Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Hiro

38results about How to "Does not affect processing speed" patented technology

Computerized tomography (CT) image metal artifact correction method, device and computerized tomography (CT) apparatus

The invention provides a computerized tomography (CT) image metal artifact correction method, a computerized tomography (CT) image metal artifact correction device and a computerized tomography (CT) apparatus. The computerized tomography (CT) image metal artifact correction method comprises the following steps that: a metal projection range caused by an interference object is determined according to an original image corresponding to original projection data; diagnosis object projection data after the removal of the interference object are obtained based on metal projection data in the metal projection range, and after that, the original projection data are corrected and a model image is constructed based on the diagnosis object projection data; and secondary correction is performed on the original projection data according to the projection data of the model image, and reconstruction is performed based on corrected target projection data and according to clinically-used scanning and image construction conditions so as to obtain a metal artifact-free target image, and therefore, the purpose of metal artifact correction can be achieved. According to the computerized tomography (CT) image metal artifact correction method of the invention, the original projection data are adopted as a correction object, and therefore, the spatial resolution and low-contrast ability of a processed image can be ensured; and the original projection data completely contain all information of the interference object, and therefore, the introduction of a new artifact can be avoided.
Owner:SHANGHAI UNITED IMAGING HEALTHCARE

Method and system for transmitting and receiving data signal of site/access point in wireless local area network

The invention discloses a method and a system for transmitting and receiving a data signal of a site / an access point in a wireless local area network, which can maintain the high throughput of a communication link in the wireless local area network in the situation that the dynamic range of the power of a radio signal is wide. The main scheme comprises the following steps: A, presetting a plurality of gains, which can ensure a receiver to work stably, for a digital AGC loop; B, selecting a corresponding preset gain and performing fine-tuning on the corresponding preset gain when necessary according to the intensity of the data signal received by the site / the access point currently; and C, setting the transmission power of a transmitter and performing the fine-tuning on the transmission power when necessary according to the intensity of the data signal received by the site / the access point currently. According to the method, a special MAC frame is added to verify the data transmitting and receiving performance, and a special method is adopted to estimate the intensity of the data signal transmitted by a channel and to calculate the data receiving ratio of the site / the access point which can serve as quantitative data for judging the fading of the channel of a wireless link, and a reference frame of the communication efficiency of a device.
Owner:BEIJING LHWT MICROELECTRONICS

Method for recovering data in distributed file system and electronic device

The invention provides a method for recovering data in a distributed file system and an electronic device. A main PG obtains a data recovery task; at least one first auxiliary PG having data recovery capability and a second auxiliary PG needing to recover data are determined from all auxiliary PGs corresponding to the main PG; a first target PG is determined from the main PG and the at least one first auxiliary PG; and the first target PG is controlled to recover a file copy corresponding to the second auxiliary PG. According to the invention, data recovery of the file copy can be carried out depending on the auxiliary PG, the file copy needing to be recovered can be recovered with no need of all depending on the main PG, and therefore the workload of the main PG is reduced, the IO processing speed of the main PG is not influenced, and the IP processing capability of the cluster is improved.
Owner:LENOVO (BEIJING) CO LTD

Switchboard system having port extending ability and method of realizing extending of port

Some dedicated chips have limited number of physical ports. This imposes restriction on the number of other devices to be connected to and the port number of the switch system, where these chips reside in. Via connecting the dedicated chips with the low-cost Ethernet exchange chips, the invention extends the port number on the dedicated chips without increasing the dedicated chips. In the net switch system, the dedicated chips possessing port-extending function offered by this invention include a port numbering process unit. The physical ports of dedicated chips are connected with the up-going port of the low-cost Ethernet exchange chips. This invention causes the port number supported by the net switch system to increase tenfold. Thus the utility of dedicated chips are raised and the cost of whole net switch system is lowered.
Owner:SUZHOU CENTEC COMM CO LTD

Log data recording method and apparatus and computer-readable storage medium

The present invention provides a log data recording method and apparatus and a computer-readable storage medium. The method comprises acquiring a plurality of log data to be recorded; recording a plurality of target log files for recording the log data to be recorded in a disk; storing the log data to be recorded into a target memory space corresponding to the target log files in a memory in parallel; and storing the log data in a disk in a storage space corresponding to the log data to be recorded; if the preset condition is satisfied, storing the log data stored in the target memory in the target log file corresponding to the target memory space in the disk. The log data recording method, the device and the computer-readable storage medium provided by the invention can record the generated log data when the log data is large, so as not to affect the processing speed of other services of the system.
Owner:NEUSOFT CORP

