31results about How to "Performance limitations" patented technology

Disclosed is a hyperspectral data classification method which is combined spectrum and spatial information. The steps comprises (1) reading the hypersectral data, (2) confirming the minimum size of structural element, (3) calculating differentiation between picture elements in neighborhood of each structural element by extended mathematical morphology expansion and corrosion operation, (4) obtaining exponential value of morphology eccentricity by the extended expansion and the corrosion operation of step (3), (5), constantly repeating the above steps with the adding of the size of the structural element to achieve the maximum size of the structural element, (6), constantly updating the exponential value MEI of morphology eccentricity in iteration process via the obtained new value, and generating a final exponential value MEI of morphology eccentricity after the iteration process is finished, (7) realizing the extraction of the data characteristic by the image of the exponential value MEI of morphology eccentricity, namely generating ground object type information, and realizing sophisticated category of the ground object by a minimum-distance classifier. The method is an unsupervised classification method for hyperspectral ground object with strong stability, high reliability and high accuracy.

The invention relates to a method for automatically extracting end members by combination of expanding morphology and orthogonal subspace projection, which comprises the following steps: (1) reading high-spectrum data; (2) determining the dimension, the initial iteration and the maximum iteration of a structural element; (3) calculating pixels with minimum mixing degree and maximum mixing degree in neighboring areas of the structural element through expansion and corrosion operations; (4) calculating and obtaining the morphological eccentricity index value through the result obtained in the step (3); (5) repeating the step (3) and the step (4) along with the increase of the iteration, and utilizing the result of expansion operation in the step (3) to update image data until the maximum iteration is reached; (6) performing binaryzation on MEI images and obtaining an end member data set; (7) calculating and obtaining a first end member by a spectral corner matching method, and updating the end member data set by the projection in an orthogonal subspace of the obtained end member; and (8) repeating the step (7) until the error requirement is met. The invention is a method for automatically extracting the high-spectrum end members with strong stability, high reliability and high precision.

The invention discloses a temperature-sensitive hepatic cell culture support material and a preparation method thereof, and relates to an intelligent tissue engineering support material and technology, in particular to a temperature-sensitive hepatic cell culture support material, a method and technology. The support material is characterized in that: the surface of the material is relatively lyophobic at the temperature of 37 DEG C and is suitable for cell adherence/spreading and cell propagation, but the surface is hydrophilic at the temperature of 32 DEG C, so cells are desorbed from the surface automatically. The preparation method of the cell culture support material comprises the following steps of: (1) surface cleaning; (2) the temperature-sensitive carboxylation modification of a TCPS surface; (3) the preparation of amido lactobionic acid (L-NH2); and (4) the saccharifying modification of the TCPS surface. Through the preparation method, the biocompatibility and cell-bound loci of the support material are enhanced, the adsorption rate of the cells on the material surface is increased, and the intelligent support material which is suitable for hepatic tissue engineering is obtained.

The invention is suitable for the technical field of terminals, and provides a method and device for guaranteeing the security of POS machine firmware. The method includes the steps that according to the type of a CPU, a loading mode corresponding to the type of the CPU is preset, and according to the loading mode, first stage starting firmware is loaded by selecting an embedded multi-medium card e MMC starting medium. With the method, loading from other starting media capable of being connected with the exterior can be avoided; it is prevented that through the starting media, the firmware in a POS machine is replaced or tampered, and therefore it is guaranteed that the POS machine meets the requirement for security.

The invention provides a media stream processing method and system and a home gateway. The method comprises the following steps that: the home gateway receives a calling request; the home gateway consults with an external network terminal through an SIP (Session Initiation Protocol) to determine a port for both parties to receive media streams; the home gateway specifies a port used for receiving data streams for a first media terminal, establishes an NAT (Network Address Translation) forwarding mapping relation between a port used by the home gateway and a port used by the first media terminal, and transmits the port used by the first media terminal, an IP (Internet Protocol) address of the external network terminal and a port used by the external network terminal to the first media terminal; the home gateway receives incoming call media streams transmitted by the external network terminal and forwards the incoming call media streams to the first media terminal according to the NAT forwarding mapping relation; and the home gateway receives outgoing call media streams transmitted to the external network terminal by the first media terminal and forwards to the external network terminal. By adopting the technical scheme of the invention, a visual telephone service can be realized in a home network or a similar network environment, and various problems existing in the prior art are solved.

