91results about How to "Reduce storage size" patented technology

The invention discloses a system and method for repairing a bad display area of a liquid crystal display panel. The system for repairing the bad display area of the liquid crystal display panel comprises an image acquisition device, a distortion-free compression module, a storage and a color spot repair module, wherein the image acquisition device is used for acquiring data of images, containing the bad display area, of the liquid crystal display panel, the distortion-free compression module is used for carrying out difference operation on the collected data of the images to achieve distortion-free compression processing, the storage is used for storing the compressed data processed through distortion-free compression processing, and the color spot repair module is used for carrying out decompression processing on the compressed data stored in the storage, carrying out color spot repair on the decompressed original data of the images to generate feedback information, and sending the feedback information back to the liquid crystal display panel. According to the system and method for repairing the bad display area of the liquid crystal display panel, due to the facts that values of the images which will be stored in the storage, and contain data of the bad display area are processed through distortion-free compression and decompression, then, color spot repair is carried out, and the information is output to the liquid crystal display panel, the storage size of the storage can be reduced, the manufacturing cost is reduced, and the De-Mura processing effect cannot be reduced.

The present invention is aimed at three specific data areas of font compression, each of whose size has become significant as other data areas have been compressed. The three data areas include model factoring, character level feature measurement (local dimensions) factoring, and typeface level feature measurement (global dimensions) factoring. In general, the invention in each area is an apparatus and method used in font compression to reduce redundant information, thereby allowing a reduction in data format (e.g., words to bytes and bytes to bits) resulting in an overall reduction in storage area for a given font collection.

The present invention provides methods and systems to protect an organization's secure information from unauthorized disclosure. The present system uses protect agents installed across various egress points (e.g., email server, user's computer, etc.) to monitor information disclosed by a user. The present system also provides the use of lightweight fingerprint databases (LFD) to maintain a database of fingerprints associated with the organization's secure data. In one embodiment, the LFD is stored locally at the site of each protect agent such that the organization's secure information can be protected even when a protect agent is disconnected from the network. Methods and systems to compress fingerprints to achieve the lightweight fingerprint databases are also provided. In one embodiment, a combined approach, utilizing both the local lightweight fingerprint database and a remote fingerprint server comprising registered fingerprints is used to achieve overall protection of the organization's secure information.

The invention discloses a human body medical model three-dimensional visualization system used for a mobile terminal and a realizing method thereof and belongs to the technical field of information processing. The human body medical model three-dimensional visualization system used for the mobile terminal comprises a modeling module, a data compaction module, a memory management module and a three-dimensional display module, wherein the modeling module and the data compaction module are realized on a server-side, and the memory management module and the three-dimensional display module are realized on the mobile terminal. The modeling module is used for modeling and generating a three-dimensional human body medical model file according to a medical section image. The three-dimensional human body medical model file is transmitted to the three-dimensional display module after redundant information is removed through the data compaction module. The three-dimensional display module is used for achieving imaging operation of the three-dimensional model according to touch operation of a user. The memory management module is connected with a memory management device of the mobile terminal and is used for real-time tracking of internal memory of the mobile terminal. According to the human body medical model three-dimensional visualization system used for the mobile terminal and the realizing method thereof, three-dimensional visualization control of the medical model is enabled to be smooth and flexible on a mobile terminal platform for the user, and important technical support is provided for the development of software relevant to three-dimensional visualization of the three-dimensional medical model, wherein the software is applied to mobile platforms.

The invention provides a railway locomotive operation control system hybrid task scheduling method based on reinforcement learning, and the method is an offline learning process. According to the method, first hybrid task set data during actual operation or in a simulation experiment of a locomotive operation control system, a hybrid task set is formed and regularization marking is performed on each task in the hybrid task set. Then the task set after regularization marking serves as input of a reinforcement learning system, and a reinforcement learning environment is formed. The reinforcement learning system applies a reinforcement learning algorithm, inspects scheduling objectives of the locomotive operation control system to perform an iterative learning process, a <state-rule> corresponding relation table corresponding to the hybrid task set is generated; and the <state-rule> corresponding relation table is stored in a database. The rule whose frequency of occurrence is the highest is selected from the database as the optimal rule of a current state, and a final <state-rule> corresponding relation table is formed. During operation of the locomotive control system, generation of a real-time scheduling sequence of hybrid tasks is guided according to the <state-rule> corresponding relation table, thereby realizing task scheduling.

