14 results about "Huffman decoding" patented technology

The application provides a JPEG data decoding method and device, equipment and storage medium, which can be used in the field of image processing. The method comprises the following steps: obtaining target JPEG data, marking the target JPEG data based on a preset segmentation length in units of target MCUs to obtain target JPEG segmented data; storing the target JPEG segmented data into a target SVM to obtain multiple initial decoded segments and synchronization parameter information; in response to the multiple initial decoded segments being correctly decoded, judging whether the number of target MCUs corresponding to the multiple initial decoded segments meets a preset number condition; if yes, sending a target decoding instruction to a target GPU to enable the target GPU to perform target Huffman decoding to obtain multiple target decoded segments; decoding each target DCU included in each target decoded segment to obtain target JPEG decoding data; and storing the target JPEG decoding data into the target SVM to obtain a target RGB image corresponding to the target JPEG decoding data based on the target GPU. The CPU running load is reduced, and the decoding efficiency is improved.

Various aspects of the present disclosure generally relate to wireless communication. Some aspects more specifically relate to performing reversed Huffman coding, whereby one or more Huffman decoding operations at a transmitter and performing one or more Huffman encoding operations at a receiver. In some aspects, the transmitter may assign, using a Huffman decoder, a sequence of symbols to an input binary sequence in accordance with the leaf nodes of a Huffman binary code tree. The transmitter may transmit the sequence of symbols to the receiver. The receiver may receive a sequence of symbols and use a Huffman encoder to perform compression on the received sequence of symbols using the Huffman binary code tree and thereby generate an output binary sequence.

The invention provides a coding method, a Huffman decoding method, equipment and a medium, and relates to the technical field of data compression, and the coding method comprises the steps: obtaining a target code table; the target code table comprises a plurality of target code words; wherein a plurality of target code words in the target code table correspond to a plurality of original Huffman code words in an original code table, and the code word length information of the original Huffman code words is implied in the plurality of target code words; and encoding the target code table to obtain an encoding result.

The invention discloses a high-throughput JPEG (Joint Photographic Experts Group) heterogeneous reasoning and hybrid parallel method on mobile equipment, and provides an FCG reasoning framework aiming at the problems of high decoding time consumption proportion, high cross-processor communication overhead and low processor resource utilization rate in the existing mobile terminal JPEG image recognition. And optimization is carried out by combining JPEG frequency domain information with characteristics of a heterogeneous processor of a mobile terminal. The framework comprises an off-line preprocessing stage, an off-line task division stage and an on-line task scheduling stage, wherein in the off-line preprocessing stage, processor operation data are collected, and a delay prediction model is established; in the off-line task division stage, a processor cluster is constructed, model tasks are distributed, and the task amount is adjusted; and in the online task scheduling stage, hybrid parallel pipeline identification is executed according to the scheme. According to the method, through multi-core Huffman decoding, heterogeneous network structure design and optimization of task scheduling, the identification delay is remarkably reduced, the throughput is improved, the energy consumption is reduced, and the batch JPEG image identification performance of the mobile terminal is effectively improved.

The invention provides an encoding method, a Huffman decoding method, equipment and a medium, and relates to the technical field of data compression, the encoding method comprises the following steps: obtaining an original code table, and deleting a plurality of original Huffman code words in the original code table to obtain a target code table; and encoding the target code table to obtain an encoding result.

The present application relates to the technical field of cloud computing, in particular to a JPEG image sequence parallel decoding and rendering method, comprising the following steps: based on CPU concurrent programming, adopting thread allocation strategy, carrying out Huffman decoding on the received JPEG image, and generating parallel processing frame sequence.In the present application, the CPU concurrent programming and thread allocation strategy are adopted, which greatly improves the Huffman decoding efficiency, saves the preliminary processing time, and significantly improves the throughput and efficiency through independent decoding of each thread.Further, combined with the isolated writing and video memory allocation strategy, the data is efficiently written to the GPU video memory, the data operation is reduced, the speed is improved, the parallel computing capability of the GPU makes the IDCT transformation and YUV to RGB processing rapid and efficient, the time delay is reduced, the output quality is improved, the shared video memory buffer strategy optimizes the hardware resource utilization, the application of OpenGL rendering guarantees the smoothness of rendering and visual effect, and the speed and quality of image processing are significantly improved through the comprehensive optimization, which lays a solid foundation for real-time application.

The invention provides a coding method, a Huffman decoding method, equipment and a medium, and relates to the technical field of data compression, and the coding method comprises the steps: obtaining a target code table; the target code table comprises n identification values, and the n identification values are used for indicating whether corresponding n original coding values in an original coding range exist in an original code table corresponding to the target code table; and encoding the target code table to obtain an encoding result.

