WO2016123813A1 - 一种智能设备的姿态关系计算方法和智能设备 - Google Patents

一种智能设备的姿态关系计算方法和智能设备 Download PDF

Info

Publication number
WO2016123813A1
WO2016123813A1 PCT/CN2015/072461 CN2015072461W WO2016123813A1 WO 2016123813 A1 WO2016123813 A1 WO 2016123813A1 CN 2015072461 W CN2015072461 W CN 2015072461W WO 2016123813 A1 WO2016123813 A1 WO 2016123813A1
Authority
WO
WIPO (PCT)
Prior art keywords
coordinate system
calibration device
calibration
angle
relationship
Prior art date
Application number
PCT/CN2015/072461
Other languages
English (en)
French (fr)
Inventor
李颖哲
宋志刚
耿卫东
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to CN201580010064.2A priority Critical patent/CN106461414B/zh
Priority to PCT/CN2015/072461 priority patent/WO2016123813A1/zh
Publication of WO2016123813A1 publication Critical patent/WO2016123813A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C25/00Manufacturing, calibrating, cleaning, or repairing instruments or devices referred to in the other groups of this subclass

Definitions

  • the present invention relates to the field of communications, and in particular, to a method and a smart device for calculating an attitude relationship of a smart device.
  • the sensor of the smart device can be used to collect sensing data such as acceleration and magnetic flux
  • the camera device of the smart device can be used to collect data such as image or video, in order to accurately calculate the angle in the attitude measurement, How to calibrate the attitude relationship between the sensor coordinate system and the camera coordinate system in the smart device is an urgent problem to be solved.
  • the invention provides an attitude relationship calculation method and an intelligent device of a smart device, which can calibrate the attitude relationship between the sensor coordinate system and the coordinate system of the camera device in the smart device.
  • the present invention provides a smart device, including: an obtaining unit, a first calculating unit, a second calculating unit, and a third calculating unit, wherein:
  • the acquiring unit is configured to acquire measurement posture information of the calibration device coordinate system in a world coordinate system when the image capturing device acquires an image of the calibration device, wherein the measurement posture information is used to represent the calibration device coordinate system The orientation in the world coordinate system;
  • the first calculating unit is configured to calculate a first attitude relationship between the world coordinate system and a sensor coordinate system when the image capturing device acquires an image of the scaling device, and calculate the camera device to collect the calibration device a second attitude relationship between the camera coordinate system and the calibration device coordinate system;
  • the second calculating unit is configured to acquire an initial posture relationship between the camera coordinate system and the sensor coordinate system, and calculate the initial posture relationship, the first posture relationship, and the second posture relationship according to the initial posture relationship
  • the calibration device coordinate system is in the Estimating attitude information in the world coordinate system
  • the third calculating unit is configured to calculate an attitude relationship between the camera coordinate system and the sensor coordinate system calibration based on the estimated posture information and the measured posture information.
  • the third calculating unit includes:
  • a comparing unit configured to compare the estimated posture information with the measured posture information
  • the device further includes:
  • a generating unit configured to generate a new attitude relationship between the camera coordinate system and the sensor coordinate system based on a preset generation rule of the generated posture relationship when the comparison result does not reach the preset comparison result;
  • a fourth calculating unit configured to calculate, according to the generated attitude relationship, the first attitude relationship, and the second attitude relationship, that the camera device collects an image of the calibration device Estimated pose information in the world coordinate system.
  • the third calculating unit includes:
  • the solution obtained by the above formula is an attitude relationship between the camera system coordinate system and the sensor coordinate system calibration
  • the min() is a minimum value function
  • the i represents the ith shot of the calibration device.
  • a position where i is greater than or equal to 1, and less than or equal to n
  • the n is the number of shooting positions at which the scaling device is photographed
  • j represents the position of the calibration device
  • the imaging device is acquired j times, wherein j is greater than or equal to 1, and less than or equal to m
  • the m is the number of times the camera device collects at each shooting position
  • w 1 , w 2 , and w 3 are presets.
  • the camera device respectively collects images of the calibration device at n shooting positions, and each shooting position is acquired m times at different angles, wherein the n is greater than or equal to An integer of 1, the m being an integer greater than or equal to 1;
  • the acquiring unit is configured to acquire measurement posture information of the calibration device coordinate system in a world coordinate system when the calibration device is in each shooting position.
