313results about How to "Reduce the number of bits" patented technology

A motion vector encoding apparatus includes: an optimum motion vector determiner for determining optimum motion vectors for multiple reference pictures with respect to a current motion vector of a current block; a motion vector encoding mode determiner for determining a motion vector encoding mode according to whether a motion vector decoding apparatus can predict the optimum motion vectors for the multiple reference pictures; and a motion vector encoder for generating and encoding motion information on the multiple reference pictures by using the optimum motion vectors for the multiple reference pictures or preset default motion vectors for the multiple reference pictures according to the motion vector encoding mode, and generating motion vector encoded data by encoding the motion vector encoding mode. The motion vector encoding apparatus can more accurately predict a predicted motion vector of the current motion vector for multiple reference pictures, so as to improve the compression efficiency.

Disclosed is a method and device for realizing quantification in coding/decoding, which mainly comprises: firstly, quantification used in image switching coefficient matrix adopts quantification form of parameter model, the parameter model is a parameter model which needs only several parameters to present and it includes a frequency band distributing structure and frequency band parameters. Thereby, the quantified matrix satisfying human eye visual properties is obtained by the control of several parameters, and the subjective quality of image after coding is controlled and quantified; secondly, the method does not need to transmit or memorize the quantified matrix in codes flow, thus the coding efficiency is improved; finally, for the quantified computation of the image switching coefficients, this method can well adapt to the character of the content information in the image. Thus, this invention can improve subjective quality of the image after quantified coding.

The invention discloses a resource allocation indication method of an LTE (Long Term Evolution) system on an unauthorized frequency band, a resource allocation indication method of the LTE system on the unauthorized frequency band, a base station and a terminal. The indication method applied in the base station comprises following steps: determining the working bandwidth of the LTE system on the unauthorized frequency band; adjusting the size of a resource block group and/or the alternative value of the resource block group corresponding to the working bandwidth; determining the size of a target resource block group according to the adjusted size of the resource block group and/or the alternative value of the resource block group; informing the terminal of the size of the target resource block group and the allocation information of the resource block group through a target signaling, thus allocating one or multiple resource block groups to the terminal. Through the technical solution of the invention, the bit quantity occupied by the resource allocation domain in the DCI signaling can be effectively reduced on the unauthorized frequency band; and the signaling cost can be reduced.

The invention applies for protecting a double coordination detection method for vacant frequency spectrum detection in a cognitive radio system, relating to a radio communication system. The method is as follows: a plurality of cognitive users which take part in coordination in a traditional coordination detection algorithm are partitioned into different groups; a cognitive user needing communication is a group leader of the group; then the purposes of reducing spending of a control channel and improving detection performance of a system are gained by coordination within groups and coordination between groups. The users within the group locally judge by using an energy-check algorithm whether a master user signal exists in a certain frequency band authorized to the master user and transmits a local judgment result to the group leader. The group leader syncretizes the local judgments to make a group judgment result. A cognitive base station further syncretizes the group judgment resultto make judgment, thus getting the final judgment of the system. Bit numbers of the judgment result transmitted to the cognitive base station are largely reduced by using the method, thus greatly saving bandwidth of the control channel, and detection performance is increased.

The invention discloses a video encoding method, which is used for increasing video encoding compression ratio. The video encoding method comprises the steps of: setting size of each maximum encoding unit corresponding to a current video frame to be greater than 64*64; dividing each maximum encoding unit into encoding units according to a quadtree structure; for each encoding unit, at least acquiring a corresponding prediction residual in at least one encoding mode through a divided prediction unit; for each encoding unit, calculating a transformation coefficient matrix corresponding to a transformation unit according to at least one prediction residual through the transformation unit; and determining an optimal dividing method according to the prediction residuals and the transformation coefficient matrixes corresponding to the maximum encoding units when different dividing methods are adopted. The invention further discloses a corresponding video decoding method and a corresponding video decoding device.

The invention relates to method and device for selecting a coding mode in layered video coding. The method comprises the following steps of: acquiring in-layer prediction mode motion information selected by an enhancement layer coding block; acquiring interlayer prediction mode motion information selected by the enhancement layer coding block on the basis of the in-layer prediction mode motion information; selecting a coding mode by judging whether the motion information of an in-layer mode is consistent with the prediction motion information of an interlayer mode or not; and coding by adopting the interlayer motion information prediction mode if the motion information of the in-layer mode is consistent with the prediction motion information of the interlayer mode or else coding by adopting the in-layer mode. The invention can be utilized to reduce the amount of calculation.

