accuracy at least meets a threshold accuracy requirement. If the threshold is met, the data compression rate is increased; otherwise, it is decreased.