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

651results about How to "Accurate acquisition" patented technology

Method of quickly segmenting moving target in non-restrictive scene based on full convolution network

ActiveCN106296728AOvercoming the disadvantages of incomplete target segmentationUnlimited sizeImage enhancementImage analysisGround truthSample image
The invention relates to a method of quickly segmenting a moving target in a non-restrictive scene based on a full convolution network, which belongs to the technical field of video object segmentation. The method comprises steps: firstly, framing is carried out on the video, and a result after framing is used for making a Ground Truth set S for a sample image; a full convolution neural network trained through a PASCAL VOC standard library is adopted to predict a target in each frame of the video, a deep feature estimator for an image foreground target is acquired, target maximum intra-class likelihood mapping information in all frames is obtained hereby, and initial prediction on the foreground and the background in the video frames is realized; and then, through a Markov random field, deep feature estimators for the foreground and the background are refined, and thus, segmentation on the video foreground moving target in the non-restrictive scene video can be realized. The information of the moving target can be effectively acquired, high-efficiency and accurate segmentation on the moving target can be realized, and the analysis precision of the video foreground-background information is improved.
Owner:KUNMING UNIV OF SCI & TECH

Wireless sensor network-based greenhouse group control system and control method thereof

The invention discloses a wireless sensor network-based greenhouse group control system and a control method thereof. Each greenhouse is provided with a plurality of wireless sensor nodes and a cluster head node for acquiring environmental factors of the greenhouses according to needs, wherein the wireless sensor nodes and the cluster head nodes autonomously form wireless sensor networks by wireless communication; the wireless sensor nodes transmit monitoring data to the corresponding cluster head nodes in a multi-hop way; the cluster head nodes transmit the data to a gateway node which uploads the data to a communication server; the communication server performs two-way communication with a control computer; and the control computer makes corresponding guidance schemes according to the data monitored by the wireless sensor networks in real time, and synchronously transmits control instructions to the wireless sensor networks according to the guidance schemes. In the system and the method, the real-time data monitored by each sensor node in the greenhouses can be timely and accurately acquired, and the control instructions can be fast transmitted to the sensor nodes after specific production guidance is made or early-warning setting is performed according to the monitoring data.
Owner:NORTHWESTERN POLYTECHNICAL UNIV

Unmanned aerial vehicle depth image acquisition method and unmanned aerial vehicle

The invention provides an unmanned aerial vehicle depth image acquisition method and an unmanned aerial vehicle. The method comprises the steps that an onboard camera is used to acquire the image sequence of a predetermined scene, wherein the Nth image and the N+1th image in the image sequence have overlapping areas and the area ratio of the area of each overlapping area to the area of the Nth image and the N+1th image is higher than a preset ratio; an optical flow method based on feature matching is used to acquire the change of the location information of each pixel of the Nth image in each overlapping area in the N+1th image, and accordingly the pixel movement speed of each pixel of the unmanned aerial vehicle in each overlapping area in a camera coordinate system is acquired; the actual flight speed of the unmanned aerial vehicle in a world coordinate system is acquired; and according to the pixel movement speed, the actual flight speed and the parameters of the onboard camera, the depth image of each overlapping area is acquired, and accordingly integrating is carried out to acquire the depth image of a predetermined scene. The method provided by the invention can be used to accurately acquire the depth images, and has the advantages of wide application scope, low cost and easy realization.
Owner:GUANGZHOU XAIRCRAFT TECH CO LTD

Unconventional oil and gas reservoir horizontal well section three-dimensional rock mass mechanics modeling method and device

InactiveCN103258091ASolve the problem that interwell information cannot be obtainedEasy accessSpecial data processing applicationsThree-dimensional spaceMechanical models
The invention provides an unconventional oil and gas reservoir horizontal well section three-dimensional rock mass mechanical modeling method and device. The method includes the steps of determining an unconventional oil and gas reservoir horizontal well section object region; obtaining three-dimensional earthquake prestack gathered data in the object region; conducting elastic parametric inversion by adopting the three-dimensional earthquake prestack gathered data and obtaining a three-dimensional space elastic parameter data body in the object region, wherein the three-dimensional space elastic parameter data body comprises a rock mass poisson ratio and Young modulus; calculating a rock mass brittleness index according to the rock mass poisson ratio and the Young modulus; establishing a rock mass mechanical parameter model based on three-dimensional grid nodes according to the three-dimensional space elastic parameter data body and the rock mass brittleness index. The unconventional oil and gas reservoir horizontal well section three-dimensional rock mass mechanical modeling method and device can simply, effectively and accurately obtain an unconventional oil and gas reservoir rock mass mechanical model.
Owner:PEKING UNIV

Steel grid construction technology based on BIM lofting and three-dimensional scanning

