132results about How to "Meet real-time needs" patented technology

The invention discloses an image-based vehicle anti-collision early warning method. An image sensor is utilized for collecting the images of a rod ahead; after a vehicle is detected successfully and tracked, according to the vehicle width variation in interframe images and combining with the influences of the relative vehicle speed and relative acceleration, the time to collision (TTC) of a target vehicle and the vehicle is calculated, the dangerous level is analyzed, and an alarm is given to a driver, and meanwhile, the relative driving state of the target vehicle and the current vehicle also can be determined. The method provided by the invention is applied to the technical field of vehicle active safety, can effective prevent rear-end collision with the vehicle ahead, also is not influenced by rugged topography, and has high safety performance.

This invention provides one car pad extraction method based on wavelet exchange complex background, which comprises the following steps: adopting rough positioning an accurate positioning; the fine positioning adopts wavelet to extract pad area high frequency information for brightness and color regression process to get one integral background and number color composed of identification number pad.

The invention discloses a switchable peep-proof display device and a backlight, and relates to the field of display. Switchover between a narrow view angle and a wide view angle can be achieved, and cost is low. The switchable peep-proof display device comprises the backlight, an optical part with a collimation or center focus effect, a switchable diffusion sheet and a display panel which are sequentially connected; the backlight is used for providing back light for the display panel, the switchable diffusion sheet can be switched between the transparent state and the scattering state, and the switchable diffusion sheet does not change the propagation direction of the back light when in the transparent state and scatters the back light when in the scattering state to obtain the wide-view-angle back light.

The invention discloses a diagnosis system and method for the vessel condition. The diagnosis system comprises a vessel-end data monitoring system and a shore-end cloud diagnosis system, wherein the vessel-end data monitoring system comprises an engine room data collection base station, an industrial personal computer, a vessel-end server and a vessel-end multi-information fusion and analysis module; the shore-end cloud diagnosis system comprises a vessel data server cluster, a data algorithm server and a user client. The processing of data is divided into two levels, wherein in the first level, real-time data are analyzed conventionally, the calculated amount is small, and the real-time performance is high; in the second level, vessel-end data are sent to the shore-end cloud diagnosis system through the vessel and shore communication, complex analysis and condition trend forecast are carried out, the analysis is more accurate, the forecast function is achieved, but the real-time performance is poor. Through the double-level data analysis mode, the requirement for real time performance can be met, and complex and deep operation of mass data is achieved. The calculation result can be used for guiding seamen to operate vessels.

The invention discloses a product database forming method based on Internet data and a system. The method includes the steps of capturing webpage data with the theme relevance higher than a preset threshold value by the adoption of the focused crawler technology, performing structuralized storage on the captured webpage data, automatically classifying the structuralized storage webpage data according to the categories which products belong to, performing statistics on the frequency and the time of occurrence of attributes of the products in the webpage data after the automatic classification, performing weighting calculation on the frequency and the time of occurrence of the attributes of the products according to preset weighting, acquiring the decision value of the attributes of the products, and determining the sort order of the attributes of the products according to the decision value of the attributes of the products. The system comprises a data capturing module, a structuralized storage module, a data classifying module and an attribute deciding module. According to the product database forming method based on the Internet data and the system, a user can acquire comprehensive and summarized information without needing to collect and sort product information on the Internet, real-time performance of data is ensured, and real-time requirements of the user are met.

The invention provides a speech and semantic recognition method of a service robot. The method comprises: system word library modeling is carried out; to be specific, a word library database is established and then a semantic frame database is established; a lip language mode base is established; a system uses a speech recognition system to collect a speech instruction inputted by a user and a face video and speech and semantic recognition and video segment lip language recognition are carried out respectively; and according to a combined recognition result, a determined Chinese semantic information is displayed at a display interface. Therefore, a noun database and a semantic frame database are established for different service fields; and on the basis of the word library database, the semantic frame database, and the lip language mode base, the robot understands the Chinese meaning expressed by the speech instruction, so that speech and semantic recognition becomes accurate.

The invention relates to a license plate recognition method based on a trusted area. K-means++ clustering is carried out to select the number and scales of initial license plate candidate frames; theobtained number and scales of initial candidate frames are combined to a YOLO-L model, so that the positioning accuracy of the vehicle area and license plate area is improved. With the YOLO-L model, avehicle area and a license plate area are located and coordinates of an upper left corner and a lower right corner are outputted; vehicles and other objects on road are distinguished by the YOLO-L model; whether the license plate area is in the vehicle area is determined based on a license plate recognition algorithm, thereby eliminating false identification of the license plate area; and if thelicense plate area is located in the vehicle area, the license plate area is detected correctly to complete license plate identification. According to the license plate recognition method provided bythe invention, the license plate and the similar objects are distinguished well; the license plate is located effectively; the misjudgment of the license plate is reduced; the misjudgment of similar objects is reduced substantially; the cense plate recognition efficiency is improved; and the accuracy of license plate recognition is enhanced.

