679results about How to "Low resolution" patented technology

An e-Petri dish comprising a transparent layer having a specimen surface and a light detector configured to sample a sequence of sub-pixel shifted projection images of a specimen located on the specimen surface. The sub-pixel shifted projection images associated with light from a plurality of illumination angles provided by an illumination source.

The invention provides a commodity recommendation method and system based on video content to solve the problem that the accuracy of commodity recommendation results is low. The method comprises the steps that keyframes are extracted from target video; object detection is conducted to each keyframe according to objectivity of objects, and the detected objects serve as candidate objects of each keyframe; commodity identification is conducted to the candidate objects of each keyframe through a convolutional neural network model, and commodities of each keyframe are obtained; keyword mapping is conducted to the commodities of each keyframe, and corresponding commodity tags are obtained to serve as the commodity tags of each keyframe; the commodity tags of each keyframe are determined as commodity recommendation information of the target video, and the commodity recommendation information is output. According to the commodity recommendation method and system based on the video content, the video content is taken into consideration when commodity recommendation is conducted, the video content is analyzed, and the obtained commodity recommendation information is closely bound up with the content of the target video, so that the accuracy of the the commodity recommendation results is improved.

Techniques for imaging a search region to detect a dielectric target include placing multiple electrodes outside the search region. At least two electrodes are activated independently of each other. Emitter circuits are connected to corresponding emitter electrodes. Each emitter circuit is configured for using its emitter electrode to produce an electric field with wavelength longer than about one hundred meters. Receiver circuits are connected to corresponding receiver electrodes. Each receiver circuit is configured for using its receiver electrode to measure a property of an electric field produced, at least in part, by an emitter electrode. A processor determines a property of a dielectric target inside the search region based on measurements from the receiver circuits using a subset of the emitter circuits. Among other uses, these techniques allow humans to be detected inside building-size regions, even when hidden by visually opaque blocking material of small dielectric constant.

The invention provides a target detection method based on SSD improvement, and the method selects ResNet-101 to replace VGG-16 as a basic network of a whole model, and provides richer feature map information. A feature fusion strategy is adopted, so that the detection effect of multi-scale prediction of a network prediction layer on a small target is improved to a certain extent; an SE-block module is added to a classification branch and used for capturing global environment information of the feature map and outputting the feature map with channel weight, so that classification is more accurate; and a Centerness parallel to the classification prediction layer is added to suppress a low-quality bounding box so as to improve the detection precision. According to the method, anchor-frame-free detection is adopted, all hyper-parameters related to the anchor frame are avoided, the hyper-parameter amount is greatly reduced through the adopted prediction bounding box strategy, the network design complexity is reduced, and the training stage becomes very simple. According to the method, a loss function adopts a focalloss function, so that the model detection precision is improved while the detection speed is kept.

A JPEG2000 compressed image is transcoded to a lower bit-rate or lower resolution, or both, without having to decompress the initial JPEG2000 image and then recompress it to a lower bit-rate and/or resolution. Instead, arithmetic decoding is performed only to the nearest higher bit-rate layer, up to the desired resolution, before performing rate-distortion optimization to produce the transcoded image.

The invention discloses a preparation method of ganoderma lucidum polysaccharide with high yield, which comprises the following steps: firstly, crushing ganoderma lucidum fruiting bodies into ganoderma lucidum powder, then sequentially extracting the ganoderma lucidum powder by adopting a microwave-assisted hot water extraction method and an ultrasonic-assisted hot water extraction method, and carrying out concentration, impurity removal, concentration, alcohol precipitation and drying on the obtained extracting solution to obtain water-soluble ganoderma lucidum polysaccharide; and extracting the ganoderma lucidum residues by adopting an alkali extraction method, and carrying out neutralization, dialysis, concentration, impurity removal, alcohol precipitation and drying on the obtained extracting solution to obtain alkali-soluble ganoderma lucidum polysaccharide. The method is used for extracting the ganoderma lucidum polysaccharide by combining microwave heating action and ultrasonic cavitation and extracting the alkali-soluble polysaccharide in the ganoderma lucidum by utilizing the alkali extraction method, thereby reducing the loss of the ganoderma lucidum polysaccharide; and the prepared ganoderma lucidum polysaccharide has the characteristics of high production yield, strong immunological activity and the like.

A printable component structure includes a chiplet having a semiconductor structure with a top side and a bottom side, one or more top electrical contacts on the top side of the semiconductor structure, and one or more bottom electrical contacts on the bottom side of the semiconductor structure. One or more electrically conductive spikes are in electrical contact with the one or more top electrical contacts. Each spike protrudes from the top side of the semiconductor structure or a layer in contact with the top side of the semiconductor structure.

The invention discloses a data quality monitoring method for a local enhancing system, which mainly solves the problem that the satellite orbit position calculation accuracy by the traditional data quality monitoring method is relatively low. The method comprises the following steps of: firstly, checking the data integrity and validity of a received broadcast ephemeris to acquire a group of accurate and valid ephemeris data; secondly, checking the consistency of the group of ephemeris data; thirdly, carrying out real-time precision ephemeris detection, new and old ephemeris detection and ephemeris-almanac detection on the broadcast ephemeris data with uniformity respectively; and finally, according to the data availability detection result, generating an available matrix and an available broadcast ephemeris group of the broadcast ephemeris. The method has the advantages of high satellite orbit position calculation accuracy and low ephemeris fault error detection probability and can effectively solve the problem of man-made interference on the broadcast ephemeris and almanac of a satellite navigation system.