Touch equipment and one-hand operation method for same

The invention discloses touch equipment and a one-hand operation method for the same. The touch equipment comprises a touch screen and a control system. The touch equipment can run under the control of the control system, and the control system comprises a virtual interface calling module, a virtual interface closing module, a touch input / positioning module, a computing module and a graphic processing module. The touch equipment and the one-hand operation method have the advantages of easiness and convenience in operation, good service flexibility, zero influence on processing speed and difficulty in causing eye fatigue.
Owner:KUNSHAN GO VISIONOX OPTO ELECTRONICS CO LTD

Fast detecting method and device for static vehicle target

The invention discloses a fast detecting method and device for a static vehicle target. The fast detecting method includes the following steps that images are collected, wherein vehicle images are obtained; the images are preprocessed, wherein normalization is conducted on the sizes of the vehicle images, and the vehicle images are converted into a grey-scale map; entropy information is extracted, wherein pixel level traversal is conducted on the images through search boxes, and pulse iteration average entropy information of the partial images in the search boxes is extracted; rough positioning is conducted on the vehicle target, wherein the images with the pulse iteration average entropy value larger than that the target entropy value are taken as vehicle target rough positioning images; the vehicle target is determined, wherein the vehicle image with the largest pulse iteration average entropy value in the same picture area is reserved as a target vehicle search result; the vehicle target is presented in an original large image, and positioning of the vehicle target is completed. According to the fast detecting method and device for the static vehicle target, robustness, rapidness and transportability of a pulse coupling neural network feature extraction method are used, the feature of the pulse iteration average entropy is selected as the judgment basis, and the speed of detecting the static vehicle target is greatly increased.
Owner:WUHAN INSTITUTE OF TECHNOLOGY

Automatic play control method and device

InactiveCN106033306ADoes not affect processing speedMeet the needs of browsing content of interestInput/output processes for data processingProcess rateMultimedia
The embodiment of the present application discloses an automatic play control method and device. The method includes: detecting a first intervention signal in a first play area, and stopping automatic play in the first play area according to the first intervention signal. The intervention signal is generated by the user's operation, and the user only needs to perform simple operations to meet the needs of browsing the content of interest without affecting the continuous playback of the playback area. Moreover, the automatic playback control method provided by the present application has a simple process, consumes less resources, and will not affect the processing speed of the interface itself and the playback area.
Owner:ALIBABA GRP HLDG LTD

Multi-sampling judgment method for fault-tolerant filtering based on remainder check

InactiveCN102567130AThere will be no problem of missed detectionDoes not affect processing speedError detection/correctionBaseline dataComputer architecture
The invention provides a multi-sampling judgment method for fault-tolerant filtering based on remainder check, which solves the problem of residual error in a fault-tolerant filtering method based on remainder check in the prior art. When filtering output data of a first branch circuit are unequal to filtering output data of a second branch circuit but the filtering output data of the two branch circuits are the same by obtaining a remainder after division, and one error branch circuit cannot be judged once, the data can be temporarily stored in a cache, other data are continued to be processed and subjected to multiple judgment until one datum in the data by obtaining the remainder after division of the filtering output data of the first branch circuit and the filtering output data of the second branch circuit is not the same as a reference datum, and at the time, the branch circuit in error and the branch circuit with a correct processing result can be judged, so that the problem of residual error is avoided.
Owner:TSINGHUA UNIV

Sub-pixel-level distortion correction method and apparatus

The invention discloses a sub-pixel-level distortion correction method and apparatus. According to the method and the apparatus, display points on a display (4) are controlled to be turned on from a known point by adopting an industrial personal computer, and a camera (1) is controlled to perform image extraction and processing. By adopting the method and the apparatus, the distortion calibration precision can be effectively improved, the sub-pixel-level distortion correction effect can be achieved, the edge calibration effect can be greatly improved, and a visual product can be applied to a more accurate occasion. By applying the method and the apparatus to a projection device, the projection distortion can be remarkably reduced, the projection effect can be improved, and the optical cost can be reduced. An algorithm is simple, and the image processing speed is not influenced, so that the method and the apparatus are very suitable for the fields of high-precision machine vision, camera imaging distortion correction, projection distortion correction and the like.
Owner:大连鉴影光学科技有限公司

