88results about How to "No time consuming" patented technology

The invention discloses a vehicle-mounted image semantic segmentation system based on a bilateral segmentation network. The system comprises a data storage module, which is used for storing a vehicle-mounted image training set and a vehicle-mounted image to be tested; a bilateral segmentation network consisting of a spatial channel and a context channel, wherein the spatial channel is used to extract the spatial information of the vehicle image, and the context channel is used to extract the context semantic information of the vehicle image; a training module for training a bilateral segmentation network using a vehicle-mounted image training set; a semantic segmentation module used for predicting the vehicle image to be tested by using the trained bilateral segmentation network, and obtaining the class to which each pixel in the vehicle image to be tested belongs. The invention discloses a bilateral segmentation network comprising a spatial channel and a context channel, wherein the spatial channel is used for extracting spatial information of an image while retaining enough spatial information, and the context channel is used for extracting context semantic information of the image while ensuring a large enough receptive field.

The invention discloses a cork pressing device. The cork pressing device comprises a support base plate and a support vertical plate, wherein the support base plate and the support vertical plate are mutually perpendicular. The support vertical plate is provided with a cork pressing mechanism, and a cork guide mechanism arranged below the cork pressing mechanism. The support base plate is provided with a locating mechanism used for fixing objects to be pressed. Centers of the cork pressing mechanism, the cork guide mechanism and the locating mechanism are collinear. The cork pressing mechanism is a crank-link mechanism. The support vertical plate is further provided with a slotted hole used for adjusting the height of the cork pressing mechanism. By rotating a crank, a cork can be pressed fast, operation is simple and convenient, much time and labor do not need to be consumed, and sealing efficiency is improved. By arranging the cork guide mechanism, the cork can be pressed correctly. Meanwhile, by arranging the locating mechanism, the objects to be pressed can be located accurately. By arranging the slotted hole used for adjusting the height of the cork pressing mechanism in the support vertical plate, adjustment can be carried out according to the height of the objects to be pressed. The cork pressing device is strong in genenrality.

The invention discloses a multimedia resource tag information generation method, apparatus, and device, and a storage medium and belongs to the technical field of Internet. The method comprises the following steps: obtaining review information of a target multimedia resource, and performing word segmentation processing on the review information; acquiring a word vector of at least one vocabulary obtained after the word segmentation; clustering the word vector of the at least one vocabulary to obtain a plurality of vocabulary classifications, wherein the different vocabulary classifications have different topic information; in the at least one vocabulary obtained after the word segmentation, extracting a key vocabulary of the target multimedia resource; based on the key vocabulary and the topic information of a plurality of the vocabulary classifications, generating tag information for the target multimedia resource. According to the multimedia resource tag information generation method, apparatus, and device, and the storage medium in the invention, complete automation is achieved when the tag information is generated, with no need to consume a large amount of manpower and time, and the intelligence is better; moreover, the generated tag information is more accurate, which improves the accuracy of subsequent multimedia resource recommendation.

The invention relates to a poisonous and harmful gas network monitoring instrument with a sensor detecting element and a monitored data processing method. The monitoring instrument comprises a shell,wherein a master control board, a detecting module, a data processing module and a power supply system are arranged in the shell; the detecting module, the data processing module and the power supplysystem are separately connected with the master control board; the detecting module comprises more than two poisonous and harmful gas sensors which detect various poisonous and harmful gases in real time; the master control board is provided with a signal acquiring unit, a wireless transmitting unit, a GPS positioning unit and a data storage unit; and the master control board is connected with a management system based on the internet of things through a wireless transmitting unit and transmits data of the detecting module to the management system through the internet of things. By the poisonous and harmful gas network monitoring instrument with the sensor detecting element and the monitored data processing method, the various poisonous and harmful gases in mixed gas can be accurately detected in real time, and the accuracy of a detection result is high.

The invention provides an electric energy meter upgrading method and system and a storage medium, and the method comprises the following steps that a client obtains the upgrading information of an electric energy meter, and generates an upgrading file package corresponding to a to-be-upgraded module in the electric energy meter according to the upgrading information; the client sends the upgradingfile package to the electric energy meter; the electric energy meter receives the upgrading file package sent by the client; and the electric energy meter determines an upgrading position corresponding to the module upgrading sub-package in the upgrading file package according to the upgrading file header information in the upgrading file package, and upgrades the to-be-upgraded module accordingto the module upgrading sub-package and the upgrading position. According to the invention, the client generates the upgrading file package corresponding to the to-be-upgraded module according to theupgrading information, and upgrades the corresponding to-be-upgraded module in the electric energy meter according to the upgrading file package sent by the client, so that in the process, the electric energy meter does not need to consume a large amount of time for overall upgrading, and the upgrading efficiency of the electric energy meter is greatly improved.

