134results about How to "Complete" patented technology

The invention provides a face recognition method capable of extracting multi-level image semantics based on the CNN (convolutional neural network). The method comprises the following steps: implementing further optimization on the basis of VGGNet, providing a new network structure, and implementing cross-level splicing on multi-level features to ensure that the finally-extracted image features have multi-level image semantics; and meanwhile, adding the extracted traditional features in the training of the CNN as additional features to ensure that the CNN feature information is more complete; then optimizing the structure of a shallow convolutional layer to ensure that the redundancy calculation is reduced and the calculation amount of models is greatly reduced; and finally, accelerating the convolutional layer by using an improved matrix decomposition algorithm to ensure that the network can be accelerated within 1 second and the accuracy rate of the models cannot be reduced when the speed-up ratio reaches four times. A face matching algorithm achieved by the invention has the characteristics of high accuracy and high real-time performance, and has higher accuracy and higher calculation efficiency compared with existing algorithms.

The invention provides a method for diagnosing a fault by a functional fault directed graph, which can be applied to the field of fault diagnosis for some systems of aircrafts (such as an air source system, an anti-icing system and an air-conditioning system of an airplane gas path). The method comprises the steps of: decomposing a diagnosis object system to determine component relation and build a system structural model, confirming the corresponding input and output based on the component functions, determining the component fault model, analyzing fault propagation path and influence relation, setting test points and test information of a sensor, monitoring parameter variation that indicates the potential fault, forming an FF-SDG model, and at last, searching the potential fault source by a hierarchical policy inference method based on the FF-SDG model to perform fault diagnosis. The method can quickly and effectively find out the system fault source of the aircraft system, discover the fault cause, and make up shortage of the fault diagnosis technique based on manuals and cases when some fault cannot be predicated and the knowledge is difficult to be acquired by domain experts.

The invention relates to a caption rendering technology in a field of producing television programs, in particular to a caption rendering method based on a curve-profiled closed loop domain and a pixel mask code matrix. The method is that: after a Bezier curve vector profile of a caption object is discretized into a polygon vector, according to the direction and embordering type of a polygon, the polygon vector used for embordering is further created, thus systematically combining the computer graphics with the digital image processing, converting the vector information of the caption object into the pixel mask code matrix based on a contour line and applying the pixel mask code matrix throughout all links of the caption rendering; therefore, the effect of the caption rendering is improved, the ever-increasing demands of the caption quality and the caption effect from the television programs producing and broadcasting organizations are sufficed, the important social property of serving the mass people by television media is reflected even more directly, and better social benefits are obtained.

The invention relates to a fast and complete automatic driving trajectory planning method. The method comprises the steps that 1) a basic path is planned; 2) collision detection is carried out on thebasic driving path, and step 3) is carried out; if collision does not exist, whether the basic driving path reaches a target point is determined; if so, step 4) is carried out; if not, step 3) is carried out; 3) after a sampling / search path planning method is used to acquire a smooth effective path, step 4) is carried out; 4) after a curvature extreme value corresponding to the path is acquired, segmentation velocity is planned to acquire a trajectory; and 5) collision detection is carried out on the trajectory in time domain and spatial domain; if collision exists, whether the distance between a vehicle and an obstacle is greater than 2 times the minimum braking distance; if so, velocity re-planning and returning to step 4) are carried out; if not, returning to step 1) is carried out; finally the path and velocity are combined to generate a trajectory and output the trajectory. Compared with the prior art, the method provided by the invention has the advantages of trajectory planninginstantaneity improvement, stability, reliability and the like.

The invention relates to a lip reading technology based lip language input method, mainly aims at Chinese characters in common use and Arabic numerals, belongs to the intelligent computer identification technology, is a typical image mode analysis, understanding and classified calculation problem and relates to multiple subjects of mode identification, computer vision, intelligent human-computer interaction, cognitive science and the like. Shot mouth lip movement video is subjected to key frame extraction, normalizing processing is performed for extracted images by utilizing gray processing, median filtering, dynamic threshold binarization processing and scanning for noisy point removal, then eigenvectors are extracted to obtain parameters with lip characteristics and are matched with a lip model library to identify the images as a Chinese phonetic alphabet sequence, and finally an input method module is combined to obtain corresponding Chinese characters or Arabic numerals.