The invention discloses a bullet screen display method, a bullet screen adding method, a bullet screen display apparatus and a bullet screen display system, and belongs to the field of online video playing. The bullet screen display method comprises the steps of: acquiring a first type of bullet screens which corresponds to an online video and are added by network medias and a second type of bullet screens added by audiences of the online video; determining an audience client for playing the online video; and sending the first type of bullet screens and the second type of bullet screens to the audience client, and carrying out visual highlighting display on the first type of bullet screens relative to the second type of bullet screens, wherein the audience client is used for displaying the first type of bullet screens and the second type of bullet screens in the process of playing the online video. Effects of partitioning users into the network medias and the audiences, carrying out highlighting display on bullet screens sent by the network medias, enabling the bullet screens sent by the network medias to be more obvious than common bullet screens sent by the audiences on the aspect of the visual effect and improving the distinguishing degree of bullet screens sent by important users are achieved.

The embodiment of the invention discloses a blood vessel segmentation method and device, an electronic device and a storage medium. The method comprises the following steps of respectively inputting an image to be segmented into at least two pre-trained set neural network models, and generating an initial segmentation result corresponding to each set neural network model; carrying out weighting processing on each initial segmentation result by utilizing a target model weight value corresponding to each set neural network model to generate a weighted segmentation result, the target model weightvalue being determined when the set neural network model is trained; and carrying out image post-processing on the weighted segmentation result to generate a segmented blood vessel image. According to the technical scheme, the Boosting modeling idea is applied to the field of blood vessel segmentation for the first time, the problem that a blood vessel segmentation result obtained through a single neural network model is low in precision is solved, and the effect of improving the blood vessel segmentation precision is achieved.

The present invention provides water soluble chitosan microspheres which can be used for loading drug and a preparation method thereof. The structure of the microsphere is that the outside of crosslinking state water soluble chitosan core which can load the drug is orderly enwrapped with a crosslinking state water soluble chitosan coating layer and a crosslinking solidified outer surface layer, which can be used for loading and controlling releasing the drug. The preparation method is that the drug which is mixed with the water soluble chitosan and corresponding water dispersion and oil component are used to form W/O type mixture which is reacted with twain-aldehyde crosslinking agent to obtain microsphere core; water soluble chitosan solution is used to form crosslinking state coating layar at the outer side of the microsphere core; finally, the crosslinking solidified outer surface layer is formed through the second crosslinking reaction. The chitosan microsphere for loading the drug has controllable particle size distribution and slow-release effect, obviously improved intensity and good biological compatibility and biological adhesion, which enlarges the application range for loading drug and has good effect of drug loading, controlling and releasing; the chitosan microsphere for loading the drug can be applied to a plurality of medication forms, such as hypodermic, intravenous and intraperitoneal injection medications, orally taking, etc., the preparation method is simple and feasible with mild condition and low production cost.

The invention relates to the field of rock fracture prediction, in particular to a shale hydraulic fracture propagation prediction method. The method comprises the steps that 1, the normal direction,tangential stress and effect stress of an inclined fracture under the action of the external stress and water pressure are calculated; 2, a strain energy density function is obtained according to thetype of the fracture; 3, a strain energy density factor is obtained according to the strain energy density function; 4, the propagation direction and propagation angle of the fracture are judged according to a strain energy density criterion; 5, influences of stratification, natural fractures and the like on the propagation direction of hydraulic fractures are obtained through numerical simulation, so that fracture propagation under the effect of shale hydraulic pressure is predicted. The prediction method is based on fracture mechanics, a hydraulic condition factor is introduced, and the influences of different stratification directions on the propagation direction of the shale hydraulic fractures are obtained by researching the relation between the fracture propagation direction and themagnitude of the hydraulic pressure and utilizing an extended finite element method, so that accurate prediction of hydraulic fracture propagation is achieved, and higher prediction accuracy is obtained.

The invention relates to the field of artificial intelligence, and discloses a medical knowledge graph construction method and a device, equipment and a storage medium. According to the method, a knowledge feature extraction mode is determined on the basis of a collected data source to analyze medical data in a targeted mode, so that accurate entity names are obtained, entities with different entity names but the same ontology are fused through a fusion algorithm, the entities are unified, and finally a knowledge graph is constructed on the basis of the fused entities. Therefore, the problem of low knowledge graph precision caused by non-uniform related medical knowledge entities is solved, powerful support and guarantee are provided for diagnosis means and methods in the medical field, and knowledge sharing and management can be completed more reasonably and efficiently. In addition, the invention also relates to a blockchain technology, and the medical knowledge graph and the entitycan be stored in the blockchain.

