33 results about "Coding distortion" patented technology

The invention discloses a multi-level full RGB just perceptible perception coding distortion prediction method for video compression, and the method comprises the steps: firstly obtaining a public video set, processing videos in the video set into RGB videos, and obtaining a perception coding distortion label; then extracting visual perception features based on the RGB video; and finally, constructing an RGB-JNCD-NET prediction model, and outputting a coding distortion prediction result by the visual perception features through the RGB-JNCD-NET prediction model. According to the method, high-efficiency and accurate distortion coding prediction can be realized without influencing the perception quality.

The invention discloses a video target tracking system and method combined with deep learning, and relates to the technical field of computer vision, and the method comprises the following steps: obtaining video frame sequence data of each monitoring device; performing segmentation processing on the time coding sequence of the video frame sequence data to obtain track fragments of corresponding monitoring equipment; and extracting a time coding subsequence of a track fragment pair formed by any two track fragments, and carrying out feature analysis on the time coding subsequence to obtain a multi-dimensional matching confidence score of each track fragment pair. According to the scheme, a shielding time window detection mechanism is introduced, shielding influences are actively recognized and quantified, the problem of time coding distortion caused by shielding in a multi-monitoring-device environment is solved, a similarity calculation result can be dynamically corrected, mismatching caused by local shielding is avoided, and the accuracy of time coding is improved by combining motion continuity constraint and a nonlinear fusion strategy. The track matching robustness in a complex shielding scene is effectively improved, and a high-confidence matching reference is provided for subsequent time axis unification.

A hybrid codec for training a neural coding system, the hybrid codec including a real codec for encoding data pre-processed by a first neural network, and for decoding the encoded data for post-processing by a second neural network; and a proxy codec that is a differentiable representation of the real codec, and is for back-propagating a loss function, representative of a coding bit-rate and a coding distortion between the post-processed data and the original data, for training the weights of at least the first neural network.

The invention relates to the technical field of satellite communication transmission optimization, and discloses a multi-band collaborative satellite communication remote data transmission system. According to the system, terminal historical data are processed through waveform decomposition, steady-state and transient characteristics are extracted, and a terminal-level dynamic transmission characteristic spectrum is constructed in combination with real-time parameters. And carrying out feature level comparison on the feature spectrum and ground station measured data, fusing time delay, frequency offset and coding distortion differences to generate a difference fusion spectrum, mapping the difference fusion spectrum to a satellite network topology, and identifying an abnormal propagation path and a corresponding physical region. And then activating multi-frequency-band cooperative monitoring, and completing accurate positioning and self-adaptive response of an abnormal region by adjusting satellite node parameters and applying controllable disturbance. According to the scheme, terminal-level behavior characteristics and network topology mapping are utilized, accurate diagnosis from signal tracing to a space path is achieved, and the accuracy of satellite network transmission anomaly management and the autonomous response capability of a system are improved.

The disclosure relates to methods of generating or modifying a bitstream. These methods comprise obtaining (e.g., generating or receiving) the bitstream, wherein the bitstream contains coded time series data, determining a measure of distortion in relation to a portion of the coded time series data, and embedding distortion metadata into the bitstream, wherein the distortion metadata includes information indicative of the determined measure of distortion. The disclosure further relates to methods of transcoding or decoding such bitstream, as well as corresponding apparatus, programs, and computer-readable storage media.

A hybrid codec for training a neural coding system, the hybrid codec comprising: a real codec for encoding data pre-processed by a first neural network and for decoding the encoded data for post-processing by a second neural network; and a proxy codec being a differentiable representation of the real codec and configured to back propagate a loss function representing the coding bitrate and coding distortion between the post-processed data and the original data, for training weights of at least the first neural network.

The application belongs to the technical field of optical communication, and discloses an environment-adaptive optical signal wavelength conversion and regeneration system and method; transient power fluctuation characteristics and rate characteristics are obtained by performing feature extraction on input signal light; the thermal environment state of a semiconductor optical amplifier is monitored by using reference light to determine a gain drift amount; the probe light power is dynamically configured based on the gain drift amount and the rate characteristics, and transient fine adjustment is performed on the bias current according to the transient power fluctuation characteristics; the input signal light and the dynamically configured probe light are coupled into the semiconductor optical amplifier, and conversion light is obtained by using cross gain modulation effect; the conversion light is subjected to code distortion evaluation, and pre-emphasis instructions are generated and fed back to the input signal light modulation link. The application realizes automatic compensation under environment change and effectively suppresses code distortion.