The invention discloses a certificateless multi-receiver signcryption method with a multivariate-based cryptosystem. The certificateless multi-receiver signcryption method is used for solving the technical problem of poor security of an existing certificateless signcryption method. The technical scheme includes that the method includes: firstly, using a key generation center to generate system parameters, and selecting a secure multivariable encryption algorithm to generate keys; and enabling a user U to obtain system private keys of the key generation center and generate the private keys to user keys prior to that a user Alice with the identity IDA sends signcryption messages to a user group L={ID1, ID2,...,IDt}, and accepting or refusing ciphertext sigma by means of un-signcryption. By the aid of the multivariate-based cryptosystem, high security of anti-quantum computation is achieved, and the certificateless multi-receiver signcryption method is still high in security facing attack of a quantum computer. Compared with the background art, the certificateless multi-receiver signcryption method is small in computation and high in security, has the advantages of completeness, confidentiality, unforgeability, forward security, backward security, receiver privacy protection and the like and is capable of resisting existing known attack.

The present invention discloses an old leaf black tea manufacturing method which relates to the field of tea processing technology. The manufacturing method is composed of the following steps: collecting and picking fresh leaves, withering, rolling, fermenting, drying, etc. The old leaf black tea manufactured by the traditional process has thick and loose twisted tea sticks, a mellow aroma and a light taste. The old leaf black tea manufactured by the present method not only conducts technical improvements in each process step to obtain complete tea appearance, but also spraying polyphenol extracts and mixed water solution of cellulase and honey in the rolling step and can effectively ensure the taste and aroma of old leaf black tea.

The invention discloses a linear pseudo-spectrum GNEM guidance and control method. According to the linear pseudo-spectrum GNEM guidance and control method, an original nonlinear optimal control problem is converted into a group of problems for solving a system of linear algebraic equations based on the concept of GNEM through linearization and Gauss pseudo-spectrum dispersion by combining nonlinear approximation model predictive control, linear quadratic optimal control and a Gauss pseudo-spectrum method. The linear pseudo-spectrum GNEM guidance and control method has the advantages that the calculation efficiency of solving the optimal control problem is quite high, and high calculation precision can be obtained just through several nodes; in addition, a final solution can be presented through a smooth function relevant to control over disperse nodes, and the method is quite suitable for online calculation. It is indicated through a simulation result that by applying the method to tail-section attack guidance with terminal angle constraint, compared with an MPSP method, the linear pseudo-spectrum GNEM guidance and control method has higher calculation efficiency and high calculation precision and can be completely applicable to a guidance framework for terminal guidance, and a smaller required overload command is generated compared with self-adaptive tail-section proportion guidance.

The invention an electric-energy quality data compression and reconstruction method based on self-adaptive dictionary learning, and self-adaptive dictionary learning is used for sparse representation of electric-energy quality data. A lot of different types of electric energy quality data form a training sample set, and the completeness and redundancy of the training sample set are ensured; basic atom sparse codes which can represent the electric energy quality data is extracted from the training sample set adaptively, and a self-adaptive dictionary is obtained via iteration; a random Gauss matrix is used to carry out dimension-reducing measurement on electric energy quality test signals, and compression sampling is realized; and on the basis of compressed perception theories, the self-adaptive dictionary is used for sparse solution to obtain a sparse representation matrix of test signals, and the original signals are decoded and reconstructed. Thus, the electric energy quality data is compressed and sampled simply and reconstructed accurately, the sampling efficiency of the electric energy quality data is improved, and storage of redundant data is reduced.

