209results about How to "Make up for the error" patented technology

The invention provides a non-contact free space eye-gaze tracking method suitable for man-machine interaction. The method comprises the steps of positioning and tracking a face and eyes in real time, extracting eye movement biological characteristic information, building an eye movement model based on the eye movement biological characteristic information, building a mapping relational model of the eye movement model and an eye-gazed object and the like. The non-contact free space eye-gaze tracking method suitable for the man-machine interaction relates to multiple crossed fields of image processing, computer vision, pattern recognition and the like and has wide application prospect in the fields of new-generation man-machine interaction, disabled people assisting, aerospace relating field, sports, automobile and airplane driving, virtual reality, games and the like. In addition, the method has great practical significance for improvement of the life and self-care level of disabled people, building of a harmonious society and improvement of independent innovative capability in the national high-and-new technical fields of the man-machine interaction, unmanned driving and the like.

The invention discloses a system and a method for collaboratively navigating, investigating and monitoring an unmanned aerial vehicle and an intelligent vehicle. The system comprises an unmanned aerial vehicle unit, a PC (personal computer) and an intelligent vehicle unit, wherein the PC controls the unmanned aerial vehicle unit to transport the intelligent vehicle unit which is adsorbed to the bottom of the unmanned aerial vehicle unit to a designated area; the unmanned aerial vehicle flies and monitors in the designated area, and acquired a ground image of the monitored area at fixed frequency; the PC splices a plurality of acquired ground images of the monitored area to obtain the picture of the whole monitored area, and converts the picture into a binary picture; a ground grid and a barrier are identified to calculate positions of the ground grid and the barrier in a two-dimensional coordinate graph of the picture of the monitored area; optimal travelling path information is calculated by utilizing the positions of the ground grid and the barrier in the two-dimensional coordinate graph of the picture of the monitored area, and transmitted to the intelligent vehicle unit for being stored; the intelligent vehicle unit performs tracking travelling according to the optimal travelling path, and performs ultrasonic automatic obstacle avoidance to travel to a designated place.

The invention relates to a calibration method for representing dense sandstone pore size distribution by adopting nuclear magnetic resonance; the method comprises the steps of measuring the rock porosity and the skeleton density; carrying out a low-temperature broken sample N2 absorption and desorption experiment, and carrying out a nuclear magnetic resonance test and a high pressure Hg injection test on a regular column sample in a saturated formation water state; converting pore volume obtained through N2 absorption into porosity components under different pore throat radiuses by using a capillary accumulated Hg inlet curve, and establishing a pore proportion summation curve by synthesizing low temperature N2-high pressure Hg injection; determining a best calibration coefficient by comparing the nuclear magnetic resonance pore proportion summation curves under the different C values to realize the conversion between nuclear magnetic resonance T2 spectrum and the pore throat radius; establishing a relation between parameters of the calibration coefficient C and a reflection pore-throat structure according to lithological characters; carrying out nuclear magnetic resonance conversion on dense sandstone reservoirs which are positioned in the same region and have the same lithological character. According to the method, the nuclear magnetic resonance T2 spectrum is calibrated by combining low temperature N2 absorption and a high pressure Hg injection experiment, and large pores, micropores and mesopores are calibrated, so that the distribution of different grades of pore sizes in dense sandstone can be effectively represented.

The present invention provides an image motion compensation method for a space optical remote sensor, relating to an image motion compensation method and solving the problem that the current image motion velocity vector calculation method used in image motion compensation cannot satisfy complex imaging tasks and the calculation result is not accurate. The process of the image motion compensation method is as follows: creating five coordinate systems starting from ground targets to image points, transferring the coordinates for a plurality of times according to the rotation and translation principles among vectors, describing the positions of the targets and the image points in the camera coordinate systems, obtaining the image motion velocity vector calculation formula and the calculation method when the satellite forms images of the targets, then obtaining the image motion velocity vector used in the image motion compensation of the space optical remote sensor. The image motion compensation method overcomes the deficiency in the prior art, dependently calculates the target positions needed in the image motion velocity vector calculation and integrally considers the orbit and attitude parameters, besides. The image motion compensation method is suitable for complex imaging tasks and general satellite orbit, and is used in the fields such as atmospheric environment monitoring, resources exploration, disaster prevention and the like.