The application provides an edge device-oriented video compression method, and relates to the technical field of rescue videos, and the method comprises the following steps: collecting a collapsed roadway video and distance data to form a front passage zone, identifying an obstacle edge and determining a remaining passage width and an interframe offset; dividing an avoidance section, a judgment section and an exploration section according to a passage width threshold; determining an obstacle edge retention range and an encoding level; performing block encoding and background multiplexing with a previous frame background sample; adjusting a next frame retention range according to an encoding distortion degree, and generating a reconstructed picture by a control end. The application can preferentially maintain the obstacle edge and the passage space required for rescue passage judgment when the bandwidth and the edge computing power are limited.

To provide a coding device, a decoding device, a coding method, a decoding method, a coding program, and a decoding program, capable of suppressing a dark part coding distortion by a gray level correction.SOLUTION: A coding device 200 includes: a gray scale correction part 201 that corrects a gray scale of a RAW image data from an imaging element having an optical black on the basis of a gamma coefficient and a value of the optical black; and a coding part 202 that codes a gray scale correction RAW image data where a gray scale is corrected by the gray scale correction part. A decoding device includes: an acquisition part that acquires a coded RAW image data obtained by coding the gray scale correction RAW image data obtained by correcting the gray scale on the basis of the gamma coefficient and the value of the optical black; a decoding part that decodes the coded RAW image data acquired by the acquisition part to the gray scale correction RAW image data and a gray scale reverse correction part that reversely corrects the gray scale of the gray scale correction RAW image data decoded by the decoding part on the basis of the gamma coefficient and the value of the optical black, and outputs the RAW image data before the correction of the gray scale.SELECTED DRAWING: Figure 2

A hybrid codec for training a neural coding system, the hybrid codec comprising: a real codec for encoding data pre-processed by a first neural network, and for decoding the encoded data for post-processing by a second neural network; and a proxy codec that is a differentiable representation of the real codec, and is for back-propagating a loss function, representative of a coding bit-rate and a coding distortion between the post-processed data and the original data, for training the weights of at least the first neural network.

The application provides a video coding processing method, device and equipment and a storage medium. The method comprises the following steps: calculating rate-distortion cost values of each coding mode of H265 intra prediction of a target coding unit; selecting a plurality of coding modes from the coding modes as a first mode queue according to the depth of the target coding unit and the rate-distortion cost values of the coding modes; determining a plurality of optimal prediction modes of the target coding unit as a second mode queue; determining a target coding mode with the minimum coding distortion from the first mode queue and the second mode queue; stopping division of a smaller unit for the target coding unit if the rate-distortion cost value of the target coding mode is less than a preset threshold corresponding to the target coding mode; and performing coding processing of intra prediction on the target coding unit according to the target coding mode. The application can improve the coding processing speed and efficiency of frame prediction video coding while ensuring the coding quality of intra prediction.

The present application belongs to the field of video perceptual coding optimization, and discloses a multi-level multi-granularity perceptual coding distortion prediction method, comprising the following steps: step 1: visual perception effect and perceptual coding distortion mapping analysis: constructing multi-level just noticeable quantization parameter dataset and just noticeable coding distortion dataset of each source video; step 2: multi-level multi-granularity perceptual coding distortion prediction: based on visual perception mechanism, qualitatively analyzing the mapping relationship between each visual perception characteristic and just noticeable coding distortion by using statistical analysis method. The present application solves the problem that different perceptual effects are not completely consistent in the perceptual effect on compressed video, increases the difficulty of perceptual coding distortion theoretical analysis under the joint constraint of video coding mechanism and visual perception mechanism, and cannot deduce the ideal JNCD threshold model by using traditional theoretical modeling, and meets the demand of multi-granularity perceptual coding optimization.

