53 results about "Omnidirectional vision" patented technology

The invention discloses a roadside parking space detection device based on computer vision, comprising an omnidirectional vision sensor and a microprocessor, wherein the omnidirectional vision sensor is installed at the roadside with parking spaces; the microprocessor is used for analyzing the parking space state according to the data of a parking space video shot by the omnidirectional vision sensor; and the microprocessor comprises an omnidirectional image acquisition unit, a parking space customizing module, a parking space state detection module based on a sampling point model, and a parking space state information releasing module. The invention provides the roadside parking space detection device which is based on the computer vision, has the advantages of wide detection range, high detection precision, good detection instantaneity, high cost performance, strong environmental adaptation capability and good expandability and sustainable development, is convenient to implement and maintain and is convenient to realize on an embedded system.

The invention discloses a calibration method of a fish-eye lens imaging system applied to target detection, which comprises the step of mathematical modeling of the fish-eye lens imaging system based on an equidistant projection and the precise calibrations of the intrinsic parameters of the imaging system (including the central point O3(u,v) of an imaging surface), the distance L between the top section of the fish-eye lens and a theoretic refractive optical center plane, the radial distortion coefficient K of the fish-eye lens, an image element aspect scale factor i, the plane torsional angle gamma between a camera coordinate system and an imaging surface coordinate system, various parameter calibration methods and matched software. The method aims at the omnidirectional vision system based on the fish-eye lens, analyzes the optical structure and the imaging principle of the fish-eye lens and introduces deformation parameters so as to establish the physical model of the fish-eye lens imaging system and then deduce the mathematical model of the system. With an independently-developed calibration procedure, three-dimensional space information is obtained from a two-dimensional image and applied to the calibration of imaging system parameters. The method is applicable to the calibration of various fish-eye type panoramic lenses, is convenient and accurate and has strong practicability.

The invention discloses a panoramic three-dimensional photographing device which is established by integrating four omnibearing photographing devices with identical imaging parameters, wherein a plane is used for connecting the four omnibearing photographing devices with identical imaging parameters so as to ensure that fixed single viewpoints of four ODVSs (Omnidirectional Vision Sensors) with identical imaging parameters are positioned on the same plane; the connection manner is as follows: four hyperbolical-surface mirror planes with identical parameters are fixed on a transparent glass face, and four cameras with identical inner and outer parameters are fixed on the same plane; and a microprocessor is used for performing three-dimensional imaging processing on images of the four ODVSs and comprises a panoramic image reading and preprocessing unit, a perspective unfolding unit and a panoramic three-dimensional image output unit; and the panoramic three-dimensional photographing device can be widely applied to a plurality of application fields such as robot vision, animated films, games and the like. The invention provides the panoramic three-dimensional photographing device with the advantages of high cost performance, simplicity for operation and capability of photographing a panoramic three-dimensional video image in real time.

The utility model discloses a three-dimensional wind measuring unit based on a comprehensive vision, which comprises a bracket, an omni-directional vision sensor and a microprocessor for detecting wind speed and wind direction. A wind cup, a reducing mechanism, a fixed connecting link, a wind mark and a wind direction orientation block are arranged in the horizontal direction and the vertical direction of the bracket. The wind cup is fixedly connected with the upper side of the input shaft of the reducing mechanism, while the lower end of the input shaft of the reducing mechanism is sleeve-jointed on the upper end of the fixed connecting link. The wind mark is sleeved on a connecting shaft; the connecting shaft is fixedly connected with the lower end of the fixed connecting link; the wind direction orientation block is arranged at the end of the wind mark; the upper end of the wind direction orientation block is positioned at the same plane with the reducing mechanism. Two pointers are respectively positioned on the input shaft and the output shaft of the reducing mechanism, and the upper end of the two pointers and the wind direction orientation block are provided with marker color. The visual range of the vision sensor comprises the plane where the vertical direction and the horizontal direction of the reducing mechanism are positioned. The utility model has the advantages of high-precision, high sensitivity, long-term reliability, high reliability in test results and effective test of 3D wind speed and direction detection.