The invention relates to a microwave processing method for chrysanthemum tea. The method comprises the following steps of: firstly, picking out satisfactory fresh chrysanthemums, and spreading the chrysanthemums at room temperature to air; performing microwave green removing, quickly reducing the temperature after green removing, and repeatedly kneading the chrysanthemums; and drying, screening, inspecting and packing the chrysanthemums to obtain the finished product of chrysanthemum tea. Compared with the traditional common green removing method through heating, distilling and drying, the method can effectively reserve important drugs and nutrition constituents such as volatile oil, flavone, chlorogenic acid, stevia, choline, amino acids, vitamins, abundant mineral elements and the like contained in the chrysanthemums, and the disinfection effect is improved.

Disclosed is a method for making cleaned flat type green tea consists of grading and gathering the tea fresh leaves, cleaning, water-removing, microwave sterilization process, and controlling the microwave water-removing to make the moisture content in the tea to be 20%-30%, and the moisture content in the end product tea to be 5%-8%.

The invention relates to a bus fire early warning and armored window glass bursting remote control system, comprising an intelligent gateway, a ZigBee network, a 3G network and a remote bus monitoring center. The intelligent gateway comprises a Leonardo processor with an ARM+DSP double-core architecture, a video collecting and decoding module and a display alarming module; the ZigBee network is composed of a coordinator node and a plurality of monitoring nodes. Each monitoring node monitors fire information such as temperature, smoke and gas, gives an alarm and controls the bursting of the armored glass in the case of emergency; the 3G network is used for transmitting video information, monitoring data and commands of the bus monitoring center which have been encoded and compressed. The ZigBee network combines real-time data monitoring with 3G network remote video transmission, thus achieving fire pre-warning and multistage alarming functions; in the case of emergency, the monitoring center monitors fire situation through video and remotely starts the bursting of the armored window glass, thus allowing passengers to escape as soon as possible, and protecting life safety.

The invention discloses a multimedia file cloud storage platform and a method for eliminating redundancy by using the cloud storage platform, and belongs to the field of information processing. The multimedia file cloud storage platform comprises a storage interface management subsystem, a cluster management subsystem and a storage management subsystem, wherein the storage interface management subsystem is used for generating an operation instruction and used for communicating with the cluster management subsystem or the storage management subsystem. The method for eliminating redundancy by using the cloud storage platform comprises the following steps: firstly, calculating fingerprint information of an uploaded file, generating a verification instruction, transmitting the verification instruction to the cluster management subsystem to be judged, if the fingerprint information is available, checking a corresponding file metadata instruction, and feeding back a result; or else verifying the fingerprint information of the file, and transmitting a result; continuously judging whether the cloud storage platform has identical files or not, if the cloud storage platform has identical files, transmitting a redundancy record adding instruction, and uploading a result; and if the cloud storage platform does not have identical files, storing the files, adding instructions, and uploading a result. By adopting the method, the storage load of a storage server can be reduced, and file transmission between a user and the cloud storage platform can be accelerated.

The present invention provides a systematic operable method for generating safety data base management system test application example. Said method includes the following several steps: (1) generating test protocol; (2) generating test template; (3) type partition; and (4) generating test vector. Said invention also provides the concrete requirements of every step.

The invention discloses a high-real-time quantitative ultrasonic detection method based on multi-angle stereo matching. Aiming at a serious restriction bottleneck between the fine quantitative representation capacity and detection instantaneity of defects of existing ultrasonic detection devices, a basic idea of a finite element is firstly introduced to rasterize a to-be-detected test piece, a plurality of ultrasonic transducers distributed in different directions of a space are utilized for sequentially irradiating grid cells, and a shifting relation between the fine quantitative representation capacity of the flexible coordinating defects and the detection instantaneity is realized by utilizing the wavelet-transform-based multi-resolution stereo matching correlation based on the acquisition of multichannel ultrasonic reflection echo signals. By collecting the multi-angle ultrasonic reflection echo signals, the high-real-time quantitative ultrasonic detection method has the advantage of complete carried information amount and is more beneficial to realization of the quantitative detection; by utilizing a basic theory of wavelet transform, the method has multi-resolution and time-frequency domain conjoint analysis capacities and is particularly suitable for time frequency signals such as ultrasonic waves and the like.