A hybrid codec for training a neural coding system, the hybrid codec comprising: a real codec for encoding data pre-processed by a first neural network and for decoding the encoded data for post-processing by a second neural network; and a proxy codec being a differentiable representation of the real codec and configured to back propagate a loss function representing the coding bitrate and coding distortion between the post-processed data and the original data, for training weights of at least the first neural network.

The invention relates to the technical field of video coding, in particular to a color gamut mapping and coding collaborative optimization method for an R scene, which comprises the following steps of: constructing a dynamic color gamut perception model, and generating a local color gamut mapping function adaptive to a current scene; executing preliminary color gamut mapping; extracting chromaticity gradient, saturation change rate and spatial frequency characteristics to construct a chromaticity distortion sensitivity graph; embedding the graph into an encoder rate distortion optimization module, and dynamically adjusting quantization parameters and conversion unit division; performing compression by adopting a standard coding framework; and iteratively optimizing the color gamut mapping function through a coding distortion feedback channel. According to the method, the color fidelity and the compression efficiency are remarkably improved, the average CIEDE2000 color difference is reduced by 32% under the same code rate, and multi-terminal heterogeneous display adaptation is supported.

The invention relates to a method for improving accuracy of voiceprint recognition of compressed coding voice, which belongs to the technical field of voiceprint recognition, and comprises the following steps: respectively extracting voice features of uncompressed voice and voice after voice coding and compression; based on the general background model, performing adaptive training by using the uncompressed voice features and the compressed voice features to obtain a corresponding uncompressed voice model and a corresponding compressed voice model; determining distortion parameters introduced in the encoding process according to the difference between the uncompressed voice model and the compressed voice model; compensating an uncompressed voice model based on the distortion parameter to obtain a universal background model of voice coding distortion compensation; and correcting the compressed voice by using the universal background model for voice coding distortion compensation, and executing voiceprint recognition based on the corrected features so as to improve the recognition accuracy. According to the invention, the accuracy of voiceprint recognition in mobile communication can be effectively improved.

In a video coding and decoding method, in a case that a transmission rate is low, image quality is deteriorated due to coding distortion, but the image quality can be improved owing to post-filter processing using a neural network. However, a problem is that there may be no method of defining a strength control value to be input in the post-filter processing. A video decoding apparatus according to an aspect of the present invention includes an image decoding apparatus that generates a decoded image, and a post-filter processing apparatus that performs post-filter processing on the decoded image. In accordance with an input format for luminance and chrominance of the decoded image to be input in the post-filter processing, a strength control value obtained in a process of generating the decoded image is transformed, and the post-filter processing is performed.

Disclosed is an apparatus and method for audio encoding / decoding that is robust against coding distortion in a transition section. An audio encoding method includes outputting a frequency domain signal by time-to-frequency (T / F) transform of an input signal, outputting a frequency domain residual signal in which a frequency axis envelope is removed from the frequency domain signal by applying frequency domain noise shaping (FDNS) encoding to the frequency domain signal, outputting a time domain residual signal in which a time axis envelope is removed by performing linear prediction coefficient (LPC) analysis based on the frequency domain residual signal, and quantizing and transmitting the time domain residual signal.

The present disclosure relates to the field of video coding. The embodiments of the present disclosure disclose a digital video coding method and apparatus. The digital video coding method includes: using a transform kernel to transform and quantize a target transform block of a video frame to obtain coefficients of the target transform block; in response to determining that the number of non-zero coefficients of the target transform block and the transform kernel do not satisfy a preset mapping relationship, determining a target adjustment method for adjusting the coefficients of the target transform block, wherein the encoding distortion cost of the target transform block after adjusting the coefficients according to the target adjustment method satisfies a preset distortion condition, and the number of non-zero coefficients of the target transform block after adjusting the coefficients according to the target adjustment method and the transform kernel satisfy a preset mapping relationship; adjusting the coefficients of the target transform block using the target adjustment method to obtain an encoding result of the target transform block. This method can effectively control the distortion caused by the coefficient adjustment of the transform block.