The invention discloses an omnibearing 3D (Three-Dimensional)modeling system based on an initiative omnidirectional vision sensor, which comprises an omnibearing vision sensor, a mobile surface laser source and a microprocessor used for carrying out 3D omnidirectional reconstruction on an omnibearing image, wherein the center of the omnibearing vision sensor and the center of the mobile surface laser source are configured on the same axes line; and the microprocessor comprises a video image reading module, an omnibearing vision sensor calibrating module, a Bird-View conversion module, an omnibearing laser information analyzing module, a joint calibrating module, a position estimating module of a linear motor of the mobile surface laser source, a computing module of point cloud geographic information of a mobile surface, a fusion module of Geographic information and color information of a point cloud, an omnibearing 3D model constructing module constructed by using position information of the mobile surface, an omnibearing 3D model generation module and a storage unit. By using the omnibearing 3D modeling system based on the initiative omnidirectional vision sensor, the perfect integration of geographic accuracy and third dimension of reconstruction of an omnibearing 3D model as well as reconstruction process automation is realized.

The invention provides a large-field-of-view scene perception method and system based on an omnidirectional vision, a medium and a device. The large-field-of-view scene perception method based on omnidirectional vision comprises the following steps of: receiving images transmitted by all cameras in real time, identifying obstacle information through image semantic segmentation, and further constructing local grid maps of all the cameras; wherein all the cameras are arranged around a robot, and the total visual angle range of the cameras covers a 360-degree environment around the robot; carrying out path planning by using the constructed local grid maps and controlling the operation of the robot according to the planned path. The method can collect the environmental information of 360-degree directions around the robot, makes up the defect of small visual range of a single camera, can adapt to more complex environments, and has excellent scene generalization and robustness.

A double-eye solid omnidirectional vision sensing device without a dead angle comprises two omnidirectional vision sensors with the same average resolving power, a connecting unit and a microprocessor used for processing an image. The omnidirectional vision sensor comprises a once catadioptric mirror face, a twice catadioptric mirror face, a transparent cover and a videography component. The once catadioptric mirror face and the twice catadioptric mirror face are arranged at the transparent cover; the videography component is arranged at a vision point at the back of the once catadioptric mirror face. The twice catadioptric mirror face is arranged in front of the once catadioptric mirror face. The once catadioptric mirror face and the twice catadioptric mirror face are both provided with a pore. The twice catadioptric mirror face is embedded into a wide-angle lens. A videography component lens, the wide-angle lens and the central shafts of the once catadioptric mirror face and the twice catadioptric mirror face are collocated at the same axial line. The backs of the two omnidirectional vision sensors are connected with each other by the connecting unit. The double-eye solid omnidirectional vision sensing device without dead angle can eliminate the dead angle in the vertical direction and capture the omnidirectional images of the large range of real time 360 degrees multiplied by 360 degrees.

The invention relates to an earthquake prediction auxiliary system based on a technology of Internet of things. The system comprises omnidirectional vision-based biological earthquake predictors placed on all observation points and an earthquake prediction server which is used for performing comprehensive analysis on monitoring results of the omnidirectional vision-based biological earthquake predictors on all the observation points, wherein all the omnidirectional vision-based biological earthquake predictors are communicated and connected, and all the omnidirectional vision-based biologicalearthquake predictors are communicated and connected with the earthquake prediction server; and the earthquake prediction auxiliary system can improve the credibility of earthquake early warning and simultaneously provide a certain reference basis for future judgment of earthquake occurrence times and earthquake occurrence areas by combining omnidirectional vision, computer vision, a database andother technologies for analyzing activities, postures, actions, behaviors and the like of animals, enabling the behaviors of the animals distributed on monitoring points of all places to speak and adopting comprehensive analysis means including time sequence of speaking, spatial positions, speaking sequence of the animals in different species and the like.

The invention relates to a fish-eye lens omnidirectional vision guidance method based on a sequence double-color lattice fairway buoy. The method comprises the establishment of a fish-eye omnidirectional vision system, the construction of a fairway buoy, the initialization of a fairway buoy recognizer, the pretreatment of images, the identification of the current effective fairway buoy, the switching of a sequence fairway buoy and a guidance strategy of a moving body. The invention relates to application of fish-eye lens omnidirectional vision navigation, ingests the whole scene of the whole hemisphere field disposably, does not need to depend on rotation of a single camera or image synthesis of a plurality of vidicons and is greatly favorable for omnidirectional observation and independent guidance of the moving body; a sequence double-color lattice type fairway buoy mode has topological characteristic close to natural scenery in indoor and outdoor structural environment, is simple, practical, easy to set and maintain and can effectively improve automatic identification and tracking efficiency of the fairway buoy in a distorted image of a large fish-eye omnidirectional vision scene; and a specific fairway buoy multi-target recognizer is supported on an embedded type DSP hardware platform to form an omnidirectional vision vehicle guidance system which realizes the application target of independent guidance of the moving body.