The invention relates to the technical field of building construction, in particular to a steel grid construction technology based on BIM lofting and three-dimensional scanning. The steel grid construction technology includes the following steps: (1) determining a BIM model of a steel grid; (2) establishing a point coordinate database; (3) importing a BIM lofting robot; (4) carrying out on-site positioning measurement by the BIM lofting robot; (5) installing the steel grid; (6) carrying out re-measurement and examination by a three-dimensional scanner; and (7) carrying out adjustment and fixation. The steel grid construction technology starts from the basic BIM. Then, automatic lofting is carried out on a construction site. Finally, the actual construction structure information obtained bythe three-dimensional scanner through on-site surveying and mapping is compared with the data aggregated by the model. The compatibility of hardware and software, the analysis of lofting accuracy through actual measurement, the multi-professional data sharing and the improvement of the measurement lofting efficiency are systematically combined. The steel grid construction technology has high construction accuracy and short construction period, improves the efficiency of measurement, saves cost, improves the economic benefit, and is worth popularizing and applying widely.
Owner:QINGDAO NO 1 CONSTR GRP

System for underground communication and staff monitoring based on TD-LTE technology

The invention relates to a system for underground communication and staff monitoring based on a TD-LTE technology. The system comprises mobile terminals, a relay station, a communication base station, a location server, a switchboard, a core network device, host computers, and transmission lines. The mobile terminals and underground base station nodes and the location server are connected with each other through wireless interfaces, and trinity communication monitoring environment is formed. The switchboard gathers a plurality of underground base station nodes through optical fibers, and connects the gathered underground base station nodes with a ground base station. The ground base station accesses a public switched telephone network through the core network device. Main transmission channels of the system uses wired optical fibers. Wireless channel uses an orthogonal frequency-division multiple access technology and a multiple-input multiple-output modulation technology. An original underground monitoring positioning algorithm is improved by using a mixed measurement mode of signal transmits-receive time measurement and angular measurement. Information of underground staff positions is obtained by using a circle coordinate setting method. The improved system and algorithm effectively solves problems of difficulties of high quality communication and monitoring positioning under complex and harsh conditions in mines.
Owner:CHINA UNIV OF MINING & TECH (BEIJING)

Complex part pose estimation system and method based on three-dimensional measuring point cloud

The invention belongs to the field of automatic measurement and specifically discloses a complex part pose estimation system and method based on three-dimensional measuring point cloud. The method forestimating a complex part pose comprises a step of calibrating a conversion matrix between a six-degree-of-freedom industrial robot end flange and a grating area array scanner measurement coordinatesystem, a step of driving a grating area array scanner to move through the six-degree-of-freedom industrial robot to scan a measured workpiece and obtaining three-dimensional point cloud data of the measured workpiece, a step of converting the three-dimensional point cloud data of the measured workpiece to a robot base coordinate system to obtain converted three-dimensional point cloud data, and astep of matching the converted three-dimensional point cloud data with a three-dimensional design model of the measured workpiece to obtain the pose of the measured workpiece relative to the robot soas to complete the estimation of the pose of a complex part. The invention has a wide measurement range, the accurate measurement of multiple regions and complete topography of the measured workpiececan be achieved, and the pose of the measured workpiece can be accurately obtained.
Owner:HUAZHONG UNIV OF SCI & TECH

Sight tracking and fatigue early warning method

The invention discloses a sight tracking and fatigue early warning method, and belongs to the technical field of intelligent traffic. The method comprises the step of obtaining the information of the driver face through a mini-type camera, the step of controlling starting and stopping of a near axial infrared controllable light source and a far axis infrared controllable light source, the step of starting the near axial infrared controllable light source at the odd number frame, the step of starting the far axis infrared controllable light source at the even number frame, and the step of determining the position of the human eye pupils in a video sequence after the difference calculation according to the human eye pupil reflection characteristics. After the pupil positioning through the image difference frequency is carried out based on an infrared effect under infrared light, the sight projection position is reversely calculated according to the pupil center position and the feature recognition points of a vehicle-mounted displayer, and fine adjustment or fine calibration can be carried out according to the user individual difference in the calculation process. Compared with other control methods, the time for obtaining the eye characteristics of a driver can be effectively shortened, the interference caused by the ambient light change is avoided, and the sight information of the driver can be rapidly and accurately obtained.
Owner:重庆大数据研究院有限公司 +3

Walking state identification method based on space vector between human body skeleton joints

The invention provides a walking state identification method based on a space vector between human body skeleton joints. The method comprises a step of identifying the joint three-dimensional coordinates of a human body skeleton after a human body is detected by using a Kinect camera, and a step of identifying a human body walking state according to the space vector formed between the joints of the human body skeleton. The step of identifying a human body walking state concretely comprises a step of extracting the three-dimensional coordinates of a left hipbone, a left knee, a right hipbone, a right knee, a left shoulder bone and a right shoulder bone needed by identification, a step of calculating the cosine value of the angle between the space vector formed by the left hipbone and the left knee and a vertical direction and a cosine value of the angle between the space vector formed by the right hipbone and the right knee and the vertical direction, a step of judging whether a target person is in a walking state through the cosine values of the two angles, and judging whether the target person in a walking state walks straight, turns left or turn right according to the angle cosine value of the space vector formed by the left shoulder bone and the right shoulder bone and a Kinect camera positive direction. According to the method, the three-dimensional coordinate data of the human body skeleton joints can be accurately obtained, and the attitude action of each frame of the target person is rapidly obtained.
Owner:NORTHEASTERN UNIV
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
Try Eureka
PatSnap group products