  • the second calculating unit includes:
  • a first estimating unit configured to calculate, according to the initial posture information, the first posture relationship and the second posture relationship, that the camera device collects an image of the calibration device, and the calibration device coordinate system is in the An estimate of the downtilt of the world coordinate system;
  • a second estimating unit configured to calculate, according to the initial posture information, the first attitude relationship and the second attitude relationship, the X of the calibration device coordinate system when the image capturing device collects an image of the scaling device An estimate of the angle between the axis and the magnetic north;
  • a third estimating unit configured to calculate, according to the initial posture information, the first attitude relationship, and the second posture relationship, Y of the calibration device coordinate system when the image capturing device collects an image of the calibration device An estimate of the angle between the axis and the magnetic north.
  • the device further includes:
  • a fifth calculating unit configured to perform, according to the initial posture information, the first posture relationship, and the first The second attitude relationship calculates a third attitude relationship between the calibration device coordinate system and the world coordinate system when the camera device acquires an image of the calibration device.
  • the first estimating unit is configured to calculate, when the scaling device is in the ith shooting position, The product of the third attitude relationship at the jth acquisition by the camera device and the vector of the Z axis of the calibration device coordinate system in the calibration device coordinate system, and the product in the direction of gravity
  • An included angle is an estimated value of a downtilt angle of the scaling device coordinate system in the world coordinate system when the j-th acquisition by the imaging device is performed as the scaling device at the i-th shooting position, the i Greater than or equal to 1, and less than or equal to the n, the j is greater than or equal to 1, and less than or equal to the m.
  • the second estimating unit is configured to calculate, when the scaling device is in the ith shooting position, a product of the third attitude relationship at the jth acquisition by the imaging device and a vector of the X axis of the scaling device coordinate system in the calibration device coordinate system, and the product is in the magnetic north direction
  • An angle is an estimated value of an angle between an X-axis and a magnetic north of the coordinate system coordinate system when the imaging apparatus performs the j-th acquisition at the i-th shooting position of the scaling device, wherein the i is greater than or Equal to 1, and less than or equal to the n, the j is greater than or equal to 1, and less than or equal to the m.
  • the third estimating unit is configured to calculate, when the calibration device is in the ith shooting position, a product of the third attitude relationship at the jth acquisition by the imaging device and a vector of the Y axis of the scaling device coordinate system in the calibration device coordinate system, and the product is in the magnetic north direction
  • the angle is an estimated value of the angle between the Y-axis and the magnetic north of the coordinate system coordinate system when the imaging device performs the j-th acquisition at the i-th shooting position, and the angle is greater than or Equal to 1, and less than or equal to the n, the j is greater than or equal to 1, and less than or equal to the m.
  • the acquiring unit includes:
  • a first acquisition subunit configured to acquire the calibration device when the calibration device is in each shooting position a measured value of the downtilt of the taxonomy in the world coordinate system
  • a second acquiring subunit configured to acquire a measured value of an angle between an X axis and a magnetic north of the coordinate system coordinate system of the calibration device at each shooting position;
  • a third acquiring subunit configured to acquire a measured value of an angle between a Y axis and a magnetic north of the coordinate system coordinate system of the calibration device at each shooting position.
  • the comparing unit is configured to use the estimated value of the downtilt angle, the calibration device coordinate system An estimate of the angle between the X-axis and the magnetic north and an estimated value of the angle between the Y-axis and the magnetic north of the calibration device coordinate system and the measured value of the downtilt angle, and the X-axis of the calibration device coordinate system Performing a weighted distance calculation on the measured value of the angle between the magnetic north and the measured value of the angle between the Y-axis and the magnetic north of the calibration device coordinate system, and using the calculated result as the estimated posture information and the measured attitude information Comparison results.
  • the comparing unit is configured to calculate the estimated posture information and the measurement posture information by using a formula Compare:
  • the T is a comparison result of the estimated posture information and the measured posture information
  • the i represents an i-th shooting position where the scaling device is located, where i is greater than or equal to 1, and is less than or Equal to the n
  • the j represents the jth acquisition of the camera device when the calibration device is located, wherein j is greater than or equal to 1, and less than or equal to the m
  • w 1 , w 2 and w 3 is a preset weight
  • the An estimated value of a downtilt angle of the scaling device when the calibration apparatus performs the jth acquisition by the imaging device at the i-th shooting position a measurement value of a downtilt angle of the calibration device when the scaling device is in an i-th shooting position
  • the present invention provides a smart device, including: a processor, a network interface, a memory, a communication bus, a camera, and a sensor, wherein the communication bus is used to implement the processor, the network interface, and the a communication between the memory, the camera device and the sensor, the processor being configured to execute a program stored in the memory; wherein the program comprises:
  • the processor performs a program for calculating an attitude relationship between the camera coordinate system and the sensor coordinate system calibration based on the estimated posture information and the measurement posture information, include:
  • the attitude relationship between the camera coordinate system and the sensor coordinate system used when estimating the posture information is calculated as the camera coordinate system and the sensor The attitude relationship of the coordinate system calibration.