Prediction method and system for capacity of power battery in new-energy vehicle

The invention provides a prediction method and system for the capacity of a power battery in a new-energy vehicle, belonging to the field of power battery capacity of new-energy vehicles. The invention aims to solve the problem of how to improve the prediction efficiency of the capacity of the power battery in the new-energy vehicle. According to the invention, the charging condition data of multiple vehicles are collected through a big data platform; the charging condition data are screened according to a preset condition to respectively obtain a training target data set required by neural network training and a training input data set having high correlation with the power battery capacity; and neural network training is performed on the training target data set and the training input data set to establish a prediction model for the capacity of power batteries in the new-energy vehicles, the current charging condition data of a vehicle having a demand for power battery prediction capacity is acquired and substituted into the power battery capacity prediction model for the new-energy vehicles to calculate the predicted capacity of the desired vehicle power battery. According to the invention, the prediction model for the capacity of power batteries in the new-energy vehicles is established, and the capacity prediction efficiency of power batteries in the vehicles is improved.
Owner:ZHEJIANG GEELY HOLDING (GROUP) CO LTD +1

Authentication method for bank certificate printed with two-dimension code

The invention discloses an authentication method for a bank certificate printed with a two-dimension code. The surface of the bank certificate is printed with a bank certificate two-dimension code. The authentication method comprises the following steps of: 101. analyzing an element item contained in the bank certificate two-dimension code information; 102. selecting a detection area; 103. printing the two-dimension code and the detection area: printing the two-dimension code and the detection area in the vacancy of the bank certificate; 104: authenticating the bank certificate: comparing whether the corresponding element item information on the detection area and the bank certificate is consistent; if yes, proving that the two-dimension code information on the bank certificate is correct; and if no, proving that the two-dimension code information on the bank certificate does not conform to information in the bank certificate. With the authentication method, the element item in the two-dimension code information serves as the detection area to be printed on the certificate in a plaintext mode, and whether the element item in the two-dimension code and the element item of the certificate are consistent can be quickly judged so as to make up the deficiency that the bank certificate can not be directly authenticated.
Owner:SHENZHEN SINOINFO TECH DEV

Cutting chamfering machine

The invention relates to a cutting chamfering machine. The cutting chamfering machine is characterized by comprising a rack, a transmission mechanism, a splitting mechanism and a chamfering mechanism;and the transmission mechanism is arranged on the rack, the splitting mechanism and the chamfering mechanism are located on the transmission mechanism and are arranged on the rack in sequence from front to back, the splitting mechanism is formed by a front power splitting cutter disc and a back power splitting cutter disc, the chamfering mechanism comprises two power grinding discs which are arranged in sequence in the transmission direction of the transmission mechanism, the two power grinding discs are inclined to the transmission surface of the transmission mechanism, and the inclined directions of the two power grinding discs are opposite. Compared with the prior art, the cutting chamfering machine has the advantages that movement is not needed, ceramic tiles or stones can be chamfered on the same machine after being split, and cracking can be reduced and even prevented in the condition that the processing speed is not influenced.
Owner:FOSHAN YONGSHENGDA MACHINERY

Primary purifying device for wastewater treatment

The invention relates to the technical field of wastewater treatment and discloses a primary purifying device for wastewater treatment. The primary purifying device comprises a water feeding pipe, a filtering drum and a driving screw panel, wherein the filtering drum is rotatably arranged; the driving screw panel is arranged on an inner wall of the filtering drum; a plurality of water outlets areformed on peripheral walls of the filtering drum; the water feeding pipe comprises a water outlet; the water outlet is located in an inner chamber of the filtering drum. The primary purifying device for wastewater treatment is capable of guaranteeing the efficiency of wastewater treatment.
Owner:襄阳先创环保科技有限公司

PCB online detecting method

The invention discloses a PCB online detecting method which is used for an image acquiring working station. The PCB online detecting method comprises the following steps: acquiring current PCB high-resolution colorful images which pass through the image acquiring working station in real time, wherein the resolution is 1,270 ppi at least; layering the high-resolution colorful images according to different colors in the images, carrying out extreme lossy compression on the images in each layer, and carrying out original lossless compression on images among the layers; and transmitting a preprocessing finish signal to an integrated control working station so that the integrated control working station can transmit compression data processing signals to idle image processing working stations in real time, wherein the preprocessing finish signal is a signal showing that compression data of high-resolution colorful images have been obtained, and the idle image processing working stations are image processing working stations which do not carry out image processing tasks at present. The image acquiring working station is matched with the distributed image processing working stations, and conflict between slow big data processing and high online instantaneity requirement can be solved.
Owner:BEIJING LUSTER LIGHTTECH