The invention discloses a control system, method and device for the exhaust temperature of a compressor. The control system comprises a temperature sensor, a controller and an electronic expansion valve, wherein the temperature sensor is arranged on an exhaust channel of an exhaust outlet of the compressor and used for collecting the current exhaust temperature of the compressor; the controller is connected with the temperature sensor and used for determining an adjusting parameter of the target opening degree of the electronic expansion valve on the basis of the current exhaust temperature and generating a control signal where the adjusting parameter is recorded; and the electronic expansion valve is connected with the controller, installed on a liquid spraying channel of the compressor and used for adjusting the opening degree to the target opening degree according to the adjusting parameter under triggering of the control signal. By means of the control system, method and device for the exhaust temperature of the compressor, the technical problem that in the prior art, precision for adjusting the exhaust temperature of the compressor is low is solved.

A vehicle lighting device includes a projector lens, a lens holder, a main body barrel, a bottom cover, a first to a fourth excitation light sources, an upper and a lower light deflector, a correction mirror, and a fluorescent unit. The first and the second excitation light source emit laser beams toward the rotational center of a light-deflecting mirror of the upper light deflector, and the third and the fourth excitation light source emit laser beams toward that of the lower light deflector. The correction mirror corrects the distortion of the laser beams scanned by the upper and lower light deflectors and reflects the corrected laser beams. A two-dimensional image corresponding to a predetermined light distribution pattern is drawn as a white image on the fluorescent unit by the corrected light beams, and the two-dimensional image is projected on a virtual vertical screen through the projector lens.

The invention discloses an experimental facility and an experimental method for a plant photosynthesis research. The experimental facility comprises a plant rowing device, semiconductor light sources, a first processor, a gas detecting device, a temperature sensor and video gathering modules, wherein the plant rowing device comprises a plurality of growing chambers, each of the semiconductor light sources which are adjustable in wave length and illumination intensity is arranged inside each of the growing chambers, so that each semiconductor light source in each plant growing chamber can directly emit light with wave length or illumination intensity, the wave length or the illumination intensity is needed by each plant growing chamber, and accuracy of a research result is improved. Moreover, the video gathering modules used for gathering videos of plant growth conditions are arranged on the positions of through holes between adjacent plant growing chambers, 360-degree real-time monitoring of plants to be researched in each plant growing chamber is achieved, growth conditions of the plants to be researched under different spectra and illumination intensities can be obtained in a continuous or real-time mode, and reliable data are provided for analysis of the growth conditions of the plants to be researched under different spectra and illumination intensities.

The invention discloses an online detection and correction machining method for an enveloping worm tooth surface. Ground finish is carried out on the enveloping worm tooth surface through four-axis linkage in combination with a virtual rotation center principle, online detection is conducted on the enveloping worm tooth surface through three-axis linkage in combination with a spherical measuring head, and machining precision of the enveloping worm tooth surface is obtained through comparison of a practical torus helical line obtained through measurement with a theoretical helical line; whether the machining precision of the enveloping worm tooth surface meets requirements is judged, source tracking processing is carried out on a machining error if the requirements are not reached, grinding program correction and machining tool adjustment is conducted on the machining error value obtained by means of source tracking, and then tooth surface correction ground finish is carried out; the process is repeated until the machining precision of the enveloping worm tooth surface meets the requirements. The method solves the problem of low precision of the enveloping worm tooth surface fundamentally, can be applied to machining tooth surfaces of planar double enveloping worms, toroid enveloping torus worms, involute enveloping torus worms and other enveloping worms, and has principle universality and wide application value.

The invention discloses a method for recognizing a clamping plate furcal hole through an intelligent forklift. The method comprises the steps that a laser scanner acquires the data of the surrounding environment of a fork arm, and passes the environment data back to a software program in the intelligent forklift in real time; the software program analyzes the received environment data, extracts relevant characteristic values of a clamping plate, carries out wave filtering on consecutive points, and filters discrete points; interference point clusters in point clusters generated after the consecutive points are subjected to wave filtering according to the relevant characteristic values of the clamping plate are filtered, cargo pier sets are found, and the positional relationship among the cargo pier sets is analyzed; and the position of the furcal hole of the clamping plate is found and tracked, and the intelligent forklift is guided to complete the furcal hole entering action. The invention further relates to a device for achieving the method. The method and the device for recognizing the clamping plate furcal hole through the intelligent forklift have the beneficial effects that the furcal hole entering accuracy of the intelligent forklift can be improved, and the intelligent forklift can be automatically guided to complete the furcal hole entering action.

The invention discloses an energy imaging system and an application thereof in a motor device fault detection method. The system comprises a plurality of collection device terminals, a plurality of edge computing devices, and a cloud server. Each edge computing device is in communication connection with at least one collection device terminal. The cloud server is in communication connection with each edge computing device. The motor device fault detection method adopts the system and a device fault classification algorithm based on a deep learning improved sparse autoencoder algorithm. The system and the method can accurately predict potential faults that may occur in the operation of the motor device, so as to achieve the gain effect of early prevention and overall repair. Meanwhile, a traditional cloud computing method with large computational amount, complicated computing process, and high requirements on hardware equipment is replaced by an edge computing method with small computational amount, fast computational speed, and low requirements on hardware, so that the computational efficiency is improved.