  • the program executed by the processor further includes:
  • the processor performs calculation of the camera coordinate system and the sensor coordinate system calibration based on the estimated posture information and the measurement posture information
  • the program of attitude relationships including:
  • the solution obtained by the above formula is an attitude relationship between the camera system coordinate system and the sensor coordinate system calibration
  • the min() is a minimum value function
  • the i represents the ith shot of the calibration device.
  • a position where i is greater than or equal to 1, and less than or equal to n
  • the n is the number of shooting positions at which the scaling device is photographed
  • j represents the position of the calibration device
  • the imaging device is acquired j times, wherein j is greater than or equal to 1, and less than or equal to m
  • the m is the number of times the camera device collects at each shooting position
  • w 1 , w 2 , and w 3 are presets.
  • the camera device separately collects the calibration at n shooting positions An image of the device, and each shooting position is acquired m times at different angles, wherein n is an integer greater than or equal to 1, and m is an integer greater than or equal to 1;
  • the program for obtaining the measurement posture information of the calibration device coordinate system in the world coordinate system when the acquisition camera performs the calibration device by the processor includes:
  • the processor performs, according to the initial posture information, the first posture relationship, and the The second attitude relationship calculation program for estimating the posture information of the calibration device coordinate system in the world coordinate system when the image capturing device acquires the image of the calibration device includes:
  • the program executed by the processor further includes:
  • the processor performs, according to the initial posture information, the first posture relationship, and the The second attitude relationship is a program for calculating an estimated value of the down-tilt angle of the calibration device coordinate system in the world coordinate system when the image capturing device acquires an image of the calibration device, and includes:
  • the third attitude relationship is a product of a vector of the Z coordinate of the calibration device coordinate system in the calibration device coordinate system, and an angle between the product and the gravity direction is used as the calibration device
  • the estimated value of the scaling device coordinate system in the z-th acquisition performed by the imaging device at the i-th shooting position is the estimated value of the downtilt angle of the world coordinate system, the i is greater than or equal to 1, and less than or equal to The n, the j is greater than or equal to 1, and less than or equal to the m.
  • the processor performs, according to the initial posture information, the first posture relationship, and the The second attitude relationship is a program for calculating an estimated value of the angle between the X-axis and the magnetic north of the coordinate system of the calibration device when the image capturing device acquires the image of the calibration device, including:
  • the processor performs, according to the initial posture information, the first posture relationship, and the The second attitude relationship is a program for calculating an estimated value of the angle between the Y-axis and the magnetic north of the coordinate system of the calibration device when the image capturing device acquires the image of the calibration device, including:
  • the performing, by the processor, acquiring the calibration device at each shooting position A program for measuring attitude information in a world coordinate system, including:
  • the processor performs a procedure for comparing the estimated posture information with the measured posture information ,include:
  • Estimating the downtilt angle, the estimated value of the angle between the X-axis and the magnetic north of the calibration device coordinate system, and the estimated value of the angle between the Y-axis and the magnetic north of the calibration device coordinate system respectively a measured value of the downtilt angle, a measured value of the angle between the X-axis and the magnetic north of the calibration device coordinate system, and a measured value of the angle between the Y-axis and the magnetic north of the calibration device coordinate system, and The calculation result is used as a comparison result between the estimated posture information and the measured posture information.