The invention relates to an energy spectrum analysis method for large-size samples. Based on an instantaneous gamma-ray neutron activation analysis technique, gamma energy spectra are obtained by a scintillator detector. The obtained gamma energy spectra are subjected to neutron self-shielding correction treatment according to the element nonlinear response. The gamma energy spectra are analyzed by a least square method, single element response energy spectra are obtained, a single element spectral library is established, and nuclides in the samples can be identified and analyzed by the singleelement spectral library. Aiming at the defect that other spectral library analysis methods require a large number of samples, the single element spectral library which is not affected by backgroundelements is obtained by a few of calibrated samples based on a physical mechanism of the influence of multielement components on PGNAA analysis results. The measured energy spectra can be fitted directly by the single element spectral library, and the fast and accurate analysis of the nuclides in complex samples having multiple components and different reaction cross sections can be achieved.

The embodiment of the invention provides an automatic text classification method and device and a storage medium. The method comprises the steps that acquring a to-be-classified text; and based on a text classification model, obtaining a text category of the to-be-classified text, the text classification model being obtained by taking the text sample set and the text category thereof as samples and adopting an LSTM long-short term memory artificial neural network for training. According to the method provided by the embodiment of the invention, the problems of low classification speed and lowaccuracy during text classification can be solved.

The embodiment of the invention provides a disassembling device of a display panel and relates to the technical field of display. The disassembling device can be used for conveniently and rapidly disassembling the display panel and has high disassembling efficiency and the disassembling damage rate is reduced. The disassembling device comprises a loading platform, a heating device, a panel fixing device, a cutting device and a height adjusting device, wherein the heating device is used for heating the display panel; the panel fixing device is used for fixing the display panel on the heating device; long-strip holes for communicating a cavity are formed in an upper surface of the loading platform; the long-strip holes are formed in two relative sides of the heating device; the cavity is internally provided with a supporting structure located below the long-strip holes; slide rails, which vertically correspond to the long-strip holes, are formed in the top of the supporting structure; the cutting device comprises a cutting part and slide blocks; the height adjusting device is used for adjusting the height, relative to the upper surface of the loading platform, of the cutting part; one end of each slide block is matched with the slide rails and the other ends of the slide blocks are used for fixing the cutting part and protrude] out of the long-strip holes; the cutting part can move along the long-strip holes under the driving the slide blocks, so as to cut frame sealing glue. The disassembling device is used for disassembling the display panel.

The invention discloses a singing synthesis method, a singing synthesis device, a singing synthesis facility, and a computer readable storage medium. The singing synthesis method comprises the following steps: acquiring a to-be-synthetized music score with the preset number format; extracting music score features of the to-be-synthetized music score; inputting the music score features into a preset Gaussian Mixture Model-Hidden Markov Model for processing, and outputting the corresponding acoustic features; and through an acoustic code synthesizer, synthetizing the output acoustic features into singing audio without accompaniment corresponding to the to-be-synthetized music score. Data used in training the Gaussian Mixture Model-Hidden Markov Model is far smaller than the data needed by the existing singing synthesis, thus abundant manpower and time consumed for collecting data are not needed, and therefore, the singing synthesis difficulty is reduced.

The invention discloses a weld joint pore defect detection method based on image processing. The method comprises the following steps: carrying out binarization processing on an input pore defect to-be-detected image to obtain a binary image imgb1; performing closing operation and opening operation on the binary image imgb1 in sequence to obtain a binary image imgb2; extracting and putting all connected domains of the binary image imgb2 into a summary set; traversing the summarized set, extracting a connected domain which transversely passes through the image, and putting the connected domain into a first screening set; if only one element exists in the first screening set, taking the connected domain as the contour of the welding joint area, and searching the pore defect area directly; otherwise, searching a steel pipe area; extracting a target welding seam area; searching an air hole defect area; and extracting all pore defect areas based on edge detection, and fitting the contour by adopting a least square method. According to the invention, the to-be-detected image of the pore defect is directly processed, the weld pore defect is detected by combining connected domain searching and edge detection, and the method has good precision and accuracy.

The invention discloses a payment rejection processing system generation method and device. The payment rejection processing system generation method comprises the steps that a trigger event set and aplurality of state nodes contained in system configuration information are acquired, and a connection relation is pre-configured for trigger events in the trigger event set and the state nodes; determining a root state node from the system configuration information, and generating a state tree according to a trigger event in the trigger event set and a connection relation pre-configured for the state node; in the state tree, the current task amount corresponding to any non-root state node is calculated according to the responsibility judgment part in the current trigger event, the corresponding responsibility judgment amount and the last task amount, and any leaf state node is a state node which cannot be migrated to the next target state node; traversing the state tree to obtain directedpaths formed from the root state node to any leaf state node, and calling the state nodes on the directed paths and the component resources corresponding to the trigger events to generate a rejectionprocessing system.