The invention relates to an iris positioning method, based on combined math altitude technique and probability stat technique, wherein said method comprises: first, binary valuing iris image, using projection to obtain the iris inner round position center; using altitude technique to extract edge to obtain the inner edge of iris inner round; using the distance between positioned round center and each point on edge as one sample, to transform search radius and radius into sample for searching minimum average value, whose average value is the round radius; positioning inner round, and using binary valuing method and math altitude method to obtain the edge of outer round; using the center of inner round to position center of outer round; and positioning outer round as positioning inner round. The invention has low error, high accuracy and high speed.

The invention relates to a robot image positioning method based on deep learning. The method at least comprises the following steps that A) deep training and learning are carried out to obtain an image model and a matching rule; B) a robot collects environment images in different orientations gradually, feature extraction and matching are carried out on the collected images, according to the obtained image model and matching rule, to obtain 3D coordinates of all matching points, and common key point information of at least two images is obtained; C) mismatching points are rejected from the matching points of the images via a random sampling consistency algorithm, and 3D coordinates of at least two key points are obtained from the images; D) the 3D coordinates of the key points are obtained by using a bubble sorting method; and E) a rotation matrix and an offset vector in every two orientations are obtained via an absolute orientation algorithm, and a moving track of the robot is positioned via iteration successively. The problem of positioning failure caused by that the robot cannot capture feature information indoor is solved. The invention also provides a deep learning based robot image positioning system using the robot image positioning method.

The invention provides a resource sharing method, a resource sharing device and a terminal. The resource sharing method comprises steps of determining position information of a demand side according to a received resource sharing request of the demand side, selecting a target supplier according to the position information of the demand side and resource information of a plurality of suppliers in a preset resource sharing database, and sending the position information of the target supplier to the demand side to enable the demand side to share resource with the target supplier according to the position information of the target supplier. According to the technical scheme, sharable resource can be quickly found for users having a demand of resource sharing in a simple and effective way, so user real-time demand can be met and user outdoor activities and life can be facilitated.

The invention relates to the technical field of image processing and discloses a method and a device for realizing conversion from a two-dimensional image to a three-dimensional image. The method comprises the following steps of: inputting the original two-dimensional image; determining interesting areas in the original two-dimensional image for a user; respectively using the interesting areas inthe original two-dimensional image for the user as prospect areas, using non-interesting areas as background areas and generating a prospect histogram and a background histogram; calculating color difference distribution of the prospect histogram and the background histogram; generating a depth map according to the color difference distribution; generating a target two-dimensional image accordingto the depth map; and generating a binocular three-dimensional image by fusion of the original two-dimensional image and the target two-dimensional image. According to the embodiment of the invention, the scheme implementation is simple, and the real-time requirement can be met.

The invention discloses a car networking real-time data transmission system and method. The system comprises vehicle navigation equipment, a memory card, a push-to-talk terminal, a mobile phone and a service background, wherein the vehicle navigation equipment is provided with a central processing unit for processing data; the memory card is connected with the vehicle navigation equipment through a universal serial bus (USB) interface, and a USB drive program, a wireless fidelity (WIFI) drive program and a service entry program are stored in the memory card; the push-to-talk terminal is connected with the vehicle navigation equipment through the USB interface and is provided with a WIFI module; the mobile phone is provided with a 2-generation (G) or 3G module and a mobile phone WIFI module and in wireless communication connection with the push-to-talk terminal; and the service background is in communication connection with the mobile phone through a 2G or 3G network. The system has high data transmission timeliness and reliability, is convenient to use and satisfies the requirements of users on efficiently obtaining information in real time.

The embodiment of the invention discloses a control method for a marine hybrid power unit, a power unit and a power system. The hybrid power unit comprises a lithium battery and a fuel cell. The control method comprises the steps that a voltage converter of the fuel cell receives a demand power command; if the demand power does not exceed the rated power of the fuel cell, the fuel cell is controlled to output power satisfying the load demand; and a voltage converter of the lithium battery determines the fluctuation power according to the change value of the bus voltage and controls the lithium battery to output or absorb the fluctuation power. According to the control method for the marine hybrid power unit, the power unit and the power system, the fuel cell and the lithium battery are combined to form the hybrid power unit, and the fuel cell and the lithium battery are controlled through the voltage converters, so that the fuel cell outputs the power meeting the load demand, and the lithium battery is controlled to switch the working mode according to the change of the bus voltage so as to output the difference power or absorb the surplus power, the bus voltage is kept within the rated voltage range, and therefore the power meeting the requirements of a ship is output in real time.

The invention provides an embedded distributed networking method which comprises the following steps: a step of forming a network topology; a step of forming an initial route; a step of real-time maintaining the route; and a step of transmitting the route. The invention also provides an embedded distributed networking system. According to the invention, the technical solution adopts four steps, which can substantially increase efficiency of copying and reading an ammeter.