The invention discloses an automatic feeding and discharging device, which comprises a distribution mechanism, a dropping mechanism and a conveyance mechanism. The conveyance mechanism is arranged between the distribution mechanism and the dropping mechanism and consists of a base, a feeding mechanism, a lifting mechanism and a conveying mechanism are disposed on the base, the lifting mechanism is arranged between the feeding mechanism and the conveying mechanism, and the feeding mechanism is connected with the distribution mechanism, and is also connected with the dropping mechanism. By the aid of the joint action of the distribution mechanism, the feeding mechanism, the lifting mechanism, the conveying mechanism and the dropping mechanism, automatic feeding and discharging of pipes can be realized, labor intensity of feeding and discharging of the pipes can be greatly reduced, production efficiency is enhanced, and production cost is lowered.

The invention discloses a method for reducing distributed machine learning communication overhead, which is based on a parameter server architecture, is suitable for multi-machine cluster distributed machine learning of a data center, and is also suitable for end cloud collaborative distributed machine learning with a server as a cloud end, a mobile phone or an embedded device as a terminal. The method comprises the following steps: firstly, calculating gradients of all working nodes, solving a global momentum by combining two rounds of parameter difference, summing the global momentum and a previous round of memory gradient to obtain a new round of memory gradient, sending a part of the new round of memory gradient to a server node, and accumulating the rest part; then enabling the server node to accumulate all sparse memory gradient sums so as to update parameters and broadcast the parameter difference of two rounds to all working nodes; and finally, enab;omh the working node to receive the two rounds of parameter difference and update the parameters. The method is based on global gradient compression, and only a part of global momentum is transmitted when the working node communicates with the server node, so that the communication overhead in distributed machine learning is reduced.

The invention relates to a flammable liquid planar free-flowing combustion simulation experimental device and method. The device comprises an oil supply system, a rogue system, an overflow system anda test system. The flammable liquid is pre-placed in an oil drum, and is flown into a flow surface tray through a peristaltic pump, a silicone hose and a steel pipe buffer tub for ignition and burning, a thermocouple tree of the test system, a camera, a bonded thermocouple, a radiant heat flow meter and an infrared thermal imager are employed to perform experiment data and image collection, and are connected with a multi-channel data collection instrument for remote data transmission, real-time display and control so as to achieve measurement of parameters such as flow burning areas, temperatures, heat radiation and the flame heights, and through change of the leakage parameters, material cushion and flow surface angles, the flammable liquid planar free-flowing combustion simulation experimental device and method are high in experiment efficiency, safe and controllable, simple to operate, convenient to use and accurate in measurement, and can develop planar free-flowing combustion state research of different substances, different flow surface types and different flow angles.

The invention provides a method and a device for searching similar images based on perceptual hash. The device comprises an input unit, a difference unit, a searching unit, a screenshot unit, a perceptual hash algorithm unit, a comparison unit, and an output unit. The device has beneficial effects that after an input original image and a screenshot image of the original image are processed by the perceptual hash algorithm, the input original image and the screenshot image are respectively compared with images in a search library, so that under the condition that the input image is the screenshot image of a certain image, the source of the original image can be found. The device and the method improve accurate rate of the perceptual hash algorithm, and provide a new thinking for searching similar images.

The invention relates to a real-time sensor signal network transmission method based on linear predictive filtering. The technical proposal is that: a sending end only sends a sampling sequence needed by signal rebuilding, the received sampling sequence is recovered on a receiving end by adopting a linear predictive filtering method, and a source signal is rebuilt by adopting the sequence, thereby building a virtual local sensor of the signal, and a user can acquire a signal sampling estimation value by sampling the virtual local sensor. The method comprises the following steps: a, clock synchronization of the sending end and the receiving end, namely carrying out the clock synchronization of the sending end and the receiving end by adopting average round-trip time of multiple synchronous frames; b, sending optimization of the sampling sequence by the sending end; c, recovery processing of the sampling sequence by the receiving end; and d, signal rebuilding on the receiving end. The method has the advantages of optimizing sending data sequence, reducing network flow needed by data transmission, and ensuring that signal data use and network transmission of signals are separated and a user does not need to directly concern signal network transmission course.