Automatic focus searching method and system for pathological images under microscope

ActiveCN112363309AHigh sharpness valueLow sharpness valueMicroscopesRadiologyFocal position
The invention provides an automatic focus searching method for pathological images under a microscope. The automatic focus searching method comprises the steps: arranging equipment and an acquisitionobject; adjusting the acquisition position of an image; roughly positioning a pathological image; and carrying out fine focus searching on the pathological image to obtain a final focus position. According to the automatic focus searching method, a preliminary positioning process and a fine positioning process exist, the step-by-step cleaning process of focus searching is met, meanwhile, operationis conducted through a definition algorithm in the preliminary positioning process and the fine positioning process, and the accuracy of automatic focus searching is improved under the condition thatthe processing speed is not affected.
Owner:杭州医派智能科技有限公司

Method and device for superposition processing of image data or audio data

The invention provides a method and device for superposition processing of image data or audio data. The method is implemented by a superposition processor which is arranged on a master apparatus and connected with a bus. The device mainly comprises a main controller part and an operator part, wherein the main controller part is mainly used for reading the data to be superposed and storing in twoFIFOs (first in first out); the operator extracts and operates the data and then stores in another FIFO; and the main controller writes the data of FIFO-3 into memory. The method is simple and convenient to implement; the superposition processor is arranged on an SOC (system on chip) and operates independently without occupying system resources or influencing the processing speed of the system, and the power consumption is low.
Owner:FUZHOU ROCKCHIP SEMICON

Assembling equipment for assembling mobile phone shells and inserts and assembling method

The invention relates to the technical field of mobile phones, in particular to assembling equipment for assembling mobile phone shells and inserts. The assembling equipment comprises a working table, a conveying belt and a supporting frame, one side of the working table is fixedly provided with a plurality of support legs; the conveying belt is fixed to one side of the working table, the plurality of mobile phone shells are arranged on one side of the conveying belt, a blocking frame A is fixedly arranged on one side of the conveying belt, a blocking frame B is fixedly arranged on the side, away from the blocking frame A, of the conveying belt, and a single material pushing device is arranged on one side of the blocking frame B; and the supporting frame is fixed on one side of the blocking frame A. According to the assembling equipment for assembling the mobile phone shells and the inserts, through the arrangement of the blocking frame A, the blocking frame B, a J-shaped positioning frame, a material pushing air cylinder, an L-shaped material pushing frame, a stamping frame, a suction cup, a linear air cylinder, a double-shaft stamping air cylinder, an L-shaped lifting frame, an L-shaped sliding frame, a reciprocating air cylinder, an electronic pressure meter, a spring, an adjusting plate and an adjusting bolt, and the assembling equipment has the functions of automatic feeding, automatic material taking, automatic inlaying, efficient machining, universality, stress adjustment and the like.
Owner:深圳鑫宏力精密工业有限公司

Gasket mounting equipment and gasket mounting method for motor with reversing mechanism

The invention discloses gasket mounting equipment for a motor with a reversing mechanism. The gasket mounting equipment comprises a rotating jig, an oil applying device, an insulating gasket feeding device and an insulating gasket pressing-in device. Distance measuring sensors are arranged at the three insulating gasket mounting stations and correspond to the discharging end of the insulating gasket feeding device; and a second rotating assembly is arranged below the jig at the feeding station and the oil applying station, and a lifting assembly is arranged below the discharging station. According to the invention, the insulating gasket is integrally assembled, and the assembly efficiency is high; and the insulating gasket can be taken accurately and quickly and stably.
Owner:NINGBO YINZHOU TEERFEI ELECTRONICS CO LTD