The invention provides a data encoding method, a data decoding method, terminal equipment and a server, relates to the technical field of electronic information, and can solve the problem of low image encoding processing efficiency. According to the specific technical scheme, the coding distortion rate is dynamically adjusted according to the visual sensitive value of the image content, so that the coding duration and the code stream size are reduced and the transmission speed is improved while the visual quality is maintained or improved by more effectively utilizing the characteristics of a human visual system.

The application discloses a mode decision method and device for chroma residual coding, and the method comprises the following steps: performing residual processing and division on an original pixel image and a preset target pixel image to obtain a sub-block; calculating the coding distortion measure of the image and the SATD value under each mode according to the sub-block; firstly, determining the skip mode or the zero mode as the optimal decision mode according to the coding distortion measure; then, determining the skip mode, the zero mode or the macro-block decision mode as the optimal decision mode according to the SATD value; and encoding the target object according to the optimal decision mode to obtain an encoding result. The application provides a mode decision method and device for chroma residual coding, and the optimal decision mode can be selected by comparing the coding distortion measure and the SATD value, the target object is encoded by the optimal decision mode, and the encoding result is obtained, so that the problem of high encoding complexity caused by the original pixel image in the original pixel image being encoded in the original mode can be solved.

A hybrid codec for training a neural coding system, the hybrid codec comprising: a real codec for encoding data pre-processed by a first neural network, and for decoding the encoded data for post-processing by a second neural network; and a proxy codec that is a differentiable representation of the real codec, and is for back-propagating a loss function, representative of a coding bit-rate and a coding distortion between the post-processed data and the original data, for training the weights of at least the first neural network.

The invention discloses a filtering method, a filtering model training method and a related device, and belongs to the technical field of coding and decoding. The method comprises the steps of determining K groups of filtering models according to quantization parameters of a target image, determining a reconstruction block corresponding to a current coding block in the target image, determining a target filtering model from the K groups of filtering models, and filtering the reconstruction block based on the target filtering model. Since the same group of filtering models is suitable for coding blocks with the same coding quality, different groups of filtering models are suitable for coding blocks with different coding qualities, and different filtering models in the same group of filtering models are suitable for coding blocks with different contents. Therefore, for the reconstruction block corresponding to the current coding block, the target filtering model can be selected from the K groups of filtering models in combination with the coding quality and the content of the coding block, and then the reconstruction block is filtered based on the target filtering model, so that the coding distortion is reduced, and the filtering performance is improved.

To solve a problem in which, when a moving image is compressed, coded, transmitted, and decoded, and then image recognition processing is performed on the decoded image, especially when the transmission rate is low, the accuracy of image recognition is significantly reduced due to coding distortion contained in the decoded image, and a method of integrating the description of the analysis result of the moving image and the coding of the moving image is not sufficient in terms of coding efficiency, and it is not possible to realize a low transmission bit rate.SOLUTION: A moving image encoding device includes an image decoding device that decodes an image from encoded data, and an auxiliary information decoding device that decodes auxiliary information that indicates at least one of the position, size, and type of a recognition target in the image decoded by the image decoding device.SELECTED DRAWING: Figure 1

The application discloses an image processing method and device, electronic equipment and storage medium, and belongs to the technical field of streaming media. The method comprises the following steps: encoding image data by using a first resolution to obtain first code stream information and a first quantization parameter; determining a first decision quantization parameter value of the image data based on the first resolution and a second resolution, the first decision quantization parameter value being a critical value of coding distortion of the image data at the first resolution, and the second resolution being smaller than the first resolution; and processing and outputting second code stream information of the image data based on the first quantization parameter and the first decision quantization parameter value, the second code stream information being the first code stream information or code stream information obtained by encoding by using a resolution smaller than the first resolution.

The embodiment of the invention provides a decoder and a carrier rocket, and the decoder comprises a jump edge detection module which is used for receiving a voltage signal which is coded by adopting a Manchester II code and is converted by a physical layer transceiver, an integrator which is used for carrying out integration by adopting a positive value when the received RX is at a high level so as to obtain an integral value, when the received RX # is at a high level, a negative value is adopted for integration to obtain an integral value; the synchronous head detection and window comparator is used for pre-judging the synchronous head according to the integral value and the starting time of the jump edge of the RX or the starting time of the jump edge of the RX # by utilizing the symmetry of the synchronous head, determining a code element value corresponding to a data bit according to the occurrence of the integral value in an upper judgment window or a lower judgment window and the jump edge of the RX or the RX # in sequence, and sending the code element value to the synchronous head detection and window comparator; and when the data bits behind the synchronous head are decoded or the decoding is completed, the judgment of the synchronous head is successful. The problem of serious code distortion caused by overlong bus can be solved, and the synchronous head and the data bit can be reliably decoded.