The invention relates to a coding method and a decoding method of a spatial domain enhancement layer with low code rate based on an interested area. The coding method is as follows: the information of the interested area is loaded from a basic layer; original input images of an enhancement layer are cut according to the information of the interested area; and the interested area obtained after cutting is coded. The invention realizes the function of coding the spatial domain enhancement layer with low code rate based on the interested area, greatly decreases a transmitting bit number after the coding of the spatial domain enhancement layer while the quality of the video information in the spatial domain enhancement layer is ensured not to be changed, and is especially suitable for the transmitting requirement of security protection monitoring.

The invention provides a base station, a user terminal and a downlink data control method and device. The control method comprises steps of setting instruction information; and carrying out first processing related to a first identifier of UE on the instruction information, wherein the instruction information is carried by a first type control channel resource unit after being subjected to first processing and the UE is capable of judging whether itself has been scheduled according to information in the first type control channel resource unit. According to the invention, the first identifier of the UE is independently separated from DCI, so the number of RE is quite small when the first type control channel resource unit is sent in a channel, complexity of UE detection is reduced and power quantity consumed in the process where the UE is used for detecting scheduling information can be reduced.

The invention discloses a self-adaption precoding matrix indexing feedback method adaptive to joint transmission; the method is characterized in that: a user terminal determines a serving base station and the adjacent cooperation base station according to the size of the channel quality indicator from all base stations to the terminal, regenerates a series of precoding matrix codebooks with different size and ensures that the code words are uniformly distributed on the unit sphere; the maximum codebook is quantized and the maximum number bit sequence is used for expressing the counted-out precoding matrix index for the link of the serving base station; for the link of the adjacent cooperation base station, the suggested formula is adopted for working out the length which expresses the bit sequence of the index, the corresponding codebook subset is used for quantification and the shorter bit sequence is used for expressing the obtained index; and finally, the bit sequence is fed back to the corresponding base station. By adopting the method in the invention, the feedback expense can be greatly reduced on the premise of affecting the frequency spectrum efficiency hardly.

The invention provides an executable code compression method of an embedded type system. The executable code compression method comprises the following steps: step S1: counting appearing times of different codes in a binary code set; step S2: sorting appearing frequencies of all the different codes to form a new sorted code frequency table; step S3: dividing all the different appearing codes into r classes according to information in the code frequency table; step S4: compressing front (r-1) classes of codes by using dictionaries with different indexing lengths and carrying out torsion circle displacement dictionary compression on the rth class of the codes; step 5: storing r constructed dictionaries and indexing sets thereof into an external memorizer respectively. The invention further provides an uncompressing system of executable codes of the embedded type system; a central processing unit is used for obtaining the needed compressed codes from the r dictionaries and the indexing sets thereof by an address mapping logic of an uncompressing logic; instruction codes in the binary code set are obtained though an uncompressing unit in the uncompressing logic.

The invention discloses a video loop filtering method based on a deep convolutional network. The method comprises the following steps: step 1, making a training data set for loop filtering network training; step 2, constructing a network model for video filtering; step 3, taking the training data set obtained in the step 1 as a training set of the network, respectively training two models for intra-frame prediction frame filtering and inter-frame prediction frame filtering, and forming a video filtering network model by the two models; training the video filtering network model by taking a minimized loss function as an optimization target; and step 4, integrating the video filtering network model trained in the step 3 into video encoding software to complete the whole video encoding process and obtain a reconstructed frame after passing through the video filtering network. Compared with a traditional filtering method, the method has the advantages that the image quality of the video reconstruction frame is improved, the accuracy of inter-frame prediction is improved, the coding efficiency is improved, the filtered image frame has higher reconstruction quality, and the video codingefficiency is greatly improved.

The invention discloses a B-frame expansion direct mode oriented to panoramic video coding. The mode comprises three expansion direct modes, i.e. (1) mode A: using the backward motion vector of a corresponding block in a forward reference frame to carry out time-domain scale to obtain the forward and backward motion vectors of the current block; (2) mode B: using the inverse mapping of the backward motion vector of a corresponding block in a backward reference frame to generate a forward motion vector, utilizing the mapped forward motion vector to scale to obtain the forward and backward motion vectors of the current block; and (3) mode C: using the forward motion vector of the corresponding block in the backward reference frame to carry out time-domain scale to obtain the forward and backward motion vectors of the current block. The invention has the advantages of effectively compensating the defects of the prior art, reducing the prediction residual and increasing the probability of selecting the direct mode, thus greatly saving the bit numbers for coding motion vectors and improving the coding performance of panoramic video sequences.