The invention discloses an image acquisition method and system for a panorama image. The image acquisition method comprises an image acquisition step in a camera pitching and/or horizontal direction,wherein the image acquisition step in the camera pitching direction comprises the following steps: S11, shooting a first pitching image by taking a camera pitching angle as beta1, and shooting a second pitching image by taking a camera pitching angle as beta2 and in the same camera horizontal angle when the first pitching image is shot; S12, calculating a first overlapping degree of the second pitching image on the first pitching image; and S13, judging whether the first overlapping degree is higher than or equal to an overlapping degree threshold value, if yes, taking a target camera pitchingangle (described in the specification) as beta2, shooting an image by a camera in the target camera pitching angle (described in the specification), if not, assigning a value of (beta2-beta0) to beta2, wherein beta0 is a first decrease angle, and returning to the step S11. The image acquisition method disclosed by the invention improves splicing success rate and image acquisition efficiency and can be applied to the field of shooting of panorama images of unmanned aerial vehicles.

The invention discloses a read command acceleration method and device of an NVMe SSD main control chip, computer equipment and a storage medium. The method comprises the steps that read data command information is cached in an acceleration module, and NFC is notified; the NFC reads corresponding data information from the flash according to the command information and sends the data information tothe acceleration module; and the acceleration module sends the processed valid data to the host. The limitation of the DDR reading and writing speed on the overall performance is avoided, and the usefrequency of a CPU is reduced, so that the data reading performance of the SSD main control chip is improved.

The invention belongs to the technical field of aluminum alloy processing, and particularly relates to a 6XXX series aluminum alloy and an aging process thereof. The aging process of the 6XXX series aluminum alloy comprises first-stage low-temperature pre-aging and second-stage high-temperature aging, wherein the temperature of the first-stage low-temperature pre-aging is 80 to 150 DEG C. The low-temperature pre-aging treatment provided by the invention, on the one hand, causes a large number of pre-beta" phases to be dispersively distributed in the alloy, and these pre-beta" phases are directly transformed into beta" phases in a subsequent aging process; on the other hand, the size of nano-scale atomic clusters formed in an original parking process is further increased, so that the nano-scale atomic clusters can provide nucleation positions for the nucleation of the beta" phases in the subsequent aging process; therefore, the precipitation density of the beta" phases in the final agedalloy is promoted, the uniform distribution of the beta" phases is promoted, the purpose of eliminating the quenching and parking effect is achieved, and at the same time, the final mechanical properties of the aluminum alloy are improved, and the task scheduling problem caused by the production capacity limitation of an enterprise is solved.

The invention relates to a fiber grating sensor package structure suitable for a high-temperature environment. The fiber grating sensor package structure comprises an upper layer package and a lower layer package; the lower layer package is provided with an optical fiber placement portion; two optical fiber fixing through holes which are spaced apart are formed in the lower layer package; the optical fiber placement portion is divided into a first placement portion, a second placement portion, and a third placement portion through the optical fiber fixing through holes; the first placement portion is located between the two optical fiber fixing through holes; the groove depth of the first placement portion is greater than the groove depths of the second placement portion and the third placement portion; the upper layer package includes two adhesive injection portions; the two adhesive injection portions are respectively corresponding to the two optical fiber fixing through holes; and adhesive injection holes are formed in the tops of the adhesive injection portions. The invention also provides a fiber grating sensor packaging method suitable for a high-temperature environment. Whenthe package structure of the invention is installed, a fiber grating sensor is directly bonded to a device to be tested, and therefore, measurement error caused by the problem of strain transmissioncan be avoided; and a two-point type bonding method is adopted, and therefore, the influence of an adhesive on the performance of the fiber grating sensor can be avoided.

The invention discloses a display screen, a manufacturing method of the display screen and electronic equipment, and belongs to the technical field of display equipment. The display screen comprises a light emitting layer, wherein the light emitting layer comprises an infrared display region, the infrared display region comprises a first visible light emitting unit and an infrared light-emitting unit, the first visible light emitting unit is provided with a mounting hole, the infrared light-emitting unit is arranged in the mounting hole, a light filtering layer and the infrared light emitting unit are oppositely arranged, and the projection of the filter layer towards the light-emitting layer covers the infrared light-emitting unit. According to the display screen, the problem of poor performance of an infrared device below the display region of the display screen can be solved.