Coal mining sewage treatment device

The invention discloses a coal mining sewage treatment device, and relates to the technical field of sewage treatment. The coal mining sewage treatment device comprises a bottom plate, wherein a filter and a storage device are fixed to the top of the bottom plate, and the filter is in a hollow cylinder shape. According to the coal mining sewage treatment device, sewage entering the filter can be filtered by arranging a rotary filter screen, so that coal fragments in the sewage stay on the surface of the rotary filter screen, the sewage can scour the surface of the rotary filter screen by guiding the flowing direction of the sewage through the water inlet pipe, coal fragments on the rotary filter screen can be tightly attached to the inner wall of the filter, when the rotary filter screen passes through a material leakage door plate, an anti-adhesion mechanism opens the material leakage door plate, so that the coal fragments filtered from the rotary filter screen automatically enter the storage device, only the coal fragments in the storage device need to be cleaned regularly subsequently, and the coal fragments on the filter screen do not need to be cleaned.
Owner:胡小林

Redis-based distributed lock repeat notification solving method

The invention provides a Redis-based distributed lock repeat notification solving method. The method comprises the following steps: S1, adopting a Redis set as a filter; s2, storing the message ID, judging whether the ID value of the message ID is an integer or not by Redis internal storage, and storing the message ID in a Redis set; s3, checking whether the messages are repeated or not; s4, informing a message queue. A Redis set is adopted as a filter, a message ID is stored in the Redis set, whether the message is repeated or not is judged by checking whether the message ID exists in the Redis set or not, a message queue is informed that the message is processed for the repeated message, the message does not need to be informed again, the effect that the message queue cannot repeatedly inform the same message is achieved, and therefore, the problem of repeated consumption of messages is solved.
Owner:银盛支付服务股份有限公司

A threshold processing method for signal detection template of electromagnetic spectrum monitoring receiver

The invention discloses a threshold value processing method for a signal detection template of an electromagnetic spectrum monitoring receiver, which belongs to the field of electromagnetic spectrum monitoring and the field of intelligent instruments. During the monitoring process, the template threshold is adjusted according to the actual situation, and a single spectrum template is derived into a template cluster containing many new templates. The present invention processes spectrum templates by means of morphological expansion, enabling users to adjust template thresholds according to actual conditions during the spectrum monitoring process, eliminating false alarms caused by signal offsets; the present invention derives a single spectrum template into a template containing many new templates The template cluster is easy to operate, and there are only two variables that the user needs to control; it improves the accuracy and flexibility of the test work, improves the practicability of the template threshold, and can effectively avoid false alarms caused by a small amount of distortion of the measured signal; The method has a low computational load and can be used in existing instruments without affecting the processing speed.
Owner:CHINA ELECTRONIS TECH INSTR CO LTD

Migration speed control method and device for distributed object storage system and medium

The invention discloses a hierarchical storage migration speed control method and device for a distributed object storage system and a computer readable storage medium. The method comprises the steps:calculating real-time time delay of a current period according to collected front-end user IO processing time delay data; calculating a matched time delay value based on the target set value of the front-end IO concurrent speed, and taking the time delay value as a reference time delay. According to a preset speed adjustment rule, a migration speed value of the cold and hot pool in the current period is calculated based on a numerical value relationship between the real-time time delay and the reference time delay; the maximum speed value of the migration data of each migration module is determined according to the total number of the migration modules in the cold and hot pool and the migration speed value; the migration speed of the cold and hot pool is controlled based on the maximum speed value of each migration module, so that the data migration speed of each level of storage area in the distributed object storage system can be effectively controlled, and the data migration does not influence the processing speed of a user request.
Owner:SUZHOU LANGCHAO INTELLIGENT TECH CO LTD

Multi-sampling judgment method for fault-tolerant filtering based on remainder check

InactiveCN102567130BThere will be no problem of missed detectionDoes not affect processing speedError detection/correctionBaseline dataComputer architecture
The invention provides a multi-sampling judgment method for fault-tolerant filtering based on remainder check, which solves the problem of residual error in a fault-tolerant filtering method based on remainder check in the prior art. When filtering output data of a first branch circuit are unequal to filtering output data of a second branch circuit but the filtering output data of the two branch circuits are the same by obtaining a remainder after division, and one error branch circuit cannot be judged once, the data can be temporarily stored in a cache, other data are continued to be processed and subjected to multiple judgment until one datum in the data by obtaining the remainder after division of the filtering output data of the first branch circuit and the filtering output data of the second branch circuit is not the same as a reference datum, and at the time, the branch circuit in error and the branch circuit with a correct processing result can be judged, so that the problem of residual error is avoided.
Owner:TSINGHUA UNIV