  • the program executed by the processor to compare the estimated posture information with the measurement posture information includes:
  • the estimated pose information is compared to the measured pose information by a formula:
  • the T is a comparison result of the estimated posture information and the measured posture information
  • the i represents an i-th shooting position where the scaling device is located, where i is greater than or equal to 1, and is less than or Equal to the n
  • the j represents the jth acquisition of the camera device when the calibration device is located, wherein j is greater than or equal to 1, and less than or equal to the m
  • w 1 , w 2 and w 3 is a preset weight
  • the An estimated value of a downtilt angle of the scaling device when the calibration apparatus performs the jth acquisition by the imaging device at the i-th shooting position a measurement value of a downtilt angle of the calibration device when the scaling device is in an i-th shooting position
  • an estimated value of an angle between an X-axis and a magnetic north of the coordinate system coordinate system of the j-th acquisition performed by the imaging device when the scaling device is in the i-th shooting position a measurement value of an angle between an X-axis
  • the present invention provides a method for calculating an attitude relationship of a smart device, which includes:
  • the calculating an attitude relationship between the camera coordinate system and the sensor coordinate system calibration based on the estimated posture information and the measurement posture information includes:
  • the attitude relationship between the camera coordinate system and the sensor coordinate system used when estimating the posture information is calculated as the camera coordinate system and the sensor The attitude relationship of the coordinate system calibration.
  • the method further includes:
  • the solution obtained by the above formula is an attitude relationship between the camera system coordinate system and the sensor coordinate system calibration
  • the min() is a minimum value function
  • the i represents the ith shot of the calibration device.
  • a position where i is greater than or equal to 1, and less than or equal to n
  • the n is the number of shooting positions at which the scaling device is photographed
  • j represents the position of the calibration device
  • the imaging device is acquired j times, wherein j is greater than or equal to 1, and less than or equal to m
  • the m is the number of times the camera device collects at each shooting position
  • w 1 , w 2 , and w 3 are presets.
  • the fourth possible aspect of the third aspect In an implementation manner, the image capturing device respectively collects images of the scaling device at n shooting positions, and each shooting position is acquired m times at different angles, wherein the n is an integer greater than or equal to 1, The m is an integer greater than or equal to 1;
  • the acquiring posture information of the calibration device coordinate system in the world coordinate system when the acquisition imaging device acquires the calibration device includes:
  • the calculating, according to the initial posture information, the first posture relationship, and the second posture relationship includes:
  • the method further includes:
  • the calculating, according to the initial posture information, the first posture relationship, and the second posture relationship The estimated value of the down-tilt angle of the calibration device coordinate system in the world coordinate system when the image capturing device acquires the image of the calibration device includes:
  • the calculating, according to the initial posture information, the first posture relationship, and the second posture relationship The estimated value of the angle between the X-axis and the magnetic north of the coordinate system of the calibration device when the image capturing device acquires the image of the calibration device includes:
  • the calculating, according to the initial posture information, the first posture relationship, and the second posture relationship The estimated value of the angle between the Y-axis and the magnetic north of the coordinate system of the calibration device when the image capturing device acquires the image of the calibration device includes:
  • the acquiring the measurement posture information of the calibration device coordinate system in the world coordinate system when the calibration device is acquired at each shooting position includes:
  • the comparing the estimated posture information with the measurement posture information includes:
  • Estimating the downtilt angle, the estimated value of the angle between the X-axis and the magnetic north of the calibration device coordinate system, and the estimated value of the angle between the Y-axis and the magnetic north of the calibration device coordinate system respectively a measured value of the downtilt angle, a measured value of the angle between the X-axis and the magnetic north of the calibration device coordinate system, and a measured value of the angle between the Y-axis and the magnetic north of the calibration device coordinate system, and The calculation result is used as a comparison result between the estimated posture information and the measured posture information.
  • the comparing the estimated posture information with the measurement posture information including:
  • the estimated pose information is compared to the measured pose information by a formula:
  • the T is a comparison result of the estimated posture information and the measured posture information
  • the i represents an i-th shooting position where the scaling device is located, where i is greater than or equal to 1, and is less than or Equal to the n
  • the j represents the jth acquisition of the camera device when the calibration device is located, wherein j is greater than or equal to 1, and less than or equal to the m
  • w 1 , w 2 and w 3 is a preset weight
  • the An estimated value of a downtilt angle of the scaling device when the calibration apparatus performs the jth acquisition by the imaging device at the i-th shooting position a measurement value of a downtilt angle of the calibration device when the scaling device is in an i-th shooting position
  • an estimated value of an angle between an X-axis and a magnetic north of the coordinate system coordinate system of the j-th acquisition performed by the imaging device when the scaling device is in the i-th shooting position a measurement value of an angle between an X-axis
  • the calibration device coordinate system is measured posture information in a world coordinate system, wherein the measurement posture information is used to indicate that the calibration device coordinate system is in the Determining the orientation of the world coordinate system; calculating a first attitude relationship between the world coordinate system and the sensor coordinate system when the camera device acquires an image of the calibration device, and calculating an image of the camera device collecting the calibration device a second attitude relationship between the camera coordinate system and the calibration device coordinate system; acquiring an initial attitude relationship between the camera coordinate system and the sensor coordinate system, and according to the initial posture relationship, the first Calculating estimated posture information of the scaling device coordinate system in the world coordinate system when the camera device acquires an image of the scaling device when the image capturing device acquires an image of the scaling device; based on the estimated posture information and The measured attitude information calculates an attitude relationship between the camera coordinate system and the sensor coordinate system calibration.