The invention discloses a CEEMDAN-ICA-SWT-based diesel engine vibration source multi-information joint identification method, comprising the steps of obtaining a vibration excitation source mixed response signal by utilizing a vibration sensor; determining an amplitude standard deviation coefficient and an overall average number of times of adding the white noise, then performing adaptive decomposition on the vibration signal by adopting a CEEMDAN method, selecting a main intrinsic mode component based on a quadratic correlation similarity coefficient method and a spectral entropy combined method, and estimating a vibration source number; forming a new observation signal as a plurality of virtual channels in independent component analysis by using the obtained eigenmode component and excitation source mixed response signal, performing blind source separation by using an ICA method, and selecting a separation result by using a similarity coefficient method in combination with the estimated number of main vibration sources; and carrying out time-frequency characteristic analysis on each independent component of the separation result by adopting an SWT technology, and identifying thevibration excitation source of the diesel engine. According to the CEEMDAN-ICA-SWT-based diesel engine vibration source multi-information joint identification method, the identification capability ofthe vibration source of the diesel engine is improved.

A test method for the performance of communication equipments, includes following steps: at the same time for aging test, makes building internal self-testing BIST, sets a control position in system to get the aging information. CPU tests whether or not the board is at the status of aging testing through the control position. According to the aging testing status controls the board to build internal self- testing BIST. Building internal self-testing BIST adopts loop testing, the result stores in relevant storage space; the saving of the results adopts double data backups marked with relevant sign position, the double data backup includes front two round results of loop testing, the sign position shows the relevant testing status at the time of getting this test result, takes effective testing result in a newest round as the criterion.

The invention belongs to the field of processing of tea leaves, and relates to a processing method of Laoshan green tea, and particularly relates to a processing method of clean and fresh type high-grade special Laoshan green tea capable of erecting after being brewed. The processing method comprises the following steps of picking, spreading, deactivation of enzymes, strip tidying, spreading for cooling and moistening regain, primary baking, secondary spreading for cooling and moistening regain, and re-baking to be enough dry, wherein a fresh leaf which is subjected to the deactivation of enzymes is directly subjected to the treatment of the strip tidying without being subjected to the spreading for cooling and temperature lowering; after the primary baking, the leaf is subjected to the secondary spreading for cooling and moistening regain; the re-baking to be enough dry is carried out by selecting and using a temperature which is lower than that of the primary baking. The Laoshan green tea processed through the processing method of the Laoshan green tea is tall and graceful in appearance, green and sleek in color and luster, apricot-green and transparent in liquor color, clean and fresh in fragrance, fresh and mellow in taste, and bright, complete and lobate in brewed leaf; after being brewed, the tea leaf erectly spreads in a cup and is high in viewing and admiring performance; the contents of an amino acid and a vitamin C which are beneficial to a human body are higher than those of ordinary Laoshan green tea.

The metering device includes electric energy distribution bus, management and control unit of measuring electric energy at input end, and management and control unit of measuring electric energy at user end. Control unit of measuring electric energy at user end for 1-30 grouped users, central management and control unit of providing functions of display, control, communication and management for users in group are assembled inside a case. The said bus connects the control unit of measuring electric energy at input end, control unit of measuring electric energy at user end for 1-30 grouped users, and central management and control unit.

The present invention discloses a design method aiming at binary program automatic parallelization of a multi-core platform. The method comprises: translating the binary in the multi-core platform into LLVM IR, and then establishing a control flow diagram to find all the cycles in a program by firstly analyzing the LLVM IR; carrying out static analysis on the parallelism of cycles, and removing the cycles difficult in parallel; before regenerating a paralleled code for the cycles, setting simple criteria for the common memory access dependency, adding a corresponding memory address overlapping judging code, thereby reducing the performance loss introduced by the dynamic analysis on the memory access dependency; and then splitting the cycles into two cores to run, and rewriting LLVM IR to generate the paralleled code. According to the method, the parallelism of the binary can be analyzed and extracted under the condition that a binary program source code cannot be obtained, and the corresponding binary program capable of being subjected to multi-core paralleling can be automatically generated, so that the effect of accelerating the original program by using the multi-core platform is achieved.