The invention discloses a detection and estimation method for fine synchronization of a downlink primary synchronization signal in an NB-IoT system. The method comprises the following steps: step (1),acquiring a time domain sampling sequence by an air interface receiver by adopting a sampling rate of 1.92MHz; step (2), performing NPSS coarse synchronization estimation by a terminal receiver; andstep (3), performing second fine synchronization at the sampling rate of 1.92Msps according to time offset and frequency offset results estimated in the course synchronization in step (2), and carrying out self-correlation operation by using an NPSS sequence and a local sequence, wherein the local sequence is generated by using integer frequency offset, and accurate time offset and the integer frequency offset are determined according to a peak position maximum value. The method of the invention has the following beneficial effects: the fine synchronization is that a correlation operation is carried out after an accumulation operation is carried out, so that the storage capacity is reduced; the fine synchronization adopts a pipeline structure based on two adders and one multiplier, so thatthe calculation amount and the storage size are reduced; and the fine synchronization frequency offset estimation stores a group of local sequences and performs iterative updating to replace multiplegroups of local sequences, so that the storage capacity is reduced.

A storage device having detachable multiple-in-one connector has a casing with one opening; a multiple-in-one plug part mounted inside the casing, exposed beyond the opening and having first and second plug assemblies detachably stacked with a gap formed therebetween; a movable assembly mounted inside the casing, connected with the second plug assembly and penetrating through the casing; and a circuit board mounted inside the casing and electrically connected with the first and the second plug assemblies, having an automatic interface switching procedure and a plurality of interface connecting procedures, having a control circuit determining the plug assembly currently in use, and selecting the interface connecting procedure corresponding to the plug assembly currently in use. Due to the multiple-in-one design, the size of the storage device can be reduced, and the storage device can be selectively plugged in a multiple-in-one or single socket connector.

Efficient buffer cache utilization frees a data buffer as soon as data buffer processing is completed, and without losing association of the freed data buffer and a descriptor buffer. Separate free buffer link lists identify the freed data buffer and any freed descriptor buffer. The data buffer is rapidly processed then freed generally before completion of processing of the descriptor buffer, freeing the processed associated data buffer before the associated descriptor buffer is freed. The association of the processed free data buffer and the descriptor buffer may be ended to enable the more frequent use of the large capacity data buffer for other update requests.

The related compression storage method for great much data comprises: detecting CPU type to select matched storage strategy; saving one piece of record with integral information in website as the reference for other records; noting the reference record difference by increment means; removing the last-version data in history to realize the dynamic conversion for history and active data; according to frequency of data value, recoding data to calculate without decoding. This invention can improve query efficiency with less storage space.

The embodiment of the invention provides an image processing method and device, electronic equipment and a storage medium and relates to the field of image processing technology. The method comprisesthe steps that S10, an image training set is acquired, and visual features of all training images in the image training set are extracted; S20, clustering is performed on the visual features to generate visual dictionaries with clustering centers being visual words, and 1 is added to the quantity of the visual dictionaries; S30, whether the quantity of the visual dictionaries is equal to a predetermined quantity is judged, if yes, the generated visual dictionaries in the predetermined quantity are output, and otherwise the step S40 is executed; S40, the visual words, closest to the visual features, in the visual dictionaries are determined; and S50, residual errors of the visual features and the closest visual words are calculated, the residual errors are used as new visual features, and the step S20 is returned to. Through the technical scheme in the embodiment, the storage scale of the visual dictionaries can be remarkably reduced, and it is convenient to perform deployment on a mobile terminal.

The invention discloses an evaluation method of an electric energy meter manufacturer based on a metering big data clustering model, which comprises the following steps: acquiring the original data ofthe electric energy meter and screening the evaluation index of the quality of the electric energy meter; Gaussian mixture model clustering algorithm is used to extract the typical evaluation index data and obtain the clustering center to form the decision matrix of comprehensive quality evaluation of watt-hour meter, which is standardized according to the benefit type and cost type respectively.Based on the weight of AHP, entropy method, Pearson correlation coefficient method and variation coefficient method, the combination weights are obtained by the combination optimization model of index weights, and the decision matrices are weighted to obtain the comprehensive evaluation matrices of the quality of electric energy meters. The quality of watt-hour meter is evaluated comprehensivelyby the method of clip angle measurement, and the quality of each watt-hour meter supplier is sorted from high to low, and the evaluation results are obtained. It can provide support for accurate accounting of energy consumption and energy revenue.