The invention discloses an optimization method of ABR video coding code rate control. The method comprises quantification parameter optimization steps for a current coding frame as follows: obtaining coding information of a coded frame; obtaining an actual code rate of the current coding frame and a quantification parameter of the current coding frame according to the coded frame information; obtaining a code rate control error of the current coding frame; and adjusting the quantification parameter of the current coding frame that whether the code rate control error of the current coding frame is less than a set error is determined, if the code rate control error of the current coding frame is less than the set error, whether the current coding frame is a key frame is determined, if the current coding frame is the key frame, the quantification parameter is reduced in an error scope, and the bit number of the current coding frame is increased, if the current coding frame is not the key frame, the quantification parameter is increased in the error scope, and the bit number of the current coding frame is reduced, and, if the code rate control error of the current coding frame is greater than or equal to the set error, the optimization steps of the quantification parameter of the current coding frame are ended. The method can effectively reduce the code rate control error.

The invention provides an uplink time domain resource scheduling method and network equipment. The method includes the steps that an uplink time domain resource allocated to terminal equipment is determined; target time domain resource indication is determined in an uplink time domain resource index table according to a time domain scheduling mode configured for the terminal equipment and the uplink time domain resource, the target time domain resource indication is used for indicating a first start symbol of the uplink time domain resource in a time slot and the number of a first symbol corresponding to the uplink time domain resource, the first start symbol is a start symbol in a start symbol set associated with the time domain scheduling mode and a time slot acquisition mode, and the number of the first symbol is the number of symbols in a symbol quantity set associated with the time domain scheduling mode, the time slot acquisition mode and the start symbol; DCI signaling carryingthe target time domain resource indication is transmitted to the terminal equipment. By adopting the method, the combination of the start symbol and the number of the symbols can be reduced to reduceDCI signaling overhead.

The embodiments of the present invention provide a estimation method of multi-look motion. The method includes following steps: dividing the frame in the video sequence into an explicitly estimating frame and an indirectly estimating frame; calculating motion vector of the explicitly estimating frame; computing motion vector of the indirectly estimating frame according to the relative position of adjacent look vidicons, the anaglyph of adjacent looks and the motion vecto of the explicitly estimating frame. The embodiment of the invention also provides another motion estimating method, and method and device of multi-llo code based on the motion estimating method, multi-look decoding method and device. Using the invention substantially utilizes time relativity and space relativity between adjacent looks in the multi-look video under this condition that the invention ensures precision of motion estimation, reducing bit rate transport layer and improving efficiency of multi-llo code.

The invention discloses methods, devices and systems for broadcasting a precoding matrix set and feeding back a precoding matrix. The method for broadcasting a precoding matrix set comprises the step of sending the information of the precoding matrix set to a terminal through a superframe precursor, or/and overhead information, or/and user indication information of a data channel. The method for feeding back a precoding matrix comprises the steps of confirming an optimal precoding matrix in the precoding matrix set, packing the index of the optimal precoding matrix and a feedback signal and feeding back a signal packet, wherein the packet head value of the signal packet is used for showing the combination of the feedback signal and the bit number occupied by the index. The device for broadcasting a precoding matrix set and the device for feeding back a precoding matrix comprise modules for realizing the corresponding functions in the methods; and the systems for broadcasting a precoding matrix set and feeding back a precoding matrix comprise corresponding base station equipment and terminal equipment. The performance of the systems can be improved and the single overhead can be saved due to the embodiment of the invention.

The invention discloses a method for deep convolutional neural network model compression. The method comprises the steps that a trained deep convolutional neural network model is retrained to remove redundant network connections; weights of remaining connections of various network layers of a convolutional neural network are coded; the weights of the remaining connections of the various network layers of the convolutional neural network are subjected to k-means clustering; clustering results are subjected to fine tuning; and results after fine tuning are saved, and a saved file is subjected toHuffman coding. According to the method, by setting a dynamic threshold, the connections in the network can be gently removed to enable the network to be recovered from the unfavorable condition thatthe connections are removed, and therefore the effect that the compression multiples is high under the condition of the same accuracy rate loss can be achieved; and in the coding process of the remaining connections, the bit number needed for representing an index value can be decreased by means of the used improved CSR coding method, therefore, the size of the compressed file can be decreased, and the compression ratio is increased.

The invention discloses a video encoding and decoding method using space zoom prediction. The encoding method comprises the following steps: by using zoom information, performing motion estimation on zoomed reference frames obtained by zooming original reference frames at different zoom ratios to find optimal matching; by using zoom prediction, estimating the zoom step and zoom window of the zoomed reference frames through preset number of zoom layers, the smallest zoom step and a predicted zoom value; and by using reference frame index prediction, predicting the numbering mode of the original reference frames and the zoomed reference frames, thus saving the bit number required for encoding the reference frame index. When being used for decoding each macro block, the decoding method comprises the following steps: performing reference frame index prediction, and finding the position of the reference frame in a reference frame matrix; performing zoom prediction, and finding the reference frame zoom ratio; generating a prediction block according to the zoom ratio and the MV (Mean Value); and finally performing steps of motion compensation and the like.