The present invention provides a data processing method and device which are applied to a shopping platform. The method comprises the steps of: obtaining first commodity information; according to thefirst commodity information, obtaining a shop information set comprising a first commodity; obtaining the position information of a user; according to the position information of the user, obtaining the first shop information in a first pre-determined distance; determining whether it is needed to perform offline experience for the first commodity or not; if it is needed to perform offline experience for the first commodity, performing distribution of the first commodity through the first shop; and after the user performs experience of the first commodity, obtaining purchasing information. Thedata processing method and device solve the technical problems that network shopping cannot perform actual experience, is poor in purchasing effect and influences the users' satisfaction in the priorart, achieve combination of network shopping and offline actual experience, improve the users' shopping satisfaction, make up errors because commodities cannot be actually checked through network shopping and ensure the user shopping experience.

The invention discloses an initiative mild grinding and polishing device and a method based on a viscous-elastic material, and mainly solves such problems as nonuniform removing quantity, imprecious positioning of a polishing head, mismatching between the shape of a polishing tool and the surface of a workpiece and generation of errors by vibration of the workpiece in the polishing process. The device consists of such several parts as an ultrasonic generation part, a laser displacement sensor detecting part, a viscous-elastic material polishing part and a worktable. The device machines an optical free curved surface, needs no frequent adjustment of the cutter position for the surface of a part with a more complex curved surface, adjusts the force application frequency to change the shape of the viscous-elastic polishing head so as to always keep a gapless bonding state between the viscous-elastic grinding and polishing head and the surface of the workpiece in the machining process, and realizes initiative mild grinding and polishing of the workpiece.

The invention discloses a mountain land surface temperature remote sensing inversion method, which comprises the following steps of 1, downloading an on-board radiance product of a TIRS sensor carriedon a Landsat 8 satellite and a surface reflectance product of an OLI sensor, and carrying out data preprocessing; 2, calculating the surface emissivity; 3, calculating topographic parameters; 4, calculating atmospheric parameters; wherein the atmospheric parameters comprise atmospheric transmissivity, atmospheric uplink radiation and atmospheric downlink radiation; 5, carrying out mountain land surface temperature inversion. Aiming at complex terrains of mountain lands, terrain parameters such as gradient, slope direction and sky visual factors are introduced into a thermal radiation transmission equation, and a mountainous region surface temperature inversion method is established, so that errors caused by inversion of the mountainous region surface temperature by utilizing a conventional thermal radiation transmission equation are effectively reduced, and the mountain land surface temperature inversion precision is improved.

The invention provides a car-reversing auxiliary system. The car-reversing auxiliary system comprises a camera for obtaining images at the back of a car, and a display for displaying the images at the back of the car. The car-reversing auxiliary system further comprises a redrawing device which allows an operator to draw auxiliary indicators used for assisting the operator in reversing the car, and the auxiliary indicators are displayed on the images at the back of the car in a superimposed way. According to the car-reversing auxiliary system provided by the invention, the operator can redraw the auxiliary indicators by self under the condition that an auxiliary line does not meet the actual conditions, so as to make the auxiliary line meet the actual conditions. On the other hand, the invention further provides a car-reversing auxiliary method.

The invention discloses a modular epidemic prevention and control type police robot. The robot comprises a mobile chassis, a monitoring system, a temperature measuring system, an interaction system, adisinfection system, a storage cabin, a positioning navigation system and a control system. The robot adopts modular design, different functional modules can be carried according to actual combat requirements, the robot can adapt to complex epidemic prevention work as soon as possible, and the problem that an existing epidemic prevention robot is weak in environmental adaptability is solved.