  • the attitude relationship between the sensor coordinate system and the camera coordinate system in the calibration smart device can be realized.
  • FIG. 1 is a schematic structural diagram of a smart device according to an embodiment of the present invention.
  • FIG. 2 is a schematic structural diagram of another smart device according to an embodiment of the present invention.
  • FIG. 3 is a schematic structural diagram of another smart device according to an embodiment of the present invention.
  • FIG. 4 is a schematic structural diagram of another smart device according to an embodiment of the present invention.
  • FIG. 5 is a schematic structural diagram of another smart device according to an embodiment of the present disclosure.
  • FIG. 6 is a schematic structural diagram of another smart device according to an embodiment of the present disclosure.
  • FIG. 7 is a schematic structural diagram of another smart device according to an embodiment of the present disclosure.
  • FIG. 8 is a schematic flowchart of a method for calculating an attitude relationship of a smart device according to an embodiment of the present invention.
  • FIG. 9 is a schematic flowchart of a method for calculating an attitude relationship of another smart device according to an embodiment of the present disclosure.
  • FIG. 10 is a schematic flowchart diagram of another method for calculating an attitude relationship of a smart device according to an embodiment of the present invention.
  • FIG. 1 is a schematic structural diagram of a smart device according to an embodiment of the present invention. As shown in FIG. 1, the method includes: an obtaining unit 11, a first calculating unit 12, a second calculating unit 13, and a third calculating unit. 14, where:
  • the acquiring unit 11 is configured to acquire measurement posture information of the calibration device coordinate system in a world coordinate system when the image capturing device acquires an image of the calibration device, wherein the measurement posture information is used to indicate that the calibration device coordinate system is The orientation of the world coordinate system.
  • the measurement posture information may be the orientation information of the inclination measuring device coordinate system in the world coordinate system as the measurement posture information, or the posture information of the calibration device coordinate system in the world coordinate system may be measured by the attitude measuring instrument.
  • the measurement posture information since the measurement posture information is measured by a special tool for measuring the posture, the measurement posture information can accurately represent the posture information of the calibration device coordinate system in the world coordinate system, or can be understood as the above measurement posture.
  • Information is highly accurate gesture information.
  • the smart device may measure the measurement posture information by using a tool carried by the smart device itself. Alternatively, the measurement status information input by the user may be received, or the measurement status information sent by another device may be received, and the like.
  • a first calculating unit 12 configured to calculate a first attitude relationship between the world coordinate system and a sensor coordinate system when the image capturing device acquires an image of the scaling device, and calculate a collecting device of the camera device The second attitude relationship between the camera coordinate system and the calibration device coordinate system when the image of the calibration device is described.
  • the first calculating unit 12 may be configured to calculate a first attitude relationship between the world coordinate system and the sensor coordinate system according to the sensing data used by the sensor.
  • the first attitude relationship between the world coordinate system and the sensor coordinate system is calculated by the acceleration and magnetic flux data collected by the sensor.
  • the above-mentioned calibration device may be an article placed at a specific position, for example, an article placed on a plane parallel to the horizontal plane, or an article placed on a plane 60 degrees from the horizontal plane, or the like.
  • the second attitude relationship between the camera coordinate system and the calibration device coordinate system is calculated by the image of the calibration device acquired by the imaging device.
  • the above calibration device may be a calibration plate.
  • a second calculating unit 13 configured to acquire an initial attitude relationship between the camera coordinate system and the sensor coordinate system, and calculate the initial posture relationship, the first posture relationship, and the second posture relationship
  • the calibration device coordinate system estimates the attitude information in the world coordinate system when the camera device acquires an image of the calibration device.
  • the initial posture relationship may be that the user inputs an attitude relationship between the camera coordinate system and the sensor coordinate system, or may receive an attitude relationship between the camera coordinate system and the sensor coordinate system transmitted by another device.