The invention discloses a manufacturing technology of purple lycium barbarum leaf tea. The manufacturing technology comprises the steps of picking and cleaning purple lycium barbarum leaves, performing deactivation of enzymes for four times, rolling for molding and drying to realize formation. To be specific, the preliminary low-temperature deactivation of enzymes is executed, then middle-temperature water bath fumigation is executed, and finally, high-temperature frying is executed; the first three deactivation of enzymes is implemented from low temperature to high temperature, and the temperature is moderate and is always below 90 DEG C, so that the activity of oxidase of the purple lycium barbarum leaves is inhibited while other substances in the purple lycium barbarum leaves cannot be damaged; rapid cooling freezing is executed, so that the deactivation of enzymes is complete, furthermore, the nutritional components and trace elements in the lycium barbarum leaves can be retained in the leaves to the maximum extent, and in particular, the content of Zn in the purple lycium barbarum leaf tea is much higher than that of common green lycium barbarum leaf tea.

The invention discloses a formal verification method for a sequencing specification of an FPGA (field programmable gate array) combinatorial logic system. The formal verification method comprises the following steps of (1) obtaining a VHDL (very-high-speed integrated circuit hardware description language) procedure according to an operation sequencing specification of the FPGA combinatorial logic system; (2) establishing a Petri net model according to the VHDL procedure; (3) establishing a reachability graph according to the Petri net model; (4) pitching on all stable gate circuit output states in the reachability graph; (5) generating a control specification comprising a logical formula for calculating tree tense according to the operation sequencing specification of the FPGA combinatory logic system; and (6) detecting whether the stable gate circuit output states in the reachability graph meet the control specification or not, determining that the states are mistaken if the stable gate circuit output states in the reachability graph do not meet the control specification, positioning mistaken statements in the VHDL procedure according to excitable changes generating the mistaken states, and determining that the VHDL procedure meets the design requirements if all the states meet the control specification. The formal verification method for the sequencing specification of the FPGA combinatorial logic system is rigorous and complete in logicality and low in complexity.

The invention discloses a tea leaf steaming machine suitable for preparing famous tea, which comprises a frame and a rotating drum. The rotating drum comprises a rotating drum cylinder section and a rotating drum conical section. Interval uniformly distributed annular electrical heating tubes are respectively arranged on the external surface of the cylinder section and conical section. The plane of the center line of each electrical heating tube is perpendicular to the axial line of the rotating drum. Each electrical heating tube is electrically connected to the temperature control unit on the top of the rotating drum. The interval between the electrical heating tubes on the external surface of the cylinder section is smaller than that between the electrical heating tubes on the external surface of the conical section. Compared with present technology, with the configuration, the invention has the advantages of no thermal inertia, uniform and fully steaming, and improved product quality.

The invention discloses a method for a virtual experiment apparatus. The method comprises the following steps of: firstly, shooting a photo of the experiment apparatus by using a digital camera, and saving the photo in the format of bmp; secondly, removing the background of the experiment apparatus by using the image background eliminating function of word2010, and saving an obtained image in the format of the bmp; thirdly, performing vectorization on an output image in the previous step by using software Inkscape, and saving an output file in the format of svg; and lastly, opening the output file in the previous step by using a text editor EditPlus, adding functional information (an interface, a display panel and the like) and multi-state information of the experiment apparatus, and saving the file in the format of xml. In the method, file format is XML-based, so that the experiment apparatus described by the file format can be used for various different virtual experiment platforms and has portability.

The invention discloses a finger vein recognition method combining a bionic texture feature and a linear texture feature. By adopting the method, the bionic texture feature and the linear texture feature of finger veins can be utilized effectively, the adverse effect of noise on the recognition accuracy is avoided, and the recognition performance and robustness of a fingerprint vein recognition system are enhanced. The bionic texture feature and the linear texture feature of a finger vein image are extracted respectively. The bionic texture feature is extracted by using a Gabor filter with good similarity of a visual cortex simple cell receptive field model of a mammal. When the method is used for performing feature recognition, the effect is similar to object recognition through human eyes, and a good effect is achieved.