To provide a decoding device that compactly stores prefix related information therein compatible to variable-length codes used in various systems. The decoding device includes a register file in which a prefix common portion and a word length of the prefix common portion, prefix individual portions and word lengths of the prefix individual portions, word lengths of suffixes, and an input bit sequence are stored, a bit matching unit that determines a prefix included in the input bit sequence from prefixes generated from the prefix common portion and the prefix individual portions, a codeword computation unit that computes the word length of a codeword included in the input bit sequence using the respective word lengths of the prefix common portion, a prefix individual portion, and a suffix corresponding to the determined prefix, and an index computation unit that computes an index in a symbol table using the respective word lengths of the corresponding suffix and the codeword.

The invention discloses a demographic statistics device and method based on boundary selection, and belongs to the field of deep learning and image processing. The invention comprises the following steps of by improving the YOLO neural network by increasing the division of the YOLO unit from 7*7 to 9*9, that is, increasing the number of detections per unit to three, and then replacing the YOLO PCneural network with the Fire module in SqueezeNet; replacing The 16th, 18th, and 24th 3*3 convolutional layers with Dire modules in SqueezeNet and decreasing the number of compressed portion of the Fire module from 128 to 96; retraining network to get a new S-YOLO-PC neural network, using the S-YOLO-PC neural network for boundary selection of human flow statistics, using the new neural network, making it more accurate in the case of greatly reduced models, can be used in a variety of occasions Human flow detection.

The invention relates to a pruning method for a deep convolutional neural network, computer equipment and an application method, and the method comprises the steps: (1), taking a layer as a unit, andexecuting a convolution kernel spectral clustering algorithm; (2) carrying out intra-layer convolution kernel pruning, dynamically calculating a spectral clustering pruning rate of this round, cuttingoff convolution kernels of non-clustering centers, and finely adjusting a certain number of rounds to recover precision; (3) calculating the importance of the convolution kernel by taking the globalsituation as a unit; (4) carrying out global convolution kernel pruning, calculating the global pruning rate of the current round dynamically, cutting off convolution kernels with low importance, andadjusting a certain round number finely to recover precision; (5) circularly executing the steps (1) to (4) until the target pruning rate is reached; and (6) keeping the weight of the descending convolution kernel as initialization, and continuing to finely adjust a certain number of rounds to obtain a final pruning network. According to the method, the problem that a large number of redundant parameters exist in an existing deep convolutional neural network is solved, and compression and acceleration of the network are achieved on a universal hardware platform.

The invention discloses a network address layered structure design and maintenance method based on geographical administrative divisions, belonging to computer application technology field. The invention is to solve problem for effectively dividing network address of whole Internet under IPv6 network address condition and designs corresponding network router strategy based thereon. The principle is that network is managed and divided according to geographical administrative divisions layered structure and IP network address of user terminal can be divided into several segments; the routing mode is divided into three mechanisms of uplink, middlelink and downlink according to layering and data packet transmission direction. The layered structure of the invention enable the network data transmit in regularity, reduces unnecessary routing spending and promotes network efficiency, implements network positioning to a certain degree, strengthens network safety, reduces memory size of routing table, saves routing address time, and effectively avoids jam of middle and lower routing system.

The invention provides an optimization method and device based on a file embedded font; the method comprises the following steps: obtaining font information of the file embedded font, wherein the font information comprises font names and corresponding font files; obtaining font file font pattern and attribute information thereof; combining font files with same font name according to the font pattern and attribute information thereof. The method collects fonts of same name, and compares font patterns so as to complete repeat font pattern delete and font combination, thus eliminating repeated font patterns in the font, greatly reducing font data and memory size, and improving file display and loading speed.

A downsized substrate may be housed in a substrate accommodation vessel (FOUP) constituting a transfer system corresponding to a large diameter substrate. An attachment includes an upper plate and a lower plate supported by a first support groove that can support an 8-inch wafer, and holding columns installed at the upper plate and the lower plate and including a second support groove that can support a 2-inch wafer (if necessary, via a wafer holder and a holder member). Accordingly, the 2-inch wafer can be housed in a pod corresponding to the 8-inch wafer, and the pod, which is a transfer system, can be standardized to reduce cost of a semiconductor manufacturing apparatus. In addition, a distance from each gas supply nozzle to the wafer can be increased to sufficiently mix reactive gases before arrival at the wafer and improve film-forming precision to the wafer.