  • the attitude relationship between the camera coordinate system and the sensor coordinate system herein can be understood as a hypothetical attitude relationship, for example, it can be an ideal attitude relationship between the camera coordinate system and the sensor coordinate system, such as the unit matrix representation. Offset rotation relationship.
  • the attitude relationship between the world coordinate system and the camera coordinate system can be obtained.
  • the second attitude relationship between the camera coordinate system and the calibration device coordinate system is known, the attitude relationship between the world coordinate system and the calibration device coordinate system can be obtained, that is, The target device coordinate system estimates the attitude information in the world coordinate system.
  • the third calculating unit 14 is configured to calculate an attitude relationship between the camera coordinate system and the sensor coordinate system calibration based on the estimated posture information and the measured posture information.
  • the estimated attitude information and the measured attitude information are both spatial attitudes of the coordinate system coordinate system in the world coordinate system
  • the measured attitude information is actually measured, that is, the measured attitude information indicates that the calibration apparatus coordinate system is in the world coordinate.
  • the high-accuracy spatial attitude is obtained, and the estimated attitude information is calculated based on the initial attitude relationship described above.
  • the third computing unit 14 can be in the measuring position
  • the initial attitude relationship may be used as an attitude relationship between the camera coordinate system and the sensor coordinate system; or when the measured attitude information and the estimated state information are different, the initial posture is After the relationship is specifically adjusted, the adjusted attitude relationship is used as an attitude relationship between the camera coordinate system and the sensor coordinate system; or when the measured posture information and the estimated state information are different, the initial posture relationship is adjusted, And generating, according to the adjusted attitude relationship, estimated posture information of the calibration device coordinate system in the world coordinate system, and calculating an attitude relationship between the camera coordinate system and the sensor coordinate system according to the estimated posture information and the measured posture information. .
  • the foregoing energy device may include, but is not limited to, a tablet computer, a mobile phone, an e-reader, a remote controller, a personal computer (PC), a notebook computer, an in-vehicle device, a network television, a wearable device, and the like ( User equipment, UE).
  • a tablet computer a mobile phone, an e-reader, a remote controller, a personal computer (PC), a notebook computer, an in-vehicle device, a network television, a wearable device, and the like
  • PC personal computer
  • UE User equipment
  • the calibration device coordinate system is measured posture information in a world coordinate system, wherein the measurement posture information is used to indicate that the calibration device coordinate system is in the Determining the orientation of the world coordinate system; calculating a first attitude relationship between the world coordinate system and the sensor coordinate system when the camera device acquires an image of the calibration device, and calculating an image of the camera device collecting the calibration device a second attitude relationship between the camera coordinate system and the calibration device coordinate system; acquiring an initial attitude relationship between the camera coordinate system and the sensor coordinate system, and according to the initial posture relationship, the first Calculating estimated posture information of the scaling device coordinate system in the world coordinate system when the camera device acquires an image of the scaling device when the image capturing device acquires an image of the scaling device; based on the estimated posture information and The measured attitude information calculates an attitude relationship between the camera coordinate system and the sensor coordinate system calibration.
  • the attitude relationship between the sensor coordinate system and the camera coordinate system in the calibration smart device can be realized.
  • FIG. 2 is a schematic structural diagram of another smart device according to an embodiment of the present invention.
  • the method includes: an obtaining unit 21, a first calculating unit 22, a second calculating unit 23, and a third computing.
  • the third calculating unit 24 includes: a comparing unit 241 and a calculating subunit 242, wherein:
  • the acquiring unit 21 is configured to acquire measurement attitude information of the calibration device coordinate system in a world coordinate system when the image capturing device acquires an image of the calibration device, wherein the measurement posture information is used to indicate that the calibration device coordinate system is The orientation of the world coordinate system.
  • a first calculating unit 22 configured to calculate a first attitude relationship between the world coordinate system and the sensor coordinate system when the image capturing device acquires an image of the scaling device, and calculate the camera device to collect the calibration device a second attitude relationship between the camera coordinate system and the calibration device coordinate system when the image is imaged.
  • the first calculating unit 22 may process the acceleration and the magnetic flux pre-data collected by the sensor by using sensor data pre-processing to obtain more accurate sensor data, and then calculate the world coordinate system and the sensor through the sensor data. The first pose relationship of the coordinate system.