The invention provides a video synthesis method which is used for electronic equipment provided with a display unit and at least two image acquisition units. The video synthesis method includes that images acquired by the at least two image acquisition units are received, and an image frame of a first video is produced by the utilization of the images; at least a part of displayed images of the display unit is obtained, an image frame of a second video is produced, the image frame of the first video and the image frame of the second video are synthesized to be as an image frame of a third video; and the steps are repeated sequentially until the image acquisition units stop image acquisition. The invention further provides a corresponding device and electronic equipment of the method, the corresponding device and the electronic equipment of the method comprise a generating unit of the image frame of the first video, a generating unit of the image frame of the second video and a video synthesis unit. The video synthesis method, the device and the electronic equipment achieve the purpose that videos are synthesized with existing videos while the videos are collected, and therefore video synthesis real-time requirements are met.

The invention discloses an automatic snapshot method and system for violation behaviors of a bus in front of a bus based on image recognition. The method mainly comprises the following steps: step 1,acquiring a real-time image in front of a bus shot by the bus; step 2, carrying out lane line detection and vehicle detection on the real-time image; step 3, judging violation behaviors such as lane occupation, line pressing and lane changing; and step 4, identifying a license plate. Under the condition of not depending on background data and manual intervention, whether violation behaviors such as bus lane occupation, solid line lane change and solid line riding and touching exist in a front vehicle or not is directly judged at a mobile terminal in real time only according to an image shot bya vehicle-mounted camera, and the violation behaviors are recorded and/or reported.

The invention relates to a control method and device of a vehicle-mounted hydrogen system and the vehicle-mounted hydrogen system, and belongs to the field of fuel cell buses. The method comprises: establishing a pressure prediction model of an output gas of the vehicle-mounted hydrogen system, and determining a pressure prediction value used for controlling the pressure of the output gas of the vehicle-mounted hydrogen system according to the pressure prediction model; calculating a difference value between the pressure prediction value and a set target pressure value to obtain a prediction error; obtaining an actual pressure value of output gas of the vehicle-mounted hydrogen system, and calculating a difference value between the actual pressure value and the target pressure value to obtain an actual error; and comparing the prediction error with the actual error, and when the prediction error is smaller than the actual error, outputting a pressure prediction value of the next momentby using the pressure prediction model as a target pressure of the next moment to control the vehicle-mounted hydrogen system to output gas. The change trend of the output pressure of the vehicle-mounted hydrogen system is judged in advance, and the pressure of the output gas of the vehicle-mounted hydrogen system is adjusted in advance, so that the supplied hydrogen can meet the requirements ofa fuel cell system.

The present invention relates to visual navigation system and method for intelligent endoscope. The system includes one endoscope, one lens deflecting mechanism, one automatic intervening mechanism, and one PC with image acquiring controller connected to the above said parts for processing image signal, displaying, and controlling lens deflection and automatic intervening. The system has simple composition, several manual and automatic switchable functions and convenient use; and the operation is fast, precise and timely.

The invention provides an encoding and decoding method which comprises the following steps that: a user interface module receives original data from the outside and defines a data packaging format required by a user according to the input of the user; a separation module separates the original data into original audio data and original video data; a plurality of conversion modules convert the original video data into video data required by the user; a processing module adds synchronization parameters of the original audio data corresponding to the video data required by the user to a head part of the video data required by the user; and a packaging module packages the video data processed by the processing module and the original audio data into data required by the user according to the data packaging format required by the user. By parallel transcoding, the data conversion rate is greatly improved and time required for transcoding is shortened so as to meet the real-time requirement of the user; and moreover, the synchronous processing carried out before the audio and video data is packaged implements the effect that in the playing process of the transcoded data, the audio and video synchronization is kept.

The invention discloses an internet voting method and device, and belongs to the field of internet voting. The method comprises the following steps: determining an ongoing voting activity at present according to a summary voting configuration statement; searching records corresponding to voting activity IDs of the ongoing voting activity in a voting activity and option mapping table, and loading the records into a memory; receiving voting information transmitted by a voter, and analyzing tetrad records or a set of tetrad records from the voting information; judging whether the voting activity IDs and voting option IDs in the tetrad records or the set of tetrad records exist in the memory or not; and regularly writing a voting option vote table in the memory into a database. The records corresponding to the voting activity IDs are loaded to the memory, so that the real-time writing operation stress of the database can be effectively alleviated by virtue of a memory caching method, the system bottle neck caused by the database also can be avoided, and the real-time requirement on simultaneously carrying out large-scale internet voting activities with multiple sessions and multiple types can be met.

The invention discloses a top view-based parking lot vehicle self-positioning and map construction method. The method comprises the steps of: taking a look-around system which is composed of a low-cost fisheye camera as an information collection system, generating a top view based on the look-around system, extracting the angular line features of parking lines nearby a vehicle in a robust manner,and carrying out tracking and map construction; and generating an angular line map in real time by utilizing a map matching technology, and carrying out real-time high-precision positioning and map construction on the vehicle by utilizing local map optimization and global map optimization. According to the method, the problem of scale drift of a monocular SLAM is avoided in principle, and real-time and high-precision vehicle self-positioning and map construction work is completed on a low-power-consumption vehicle-mounted processor through a low-cost sensor by means of an existing vehicle-mounted system so as to assist in completing an autonomous parking task. The invention further provides a construction device, a construction system, an automatic driving vehicle and an autonomous parkingsystem.