A device, system, and method for retrieving a sample of oil from an engine, comprising, a chamber having an inside wall and at least one oil inlet, a piston head received in said chamber and slidable between an empty position in which it is disposed substantially at said inlet and a filled position in which it is a first distance from said inlet, a handle affixed to said chamber, and a piston member extending between said handle and said piston head, said piston member comprising a flexible member having a proximal end and a distal end, said distal end being affixed to said piston head.

The invention provides a data compression method and device and a data decompression method and device, and the method comprises the steps: dividing original data into a plurality of to-be-compresseddata blocks, and enabling each to-be-compressed data block to be stored according to a plurality of preset bytes; according to data information of non-zero bytes in each to-be-compressed data block, determining first compressed data corresponding to the to-be-compressed data block; and splicing the first compressed data corresponding to the plurality of to-be-compressed data blocks to obtain second compressed data corresponding to the original data, thereby reducing the storage size of the data, saving the storage space of the data, and improving the storage efficiency and transmission efficiency of the data.

The invention relates to a self-adaptive parameter-adjusting electric arc additive manufacturing method and system. The method comprises the steps: obtaining a three-dimensional model; performing hierarchical slicing processing to obtain hierarchical paths, forming a preliminary path point set, judging whether to execute hierarchical path adjustment, and if so, performing path point simplification or subdivision based on the preliminary path point set; conducting welding path planning to obtain an adjusted path result; setting welding parameters according to the path result, and selecting a welding process; conducting simulation verification on the welding process, and judging whether the path result continues to be adjusted or not; generating a corresponding welding instruction according to the verified path result, adding a welding platform preheating instruction before the welding instruction, obtaining an execution instruction set and sending to a welding machine and a robot; and responding to the received execution instruction set, and implementing additive manufacturing. The efficiency and precision can be effectively improved, and the difficulty of electric arc additive manufacturing is greatly reduced.

The invention discloses a multivariable public key signature system and a multivariable public key signature method. The signature system comprises a first processor, a first linear affine transformation part, a trap door part, and a second linear affine transformation part. The first processor is used to generate a random number, and is used to receive to-be-signed massage, and is used to transmit the to-be-signed massage to the first linear affine transformation part. The first linear affine transformation part is used for the affine transformation of the to-be-signed massage, and the trap door part is used to generate a multivariable polynomial equation group according to a private key parameter, and is used to substitute the received random number and the affine transformation result into the multivariable polynomial equation group for the solution, and the second linear affine transformation part is used for the affine transformation of the solution acquired by the trap door part, and a signature is acquired. Under a condition of guaranteeing the unforgeability of the signature, the speed of the multivariable public key password signature is accelerated, and at the same time, the memory size of the private key is reduced.

The invention discloses an image processing method and device, equipment and a storage medium. The method comprises the steps that acquring a current frame image collected by a camera, and extractingvisual features of the current frame image; generating a feature vector of the current frame image according to the visual feature of the current frame image; dividing the feature vector of the current frame image into a plurality of sub-vectors, quantifying the plurality of sub-vectors, and generating a feature index of the visual feature of the current frame image; matching the feature index ofthe visual feature of the current frame of image with the feature index of the visual feature of each training image, and determining a matching feature pair of the current frame of image and each training image; wherein the feature index of the visual feature of each training image is obtained based on a sub-code book; and determining the training images of which the number of the matched featurepairs is greater than a first preset threshold value as similar images of the current frame of image. According to the technical scheme, rapid image recognition can be realized.

The invention relates to a data compression method and device for an application program and a computer storage medium. The data compression method of the application program comprises the steps of traversing a plurality of folders included in a code branch, and obtaining a first file meeting a data compression requirement; Compressing the first file to obtain a corresponding second file; When thecompression ratio of the second file is higher than a first threshold value, replacing the first file with the second file, Wherein the code branch is a related file of the application program, and the plurality of folders are folders needing to be packaged into the application program. According to the data compression method of the application program, the first file is selectively compressed by presetting the first threshold value, so that useless compression is reduced, the compression efficiency of the application program is improved, and meanwhile, the user experience is improved.