The invention discloses an active three-dimensional omnidirectional vision-based river width measuring device, which comprises a GPS sensor, a sonarprobe, an ominidirectional vision sensor with a fixed and single viewpoint, a full-view color body structure optical generator with fixed and single transmitting central point, and a microprocessor used for carrying out three-dimensional photographic measurement on the river width, acquiring river map data and fusing river depth data; the viewpoint of the omnidirectional vision sensor and the transmitting central point of the full-view color body structure optical generator are configured in the same axis line; and the omnidirectional vision sensor, the full-view color body structure optical generator, the GPS sensor and the sonarprobe are fixed in the extended line of the same upright post. The active three-dimensional omnidirectional vision-based river width measuring device has the advantages of quick measurement, good real-time, strong practicability and high robustness.

The invention discloses a coordination and cooperation control method for football robots and a system thereof. The method comprises the following steps: I, self-positioning of the football robots is performed according to the quantum immune algorithm; and II, coordination and cooperation are performed by the football robots in the antagonistic process of the game according to the algorithmic game theory. The system comprises a data processor, a data memory, a serial port communication module, a wireless network card, a 1394 control card, a compass, a milemeter, an omnidirectional vision module, an intelligent power module, motion control cards and motor drive circuit modules; the output ends of the motor drive circuit modules are connected with motors; the motors are connected with motor encoders; the motor encoders and a football robot kicking mechanism are all connected with the intelligent power module; the wireless network card is connected with a football robot trainer. According to the invention, the design is novel and reasonable, the coordination ability among multiple robots is improved, the collaboration efficiency is improved, the goals are increased, the winning probability for a football robot match is improved, the using effect is good, and the popularization and use are facilitated.

The invention relates to a cave detection device based on a binocular stereo omnidirectional vision sensor, which comprises a cave detector immersed in a cave to be detected and a portable microprocessor arranged in the edge of a cave entrance; geological feature video information, temperature and humidity information and oxygen concentration information inside the cave are transmitted to the portable microprocessor through a wireless transmission mode; the cave detector comprises two omnidirectional vision sensors with same imaging parameters and a microprocessor for compressing and transmitting images of the two omnidirectional vision sensors to form the binocular stereo omnidirectional vision sensor; and the portable microprocessor obtains the oxygen content, the temperature and humidity and other environment data of the cave by the wireless video mode and integrates the detection of the geological feature environment and the biological living environment of the cave together. The cave detection device can transmit the ambient panoramic shape image, the oxygen concentration, the temperature and humidity and other environmental information of the cave to an observer in the cave entrance by the video mode and has good reliability.

The invention relates to an omnidirectional vision intelligent trolley convenient for secondary development. The omnidirectional vision intelligent trolley is mainly composed of a power supply module,a main control module, a sensor module, a motor driving module, an omnidirectional vision system and a trolley mechanical structure. The omnidirectional vision intelligent trolley is characterized inthat the power module is connected with the main control module, the sensor module, the motor driving module and the omnidirectional vision system, the main control module is connected with the sensor module, the motor driving module and the omnidirectional vision system, and the motor driving module is connected with the trolley mechanical structure; the sensor module comprises a sub-processor and three ultrasonic sensors; the motor driving module comprises two driving chips; the omnidirectional vision system comprises a camera and a light reflective structure. The intelligent trolley has the advantages that the trolley is combined with the deep learning to realize the operation of high intelligence, has the functions of secondary development, omni-directional vision and omni-directionalmovement, and can be conveniently developed into robots with various different functions.

The invention discloses an all direction sighting device that includes all direction reflector and camera. All direction reflecting lens and camera are installed in the tunnel body. The center pointer of the al direction reflecting lens is located on the optic axis of the camera, and the camera installation base is set on leveling regulation mechanism. The all direction reflection lens is made up of level geometric proportion lens and vertical geometric proportion lens. The invention could realize the distance relation of panorama and real scene linear.

The invention discloses an animal behavior detection device based on omnidirectional vision. The animal behavior detection device comprises an omnidirectional vision sensor and a microprocessor for analyzing, processing and judging the activities, postures, actions and behaviors of animals according to a panoramic image shot by using the omnidirectional vision sensor and relevant input information, wherein the omnidirectional vision sensor is arranged above a monitoring environment, and is used for acquiring a panoramic video image in a real-time monitoring environment and providing a remote animal monitoring and observing measure; technologies such as omnidirectional vision, computer vision and a database and the like are used for analyzing the activities, postures, actions and behaviors of animals, so that the automation and intelligence levels of animal observation and detection are raised; an animal behavior PAE (Physical Address Extension) encoding way is adopted, any animal behavior is described in a data way, and measures are provided for the decomposition, identification and description of animal behaviors by using PAE encoding; and due to the adoption of the device, an effective animal behavior research meathod is provided for animal behavior researchers.