A device may be configured to signal a cross-component adaptive loop filter coefficients. In one example, a device may be configured to signal cross-component adaptive loop filter coefficients for a number of cross-component filters for a chroma component using a continuous Huffman coded bit string. A device may be configured to parse a continuous bit string and calculate filter coefficient values for filter coefficients for each of a number of cross-component filters by applying a Huffman decoding process.

The application belongs to the technical field of mechanical fault diagnosis, and specifically discloses a wireless telemetry and edge diagnosis method and system based on compressed lightweight sensing, which comprises the following steps: collecting vibration and temperature signals of equipment, and pre-processing original signals; performing envelope detection on the pre-processed vibration data at the edge end, calculating the kurtosis values of the temperature data and the vibration data, and performing abnormality detection using the pre-set kurtosis values, and outputting fault information if there is abnormality; using discrete cosine Huffman dual-mode coding to compress and encode the pre-processed vibration data, and sending the encoded data by the telemetry node; and decoding and reconstructing the encoded data by the data receiving end through Huffman decoding and discrete cosine inverse transformation, and using a neural network to recognize faults. By using the technical scheme, the amount of transmission data is significantly reduced through data compression technology, thereby reducing the bandwidth requirement and transmission delay, and the fault diagnosis can be more efficiently performed.

A device may be configured to signal a cross-component adaptive loop filter coefficients. In one example, a device may be configured to signal cross-component adaptive loop filter coefficients for a number of cross-component filters for a chroma component using a continuous Huffman coded bit string. A device may be configured to parse a continuous bit string and calculate filter coefficient values for filter coefficients for each of a number of cross-component filters by applying a Huffman decoding process.

Provided herein are techniques to performing lossless compression on neural network model parameters using grouped Huffman encoding. One embodiment provides a graphics processor comprising a host interface, a graphics processing cluster including a plurality of processing resources, at least one processing resource of the plurality of processing resources configured to receive a decode table and plurality of blocks of encoded parameters of a neural network model, perform a parallel Huffman decoding operation on the plurality of blocks of encoded parameters based on the decode table to concurrently decode multiple blocks of the plurality of blocks of encoded parameters, the plurality of blocks of encoded parameters decoded into a plurality of parameter groups, and unpack each of the plurality of parameter groups into a plurality of parameters of the neural network model during an inference operation performed via the neural network model.

The application discloses a hardware compression and decompression method based on Huffman decoding table, wherein original data is compressed on a PC end first, a Huffman decoding table and compressed data are dynamically generated in the compression process, and the two parts are burned into a hardware storage device; a decompression process is carried out in an embedded system, the embedded system sends corresponding instructions and addresses to a hardware decompression module, the decompression module addresses and searches the Huffman decoding table according to the corresponding instructions to decompress the compressed data, and the decompressed data is sent into the embedded system. The open source Gzip compression code is improved, a dynamic Huffman decoding table suitable for the hardware decompression module of the application is generated in the compression process, and the dynamic Huffman decoding table is stored in a corresponding format, so that the decoding table is directly used in the decompression process, and the decompression efficiency and accuracy are greatly improved.

The invention discloses a nondestructive storage method of a fluorescence spectrum binary concentration image sequence based on multichannel bit mapping and entropy coding. Firstly, N frames of binary edge images with the same resolution are fused bit by bit according to pixels to form a fusion matrix V; and determining the storage bit width of the V, carrying out run length coding on the V, constructing a Huffman code table based on the statistical probability, carrying out entropy compression, and finally generating a compressed file. The header of the file records information such as the image size, the frame number, the data type and the CRC check code, and the integrity and the decompressibility of compressed data are ensured. And during decompression, performing Huffman decoding, inverse RLE and bitwise disassembly according to an inverted sequence to realize complete lossless recovery of an original sequence. By using the method, redundancy among multiple frames of images can be fully utilized, the compression ratio is improved, lossless reconstruction of original data is realized, and the method is particularly suitable for a binary image sequence scene with strong inter-frame correlation. In a multi-frame related scene, the compression ratio of the method is obviously superior to that of a traditional frame-by-frame PNG method, and the reproduction degree reaches 100%.

The application provides a JPEG image load positioning method based on ensemble learning, and specifically comprises the following steps: step 1, collecting N pieces of load-bearing JPEG images based on the same embedding position key and embedded using the same DCT domain steganography algorithm; step 2, performing Huffman decoding on all images to obtain a quantized DCT coefficient matrix; combining the coefficients at the same position of all coefficient blocks in each image to obtain 64 pieces of same-frequency subgraphs; step 3, extracting 8-dimensional residual feature vectors for the same-frequency subgraphs in step 2; step 4, extracting 72-dimensional SPAM feature vectors for the same-frequency subgraphs in step 2; step 5, splicing the 8-dimensional residual feature vectors in step 3 and the 72-dimensional SPAM feature vectors in step 4 into 80-dimensional feature vectors; step 6, using stacking combined with K-fold cross-validation to perform model fusion; and using the 80-dimensional feature vectors in step 5 to train to obtain a final model.