The application discloses a kind of multi-level multi-module collaborative video sensing coding optimization method and device, and the derivation of frame level coding distortion prediction and frame level quantization parameter is carried out by original video coding distortion prediction;The image of original video is intraframe / interframe prediction, and the difference calculation is carried out to the predicted image and original image obtained, and residual image is obtained, and the residual image is filtered by the predicted coding distortion, and the residual image after filtering is based on residual block transformation, and then according to the predicted frame level coding distortion and frame level quantization parameter, perceptual quantization is carried out;Rate distortion optimization is carried out based on perceptual quantization parameter, and intraframe / interframe prediction is optimized;Perceptual quality enhancement network is constructed, and is used to optimize intraframe / interframe prediction;After the image of original video is predicted, difference calculation, residual filtering, transformation, perceptual quantization based on optimized intraframe / interframe prediction, entropy coding is carried out.

This application discloses a filtering method, a filtering model training method, and related apparatus, belonging to the field of encoding and decoding technology. The method includes: determining K sets of filtering models based on the quantization parameters of the target image; determining the reconstructed block corresponding to the current coded block in the target image; determining a target filtering model from the K sets of filtering models; and filtering the reconstructed block based on the target filtering model. Since the same set of filtering models is suitable for coded blocks of the same coding quality, different sets of filtering models are suitable for coded blocks of different coding qualities, and different filtering models within the same set are suitable for coded blocks with different content, for the reconstructed block corresponding to the current coded block, a target filtering model can be selected from the K sets of filtering models by combining the coding quality and the content of the coded block, and then the reconstructed block can be filtered based on the target filtering model, thereby reducing coding distortion and improving filtering performance.

The application provides a video encoding method and device, electronic equipment and medium, and relates to the technical field of video encoding. The method comprises the following steps: determining a prediction encoding distortion degree of a to-be-encoded unit based on pixel data of a reference encoding unit in a to-be-encoded video, the reference encoding unit being an already-encoded encoding unit in the to-be-encoded video, and the encoding unit being any one of an encoding block, an encoding frame and an encoding frame group in the to-be-encoded video; determining a rate-distortion trade-off coefficient of the to-be-encoded unit based on the prediction encoding distortion degree and a reference rate-distortion parameter; determining an encoding quantization value of the to-be-encoded unit based on the rate-distortion trade-off coefficient and a preset distortion degree mapping relationship, so as to encode the to-be-encoded unit according to the encoding quantization value and obtain a target encoding video. The method realizes reasonable allocation of encoding quantization values to video frame images by taking the encoding distortion degree as a prediction basis, and effectively improves the stability and adaptability of the video encoding effect.

A hybrid codec for training a neural coding system, the hybrid codec including a real codec for encoding data pre-processed by a first neural network, and for decoding the encoded data for post-processing by a second neural network; and a proxy codec that is a differentiable representation of the real codec, and is for back-propagating a loss function, representative of a coding bit-rate and a coding distortion between the post-processed data and the original data, for training the weights of at least the first neural network.

To provide a video encoder capable of selecting optimal resolution, and a video encoding program.SOLUTION: A video encoder according to one embodiment has a video reducing part which reduces resolution of an input video and outputs video data of a plurality of videos different in resolution. Also, the video encoder performs parallel encoding of respective video data of a plurality of videos of different resolutions and outputs video data of the plurality of encoded videos. Moreover, the video encoder selects as an optimal resolution resolution where encoding distortion of the encoded video becomes minimum when a bit amount of the encoded video data is a target rate with respect to video data of the respective encoded videos, and has a resolution selection part outputting the encoded video data of the selected resolution.SELECTED DRAWING: Figure 1