A lightning arrester inspection transportation line

The invention discloses a lightning arrester inspection transport line comprising a plurality of placing racks provided with road wheels at lower ends, a transportation track and a circulation track;two ends of the transportation track are respectively provided with a first and a second rotating disks; the placing rack moves on the transportation track and the circulation track; a transition rackis installed on a front end of the placing rack; a guiding board is installed on the transition rack; a guiding gear rack is installed on the guiding board; a slide pedestal moves along the guiding gear rack; the slide pedestal is provided with an inclined transition slot; a lower end of the inclined transition slot is fixed on the slide pedestal through a cylinder; an inspection transport rack is provided with a first connecting board and a second connecting board, wherein the inclination angle between the first connecting board and the inclined transition slot bottom and that between the second connecting board and the inclined transition slot bottom are the same; a conveyor belt is installed at the middle of the inspection transport rack; an electric sucker of the fixed placing rack isinstalled at a lower end of the inspection transport rack. The lightning arrester inspection transport line of the invention has simple structure and reasonable design; it can not only reduce the cost of production and the labor burden of workers, but also avoid the occurrence of missed inspection and the inflow of unqualified products into the market.
Owner:句容市大唐电气有限公司

An entrance and exit control system and method meeting evacuation requirements

The invention discloses an entrance and exit control system and method that meet evacuation requirements. The system includes a locking device for turning on and off the electric lock, a locking device for turning on and off the electric lock, an entrance and exit controller, a reading device and / or a request to leave device; the locking device for power-on lock and power-off is installed on the door leaf and the door frame, forming a locking point between the door leaf and the door frame; The mechanical lock cooperates to form another locking point of the door leaf and the door frame; the locking device of the power-on lock is connected to the lock power supply through the normally closed output contact of the relay of the entrance and exit controller; the locking device of the power-on and power-off electric lock The normally open output contact of the relay of the entrance / exit controller is connected to the power supply of the lock; the reading device and / or the device for requesting to leave is connected to the entrance / exit controller. The invention can improve the anti-technical opening ability of the entrance and exit, and can effectively improve the safety protection ability under the premise of meeting the requirements of safe evacuation.
Owner:THE FIRST RES INST OF MIN OF PUBLIC SECURITY

Method and system for transmitting and receiving data signal of site/access point in wireless local area network

The invention discloses a method and a system for transmitting and receiving a data signal of a site / an access point in a wireless local area network, which can maintain the high throughput of a communication link in the wireless local area network in the situation that the dynamic range of the power of a radio signal is wide. The main scheme comprises the following steps: A, presetting a plurality of gains, which can ensure a receiver to work stably, for a digital AGC loop; B, selecting a corresponding preset gain and performing fine-tuning on the corresponding preset gain when necessary according to the intensity of the data signal received by the site / the access point currently; and C, setting the transmission power of a transmitter and performing the fine-tuning on the transmission power when necessary according to the intensity of the data signal received by the site / the access point currently. According to the method, a special MAC frame is added to verify the data transmitting and receiving performance, and a special method is adopted to estimate the intensity of the data signal transmitted by a channel and to calculate the data receiving ratio of the site / the access point which can serve as quantitative data for judging the fading of the channel of a wireless link, and a reference frame of the communication efficiency of a device.
Owner:BEIJING LHWT MICROELECTRONICS

Method and device for rapid detection of static vehicle targets

The invention discloses a fast detecting method and device for a static vehicle target. The fast detecting method includes the following steps that images are collected, wherein vehicle images are obtained; the images are preprocessed, wherein normalization is conducted on the sizes of the vehicle images, and the vehicle images are converted into a grey-scale map; entropy information is extracted, wherein pixel level traversal is conducted on the images through search boxes, and pulse iteration average entropy information of the partial images in the search boxes is extracted; rough positioning is conducted on the vehicle target, wherein the images with the pulse iteration average entropy value larger than that the target entropy value are taken as vehicle target rough positioning images; the vehicle target is determined, wherein the vehicle image with the largest pulse iteration average entropy value in the same picture area is reserved as a target vehicle search result; the vehicle target is presented in an original large image, and positioning of the vehicle target is completed. According to the fast detecting method and device for the static vehicle target, robustness, rapidness and transportability of a pulse coupling neural network feature extraction method are used, the feature of the pulse iteration average entropy is selected as the judgment basis, and the speed of detecting the static vehicle target is greatly increased.
Owner:WUHAN INSTITUTE OF TECHNOLOGY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products