  • the first attitude relationship between the world coordinate system and the sensor coordinate system can be calculated by the following formula:
  • the first attitude relationship between the world coordinate system and the sensor coordinate system can be calculated by the above formula, and specifically, a rotation matrix can be obtained.
  • the first calculating unit 22 may calculate a second attitude relationship between the camera coordinate system and the calibration device coordinate system by using a camera to take a picture. This is a well-known technical feature and will not be described in detail.
  • a second calculating unit 23 configured to acquire an initial posture relationship between the camera coordinate system and the sensor coordinate system, and calculate the first posture relationship, the first posture relationship, and the second posture relationship according to the initial posture relationship
  • the calibration device coordinate system is in the The estimated pose information in the world coordinate system.
  • the comparing unit 241 is configured to compare the estimated posture information with the measured posture information.
  • the calculating sub-unit 242 is configured to, when the comparison result reaches a preset comparison result for calibrating the attitude relationship, calculate an attitude relationship between the camera coordinate system and the sensor coordinate system used when calculating the estimated posture information The attitude relationship between the camera coordinate system and the sensor coordinate system calibration.
  • the first attitude relationship between the world coordinate system and the sensor coordinate system, the second attitude relationship between the camera coordinate system and the calibration device coordinate system, the attitude relationship between the camera coordinate system and the sensor coordinate system, and the calibration device coordinate system are The relationship between the four pieces of attitude information in the world coordinate system is constant at the same time, so that a coordinate device coordinate system can be obtained by a hypothetical attitude relationship between the camera coordinate system and the sensor coordinate system.
  • the estimated attitude information in the world coordinate system, when the estimated attitude information is the same as or similar to the measured attitude information, the attitude relationship between the assumed camera coordinate system and the sensor coordinate system may be determined as the camera coordinate system and the sensor coordinates.
  • a calibrated attitude relationship is the attitude relationship between the assumed camera coordinate system and the sensor coordinate system.
  • the foregoing comparison result for calibrating the attitude relationship may be that the estimated posture information is the same as the measurement posture information, or the similarity between the estimated posture information and the measured posture information is greater than or equal to a specific threshold, or The difference between the estimated attitude information and the measured attitude information is less than or equal to a specific threshold, wherein the specific threshold may be a default or preset threshold of the smart device, for example: 98% or 99%, and the like.
  • the foregoing apparatus may further include:
  • the generating unit 25 is configured to generate a new attitude relationship between the camera coordinate system and the sensor coordinate system based on a preset generation rule of the generated posture relationship when the comparison result does not reach the preset comparison result;
  • a fourth calculating unit 26 configured to calculate, according to the generated attitude relationship, the first posture relationship and the second posture relationship, the calibration device coordinate system when the image capturing device collects an image of the calibration device Estimated pose information in the world coordinate system.
  • the difference between the estimated posture information and the measurement posture information may be large, and a new attitude relationship may be generated according to the preset rule.
  • a new attitude relationship may be generated according to the preset rule.
  • the estimated attitude information calculated according to the initial attitude relationship is different from the measured attitude information, it is indicated that the initial attitude relationship is not high accuracy of the camera coordinate system and the sensor coordinate system.
  • the attitude relationship, or the accuracy of the initial pose relationship described above, is relatively low, so that a new pose relationship needs to be generated based on the initial pose relationship.
  • the foregoing generating rule may include:
  • a new attitude relationship is generated based on the above comparison result.
  • the posture relationship between the camera coordinate system and the sensor coordinate system used when calculating the estimated posture information can be greatly adjusted;
  • the difference between the estimated attitude information calculated using the initial attitude relationship and the measured attitude information is large, the initial attitude relationship can be greatly adjusted.
  • the attitude relationship between the camera coordinate system and the sensor coordinate system used when calculating the estimated posture information may be slightly adjusted.
  • the initial attitude relationship can be adjusted to a small extent.
  • the foregoing preset rule may further include:
  • the attitude relationship between the camera coordinate system and the sensor coordinate system is randomly generated within a preset range.
  • the generating rule may further include:
  • the attitude relationship between the camera system coordinate system and the sensor coordinate system is generated by the Levenberg-Marquardt algorithm.
  • the attitude relationship between the camera coordinate system and the sensor coordinate system may be a rotation matrix R, and the rotation matrix may be transformed into a triple, wherein the transformation may be a Rodrigues transformation.