The invention relates to a panoramic vision monitoring system and method suitable for a heavy commercial trailer. The system comprises a central control host, a driving data recorder, a display module, and an image acquisition module; the central control host is electrically connected with each of the driving data recorder and the display module through a CAN bus, the image acquisition module is electrically connected with the central control host, and the driving data recorder is wirelessly connected with a background monitoring terminal. Through the panoramic vision monitoring system provided by the invention, the central control host directly acquires time information and vehicle location information from the driving data recorder and acquires vehicle model information and vehicle driving information from a trip computer, six paths video image information around the vehicle are acquired through the image acquisition module, and the image processing is performed on the six paths of video image information through the central control host, and meanwhile, the above information is overlapped to generate synthesized image information; the synthesized image information is displayed through the display module to facilitate the driving of a driver; and the six paths of video image information are sent to the background monitoring terminal through the driving data recorder to facilitate the remote monitoring.

The present invention provides a shielding disc and a reaction chamber. The shielding disc comprises a rotation subsection and a body. The rotation subsection is composed of a fixed end and a movable end. The fixed end is arranged along the circumferential direction of the body. The movable end of the rotation subsection is located at an original position under the action of the gravity force thereof. The original position is defined to be the position of the movable end when an included angle between the lower surface of the rotation subsection and the lower surface of the body is preset to be an obtuse angle. When a to-be-shielded object is shielded by the shielding disc, the movable end of the rotation subsection is jacked up from the original position to a preset highest position. In this way, the shielding disc capable of shielding the to-be-shielded object is formed by the rotation subsection and the body. The shielding disc is smaller in size when to be stored. Therefore, no garage for storing the shielding disc is required to be arranged in a reaction chamber, so that the cost of the reaction chamber is reduced. The economic benefit is further improved.

The invention discloses a storage optimization method for a Hadoop distributed file system, and the method specifically comprises the following steps: firstly, selecting an INFO-level log file which comprises a specific execution timestamp and file name information, and obtaining an access record and a deletion record of the INFO-level log file; extracting and sorting all information containing keywords in the IFNO level log, and then sorting and numbering according to timestamps; then determining feature tags, selecting features, constructing feature vectors, and forming a sample set of a training file elimination model; selecting three feature values of the feature vector as three classification nodes of a decision tree in sequence, establishing the decision tree by adopting an ID3 algorithm, and constructing a file elimination model by the decision tree; and finally, predicting the reusability of the file by using the established file elimination model. According to the method, the storage efficiency of the distributed file system is optimized, the data storage scale is reduced, and the storage efficiency of the HDFS is improved.

A technique for displaying 3D videos creates a representation of image deformation, such as depth maps, in terms of a function for overlaying kernels with variable support and order. By optimizing the kernel type, order and support, which are allowed to be varied across different region of the deformation, it is able to approximate the image deformations in terms of the kernel parameters. Since the number of kernel parameters is usually much smaller than that of the number of pixels, this allows a significant reduction in the storage size of the image deformation.

The invention relates to an estimation method for a defect quantification parameter. The estimation method mainly aims at an inner defect of a material or a structural body. A simple method for estimating the defect parameter via an impulse ultrasonic echo is provided. The method is mainly characterized in that: the data collection amount is quite small; and the ultrasonic echo model is simple and convenient, and can be determined by a triangular model. A little of collected echo signal data is used for estimation to obtain the wave arrival time, the maximum echo peak, the echo climbing speed, the echo lowering speed, the echo width, and the echo bandwidth, which are relative to the defect quantification parameter. The estimation method can be used in the situations, such as defect detection and parameter quantification, of the field of nondestructive examination and estimation.

The invention discloses a circuit board health condition monitoring method based on volt-ampere characteristic curves. The method includes: data completeness can be guaranteed while a representing method of a digital image is applied to a volt-ampere characteristic curve between any two points on a circuit board for storage, and node grouping curves serve as characteristic information of the circuit board in a manner of a circuit gray graph. Voltage and current data in the volt-ampere characteristic curves are subjected to per unit processing, all curves formed by element pins are classified according to nodes, the curves are respectively subjected to aliasing according to the nodes, placement in each pixel region is summed up to be represented as gray, a set of all pixels forms the digital image of characterization node inherent characteristic, and correlation of the circuit gray graphs of the nodes under different statuses is determined to judge health condition of the circuit board. By the circuit board health condition monitoring method, conversion and compression of mass data are achieved, storage scale of the data is downsized greatly, and speed, efficiency and accuracy of monitoring the circuit board health condition are increased.