The present invention provides a formula of composite seasoning comprising red rice, chrysanthemum leaf, pepper, sodium glutamate, cassia twig, aniseed, orange peel, garlic, ginger, white spirit, pricklyash, sesame oil, salt, Tsaoko, chicken bouillon and fennel. The present invention may be used in making several kinds of flavored food, such as delicious pigí»s trotters with health functions of invigorating spleen, invigorating stomach and beautifying.

In order to solve the technical problems existing in current technical proposals, the invention provides a preparation method of green buds. According to the method, the enzyme deactivation isthorough and does not affect the tea quality, after fragrance distilling, the moisture content is moderate, the fragrance retention time is prolonged, and the existing advantages of green buds can be ensured. Specifically, the preparation method of the green buds includes the steps of: tea harvesting, cleaning, tea tedding, rolling, enzyme deactivation, drying, grading, and fragrance distilling. Multiple enzyme deactivation is unnecessary, the enzyme deactivation is thorough, no scorching phenomenon can appear, quality of the green buds is guaranteed, and the relative content of tea polyphenol and catechin in the tea leaves is reduced, so that the bitter taste of tea is removed, the nutrients in the tea can be soaked out more easily, and the mouthfeel of the tea can be improved.

The invention relates to the technical field of tea processing, and specifically discloses a tea fixating apparatus and a tea processing method. The tea fixating apparatus comprises the following parts: a fixating tube; a first fixating screen, a second fixating screen and a third fixating screen which are sequentially arranged inside the fixating tube from the top to the bottom; and a vibrating part which is used for allowing the first fixating screen or the second fixating screen to vibrate up and down. A hot air circulation chamber is arranged, in isolated arrangement, at an inner bottom ofthe fixating tube via a clapboard with through holes; and a draft fan is arranged inside the hot air circulation chamber. The tea fixating apparatus disclosed by the invention is capable of carryingout hot-air fixation when tea leaves are vibrated to fall; so that, piling of tea leaves during fixation is avoided. Moreover, uniform hot air circulation and distribution are ensured, so that, high tea fixation efficiency and full fixation effect are achieved so as to have tea quality improved; moreover, the tea fixating apparatus also has a dust-removing effect.

The invention discloses a method and a device for encrypting credit assets by adopting a block chain technology, electronic equipment and a computer readable storage medium. The method comprises the following steps that a block chain supporting a smart contract is selected or created as an asset hosting chain; asset hosting registration is performed, namely the entrusting party generates a hostingtransaction on the encrypted asset chain by using a private key signature, and broadcasts the hosting transaction; asset transaction is performed, namely the asset hosting contract matches hosted assets according to a hosting strategy set by the entrusting party; two suggested transactions are generated and broadcast on the asset hosting chain; wherein the two suggested transactions are used forrealizing encrypted asset replacement in different entrusting party accounts, the corresponding entrusting party signs the suggested transactions and then broadcasts the signed transactions on the asset hosting chain, and the asset hosting contract attaches signatures of the entrusting party after receiving the two suggested transactions signed by the entrusting party and broadcasts the signatureson the corresponding encrypted asset chain.

The invention discloses a semi-supervised image retrieval method based on a self-coding network and robust kernel hashing. Step 1, depth image features are extracted by adopting a self-coding networkstructure; step 2, a robust kernel Hash mechanism is taken as a retrieval mechanism; and step 3, the self-coding network and the robust kernel Hash are combined to complete the retrieval process, andcomplete image features are extracted by using the self-coding network, and the image does not need to be labeled manually, so that the method has good learning ability and expression ability for different data sets, and has a better effect for large-scale data sets. Meanwhile, a Robust Hashwith Kernels (RSH) retrieval mechanism is proposed to be used, the robustness of the KSH method is improved,the calculation amount in the retrieval process is reduced, and finally the retrieval precision is improved and the storage space and the calculation complexity are greatly reduced through the imageretrieval algorithm combining the self-encoding network and the robust kernel Hash.