The invention discloses a transverse-structure IMPATT diode and a preparation method of the IMPATT diode. The IMPATT diode includes: a substrate layer, an epitaxial layer, a drift layer, an n-AlGaN barrier layer, an n+-GaN barrier layer, a left ohmic contact layer, a right ohmic contact layer, a left ohmic contact electrode, a right ohmic contact electrode, a passivation layer and a Schottky contact electrode. When the transverse-structure IMPATT diode is electrified, the current direction is along the transverse direction of the epitaxial layer; the drift layer is a two-dimensional electrongas thin layer formed on the top of the epitaxial layer, and the transition process is limited in the drift layer rather than in a body material. The IMPATT provided by the invention improves the oscillation frequency, the transverse circuit compatibility and the frequency flexibility compared with a traditional vertical-structure IMPATT by utilizing a transverse structure under the same material.

The invention discloses a method for testing the normal strength between the layers of a composite material, and belongs to the technical field of composite materials and material performance testing.According to the method, the specimen is a square flat plate block-like body; during the paving adhesion, a polytetrafluoroethylene film doing not cover the test area is inserted between the upper paving layer and the lower paving layer of the predetermined testing interface, and curing molding is performed; and the strength between the layers of the composite material is tested by using a testing device. According to the present invention, the method is used for testing the normal strength between the layers of the composite material, and breaks through the limitation on the interlayer performance of the to-be-tested specimen in the conventional method; and the corresponding size of the hollowed-out zone of the PTFE film can be designed according to the interlayer performance of the to-be-tested composite material, such that the normal damage load between the layers in the test can be controlled, and the interlayer damage of the composite material precedes the adhesive damage betweenthe metal block and the specimen so as to ensure the damage mode.

The invention discloses an information system for penetration test and bug repair and a method thereof, and the system comprises a target system penetration unit which is used for enabling a bug device to penetrate a target system. An intermediate terminal and the information system used for penetration test and bug repair are located in the same office or studio, and the intermediate terminal andthe information system used for penetration test and bug repair are both connected with a receiving terminal of a client. The plurality of electronic documents can be transmitted synchronously, a handshake link can be constructed for each electronic document, and the plurality of electronic documents can share one handshake link. The defects that in the prior art, the use performance is poor dueto the fact that the use rate of a large number of CPUs and memories is consumed, and the performance of synchronously transmitting the first electronic document or the bug repairing unit is greatly limited due to the fact that a current transmission mode is continuously increased along with a test terminal or an intermediate terminal are effectively avoided.

The invention discloses an image processing method and device, computer equipment and a storage medium, and the method comprises the steps: carrying out the feature extraction of a received first image, and obtaining a plurality of image original features of the first image; generating an identity identification feature of the first image by using the plurality of image original features, determining a feature distance between each image original feature and the identity identification feature, and determining an optimization parameter according to the feature distance; and updating a loss function output result used for iterative training of the image recognition model by using the optimization parameter, wherein the trained image recognition model is used for performing image recognition processing on a to-be-recognized second image. The device can comprise an original feature extraction module, an identification feature generation module, an optimization parameter determination module and a loss output updating module. Compared with the prior art, the image recognition processing capability of the image recognition model can be remarkably improved, and the image recognition effect is remarkably improved.

A data processing device comprising several processing circuits each operating under the control of its periodic clock signal so that the clock signal may have different frequencies and/or may be automatic. Each of the several processing circuits has an output that outputs memory access requests, each request held at the output for an interval of validity duration defined by the clock signal of the particular processor. A multiplexing circuit multiplexes these access requests to a certain memory. A memory requires a minimum memory repeat cycle before it can accept an access request following the acceptance of a previous access request. The clock period of the processing circuit is longer than the minimum memory repetition period. The timing circuit selects the acceptance time point of each specific access request from the first data processing circuit. The point in time at which this specific request is accepted is always within the validity duration interval for which this specific access request was made. The timing circuit shifts the location of the accept time point in the validity duration interval, thereby delaying the location to allow for a previous acceptance of an access request from another processor. Subsequently, during application of sequential access requests from the first data processing circuit, the position is moved back to the start of the validity duration interval using sequential steps.

The embodiment of the invention discloses a message security processing system and a method. The message security processing system comprises a processor, a hardware security co-processing module anda network processing module, and the processor is integrated with a software security protocol and a control module; the network processing module is connected with at least one network interface, andsends the message received from the network interface to the hardware security co-processing module for security processing; a message fed back after processing of the hardware security co-processingmodule is completed is sent out through a network interface; and the hardware security co-processing module is used for performing security processing on the received message sent by the network processing module and feeding back the message to the network processing module. According to the system of the embodiment of the invention, through interaction between the hardware security co-processingmodule and the network processing module, the effect that the message security processing does not need to occupy the running memory of the processor is realized, so that the processing performance of the message security characteristic is not limited by the performance of the processor.