The invention relates to a disinfecting device, in particular to a self-blowing type foot flushing and disinfecting device. The technical problem is how to design a self-blowing type foot flushing anddisinfecting device which can uniformly spray a disinfectant in each position of feet for disinfection during disinfection and is fast in disinfection process. The self-blowing type foot flushing andsterilizing device comprises a base, a liquid storage tank, connecting racks, a liquid inlet one-way pipe, a piston block, a shaft sleeve and a cylinder, wherein the liquid storage tank for storing the disinfectant is arranged on the base, the connecting racks are arranged on two sides of the top of the liquid storage tank, and the cylinder is arranged above the connecting racks. Feet are fixed on the disinfecting device by an opening and closing mechanism, a support net frame is stepped to drive a nozzle to spray the disinfectant to perform comprehensive disinfection on feet, after disinfection, a clamping mechanism is loosened and a rotary mechanism is stepped to operate to add the disinfectant, residual disinfectant on feet is washed away, and the disinfectant remaining on feet is avoided.

The invention relates to the important fields of artificial intelligence and mode identification, namely, image identification and human face identification, in particular to a color image over-complete block feature extraction method based on a non-iterative bilateral two-dimensional principal component analysis (NIB2DPCA) method. The method comprises: performing over-complete block segmentation on a color image; then performing feature extraction and reconstruction on a sub-image module with the NIB2DPCA method from R, G and B channels; and performing multi-module fusion to finally obtain a classification feature matrix. According to the method, the amount of extracted information is far greater than that of an original image, so that the identification rate of the color image is increased. The method has higher identification accuracy and higher identification speed. The method is applied to the field of human face identification, and the identification speed of the method is increased by a few orders of magnitudes in comparison with that of an over-complete block segmentation based human face identification method for deeply hidden identity characteristics.

The invention discloses a fixing device of a box body. The box body is detachably connected with the fixing device; when the box body is installed, a cam is rotated through a handle, so that two slide blocks slide towards two opposite directions so as to be opened; an installing stand column on the box body is inserted from a second through hole on a bottom part of a shell body; the handle is loosened; the slide blocks slide under the action of a reset mechanism so as to clamp the installing stand column; and a locking part penetrates through a third through hole on the shell body and a first through hole on the installing stand column so as to clamp and fix the installing stand column. When the box body is detached, the locking part is pulled out, so that the box body can be separated from the fixing device. The fixing device of the box body exists in an assembly form, the structure is simple, the box body can be simply and conveniently installed and detached through the fixing device, less labor force and time are consumed, the box body can be installed and detached easily, the time and the labor are saved, and the use is convenient.

The present invention discloses a drying device capable of uniformly drying yellow croaker fillets. The drying device comprises a base seat and a cover plate, an upper part of the base seat is connected with a device main body via a telescopic support rod, a heating cavity is arranged inside the device main body, besides, a drying chamber is arranged inside the heating cavity, besides, a rotatingshaft is arranged inside the drying chamber, upper and lower sides of placing plates are all connected with placing frame bodies through fixed shafts and rotating sleeves, clamping plates are arrangedin middle of the placing frame bodies, and bottom parts of the clamping plates are connected with fixed plates via compression springs. The drying device capable of uniformly drying the yellow croaker fillets can very good conduct uniform and thorough drying work of the yellow croaker fillets during use processes, can very well fully utilize spaces above the placing plates during the use processes, and facilitates disassembly and installation of the placing plates during the use processes.

The invention discloses guiding forceps convenient to clamp and fix for use. The guiding forceps comprise forcep handles, wherein one end of each forcep handle is fixedly connected with a force application hole, the other end of the force application hole is hinged with a clamping handle, a C-shaped groove is formed in the side surface of the clamping handle, a drive box is fixedly mounted on the side surface of an inner cavity of the C-shaped groove, the bottom of an inner cavity of the drive box is fixedly connected with a fixing clamp, a drive motor is in inserting connection to the top of the fixing clamp, an output shaft of the drive motor is fixedly connected with a screw rod rotationally connected with the top of an inner cavity of the fixing clamp, the screw rod is sleeved with a screw rod sleeve, a displacement rod penetrating the drive box and extending to the side surface of the drive box is fixedly mounted on the side surface of the screw rod sleeve, and a buffer layer is fixedly connected with the end, far from the screw rod sleeve, of the displacement rod. By means of the C-shaped grooves in the clamping handles, ends of a shunting tube can be inserted into the C-shaped grooves, so that the shunting tube can reach the abdominal cavity, and the life safety of a patient is protected to a certain extent.