A method for detecting and tracking a movement target based on omnidirectional vision of a robot relates to the technical field of robots. The method comprises the following steps: step 1: acquiring and analyzing an image; step 2: detecting the movement target: (2.1), detecting the movement target by adopting an inter-frame differential method, and analyzing an image sequence through an optical flow method after detection; (2.2), comparing two types of detection images in the step (2.1) to obtain incomplete data; (2.3), constructing the incomplete data into complete data by adopting a Gaussian mixture model; detecting the movement target through the Gaussian mixture model; and step 3: tracking the movement target: (3.1), establishing a target reference object, and carrying out target tracking and detection through the reference object, and filtering target color; and (3.2), carrying out positioning processing on the tracked target, and calculating the sequence of a positioned target. By virtue of the method, the target can be conveniently and quickly detected and tracked; and the positioning is accurate and convenient, the efficiency is improved and the safety is high.

Disclosed in the invention is a DSP-based omnidirectional visual sense processing system. On the basis of a design for omnidirectional imaging of a robot, the system is composed of an omnidirectional vision sensor, a CCD camera, and a DSP digital image processing system. Environment information (light) passes through the omnidirectional vision sensor firstly and is reflected by the mirror surface of the omnidirectional vision sensor; and because of utilization of the omnidirectional vision sensor with a hyperbola rotator, the light is emitted to another focus of hyperbola and the optic center of the camera is just located at another focus of the hyperbola, so that the external scenery can be imaged on the camera. The DSP digital image processing system collects the image to an SRAM of a digital image collection system after SAA711AHZ analog-to-digital conversion; the DSP carries out image pretreatment and image resolving on the image and then stores the processed image to an SDRAM; and the processed image in the SDRAM or the original image in the SRAM can be transmitted to a robot PC104 through an RS232I/O port of the DSP digital image processing system by programming. And the robot PC104 carries out robot positioning, target identification and tracking, obstacle avoidance, and path planning and the like based on the received image.

The invention discloses a novel aspheric reflector design method for an omnidirectional vision system. The method is characterized by comprising steps: the diameter, the thickness and the maximum incidence angle of the reflector are determined, a proper sampling point number is selected according to the maximum incidence angle and required processing precision, the diameter and the thickness are subdivided according to the sampling point number, as for a cluster of coaxial hyperbolic curves with the same focus and with real axis lengths increased linearly, curve segments are intercepted horizontally for finite element splicing to obtain a surface-type curve meeting certain requirements, an iterative formula is substituted, coordinates of the sampling points are obtained, the sampling points generated by an iterative equation are used for fitting an aspheric contour curve equation, and the equation is converted into ZMAX recognition optical equation.

The invention comprises a vertically hanging reflector, a transparent barrel, a video cameral, and a mounting rack having an upper securing structure. The reflector is located on the upper portion of the transparent barrel; the lower portion of the transparent barrel there is a flat reflector; in the center of convex mirror of the reflector there is a through hole; the video camera is located above the convex mirror of the reflector; a black taper whose diameter gradually goes to small is set between the flat reflector and the reflector, and the taper is secured in the center of the flat mirror; the reflector, flat reflector, taper and lens of video camera all are set on a same central axis; the output of the video camera is connected to a image processing MPU. The invention also comprises an omnidirectional vision sensor.

The invention discloses an intelligent walking stick based on grid laser omnidirectional vision. The intelligent walking stick structurally comprises a base, a supporting rod, a handle, a help seeking button, a voice broadcast opening, a noctilucent ring, a glove and a temperature control button, due to the fact that the distance between a refractor and the ground is changed, the area of an obtained light source is changed, and by changing the installation mode of a mirror plate of the spherical refractor, the mirror plate is limited through the limiting mechanism, the angle of the mirror plate can be changed according to different users, the area of a light source can be obtained after the distance between the refractor and the ground is changed, the front light source can be obtained and refracted into the camera in time, and the walking stick can normally provide front road surface information for the users; when a user passes through a wet muddy ground, the friction force between the base and the muddy ground is small, the borne supporting force is reduced, and the slipping phenomenon is likely to occur, the inserting rods in the sleeve plates are inserted into the ground, the slipping phenomenon occurring when the base passes through the wet muddy ground is reduced, and safe use of personnel is facilitated.