The invention discloses a system for detecting sediment content by adopting capacitive differential pressure sensors. The system comprises a water sample pretreatment system, a signal acquisition system, a control treatment system, and a touch screen display system, wherein the systems are orderly connected, the touch screen display system is connected to a data transmission system; the data transmission system is connected with a configuration monitoring system; the signal acquisition system comprises a depth sensor, a sediment content sensor, a flow velocity sensor, a water temperature sensor and an analog quantity input/output module; and the depth sensor, the sediment content sensor, the flow velocity sensor and the water temperature sensor are connected with the analog quantity input/output module respectively; and the analog quantity input/output module comprises four analog signal input channels. The system comprehensively considers a plurality of environmental factors, realizes the high-accuracy sediment content measurement, can be applied in the field of general river sediment detection, and can also be applied in fields such as water quality management of reservoir areas and channel cleanout of yellow river basin, sludge discharge management of sewage treatment plants and the like.

The invention discloses a vehicle-mounted non-blind-area positioning navigation and travel posture measuring device and a measuring method. The measuring device comprises a satellite positioning module, a magnetic resistance sensor, a tilt angle sensor, a CAN bus interface, a micro controller and an external input/output interface; the outputs of the satellite positioning module, the magnetic resistance sensor and the tilt sensor are connected with the input of the micro controller; the micro controller is in two-way connection with a whole-vehicle CAN bus through the CAN bus interface; the micro controller is in two-way connection with an external device through the external input/output interface. The measuring device has the benefits of no positioning blind area, low development and design cost and low maintenance cost.

The invention discloses a multi-source sensor joint calibration method based on a three-dimensional regular octagonal structure. The method comprises the steps of calibrating internal parameters of acamera; constructing a three-dimensional laser point cloud model; correcting the three-dimensional laser point cloud based on inertial data; optimizing a laser jitter error superposition mean value; optimizing the laser stepping angle error increment; and constructing a multi-line-segment joint external parameter calibration model, and solving an external parameter optimal solution through a nonlinear method. The method can effectively improve the joint calibration precision and the data fusion effect of the multi-source sensor in a complex agriculture and forestry environment.

The invention belongs to a building reinforced concrete cutting apparatus, in particular to a hydraulic concrete cutting machine. The hydraulic concrete cutting machine is characterized by comprising a hydraulic supply source, an electric cabinet, a guide way component (1), a cutting feed component (4), a rotation component (7), a speed reducing component (12), a base component (14) and a hydraulic motor (16); the guide way component (1) is fixed on a guide way supporting frame (2); the base component (14) and the cutting feed component (4) are respectively installed at two sides of the guide way component (1); the rotation component (7) and the speed reducing component (12) are vertically connected and fixed above the base component (14); the hydraulic motor (16) is arranged at one end of the rotation component (7) and communicated with the hydraulic supply source through a hydraulic oil pipe; and the electric cabinet controls the operation of the cutting machine through a control cable. The invention adopts the cable control of the electric cabinet to automatically cut, is safe, has high efficiency, low cutting cost, convenient operation, no dust during operation and no repair for cutting processing.

The present invention provides a method of making up for walking two-value image distortion, main resolving the problem of human information defect contained in the walking two-value image of the walking database. The particular courses of the method is: processing denoising, cutting pretreatment to the walking two-value image to obtain a normal walking image; acting the difference value of the present representation walking image and a prevFrame normal walking image as frame difference of the normal walking image; adding each walking image in the walking cycle, and getting its average to obtain the walking energy image of the cycle; filtering the walk energy image according to the set threshold, removing the noise or weaker signal of the energy image to obtain a walking main image; combining the walking main image and the normal walking image to obtain frame difference main image to make up for distortion of the walking two-value image caused by human information defect, and embodying changes of the form under the walking course. The invention can effectively attribute the walking two-value image and improve discrimination, also can used for the image processing course of the walking identification.

The invention relates to a system and a method for improving the angle measuring accuracy of an inductosyn, belonging to the technical field of an inductosyn angle measuring system. A digital potentiometer, a FPGA and an active phase shift circuit are used to form a programmable signal phase adjustment unit, and the phase difference of two signals is adjusted through FPGA programming, so that theangle measurement accuracy of the system is improved.