  • the calculating sub-unit 242 can calculate the attitude relationship between the camera coordinate system and the sensor coordinate system calibration using the estimated posture information calculated by the fourth calculating unit 26.
  • the shooting position of the above-mentioned calibration device covers the entire three-dimensional space as much as possible.
  • the camera device separately collects images of the calibration device at n shooting positions, and each shooting position is acquired m times at different angles, wherein the n is an integer greater than 1, and the m is greater than 1 The integer.
  • the camera device places the calibration device horizontally on the horizontal plane, the vertical horizontal plane of the calibration device, and the lowering angle of the calibration device at 60 degrees.
  • the timing device performs the acquisition, and in the above three positions, the calibration device may also rotate the calibration device once every 180 degrees, and then use the rotated calibration device.
  • each shooting position can be collected multiple times at different angles, for example: images are taken every 30 angles. In this way, six shooting positions can be obtained by the above method, and each shooting position is collected 12 times at different angles, and one image is collected 72 times.
  • the acquiring unit 21 may be configured to acquire measurement posture information of the calibration device coordinate system in the world coordinate system when the calibration device is in each shooting position.
  • the second calculating unit 23 may include:
  • a first estimating unit 231 configured to calculate, according to the initial posture information, the first attitude relationship, and the second posture relationship, that the camera device collects an image of the calibration device An estimate of the downtilt angle of the world coordinate system;
  • a second estimating unit 232 configured to calculate, according to the initial posture information, the first posture relationship, and the second posture relationship, that the camera device collects an image of the calibration device when the image of the calibration device is An estimate of the angle between the X-axis and the magnetic north;
  • a third estimating unit 233 configured to calculate, according to the initial posture information, the first posture relationship, and the second posture relationship, that the camera device collects an image of the calibration device when the image of the calibration device is Estimated value of the angle between the Y-axis and the magnetic north.
  • the attitude relationship between the camera coordinate system and the sensor coordinate system can be calculated by the above-described downtilt angle, the angle between the X-axis and the magnetic north, and the estimated value of the angle between the Y-axis and the magnetic north, and the above-described measurement and measurement posture information.
  • the above-described measurement and measurement posture information may include the above-described downtilt angle, the angle between the X-axis and the magnetic north, and the measured value of the angle between the Y-axis and the magnetic north.
  • the foregoing device may further include:
  • a fifth calculating unit 27 configured to calculate, according to the initial posture information, the first posture relationship and the second posture relationship, the coordinate system of the calibration device when the image capturing device acquires an image of the calibration device The third attitude relationship of the world coordinate system.
  • the first estimating unit 231 may be configured to calculate the third attitude relationship and the calibration device coordinate when the j-th acquisition by the imaging device is performed by the calibration device when the calibration device is in the i-th shooting position.
  • the product of the Z-axis of the vector in the calibration device coordinate system, and the angle between the product and the gravity direction is used as the imaging device when the scaling device is at the i-th shooting position.
  • the estimated value of the scaling device coordinate system in the world coordinate system at the time of j acquisition, the i is greater than or equal to 1, and less than or equal to the n
  • the j is greater than or equal to 1, and less than Or equal to the m.
  • the second estimating unit 232 may be configured to calculate the third posture relationship and the calibration device coordinate when the j-th acquisition by the imaging device is performed by the calibration device when the calibration device is in the i-th shooting position
  • the product of the X-axis of the system in the calibration device coordinate system, and the angle between the product and the magnetic north direction is used as the jth of the imaging device when the scaling device is at the i-th shooting position.
  • the third estimating unit 231 may be configured to calculate the third attitude relationship and the calibration device coordinate when the j-th acquisition by the imaging device is performed by the calibration device when the calibration device is in the i-th shooting position.
  • the product of the Y-axis of the system in the calibration device coordinate system, and the angle between the product and the magnetic north direction is taken as the jth of the imaging device when the scaling device is at the i-th shooting position.
  • the three-dimensional vector of the calibration device coordinate system X-axis, Y-axis, and Z-axis in the world coordinate system is calculated at the j-th acquisition by the imaging device when the scaling device is in the i-th shooting position.
  • the above-mentioned downtilt angle, the angle between the X-axis and the magnetic north, and the estimated value of the angle between the Y-axis and the magnetic north can be obtained.
  • the obtaining unit 21 may include:
  • a first acquisition subunit 211 configured to acquire a measured value of a downtilt angle of the calibration apparatus coordinate system in the world coordinate system when the calibration apparatus is at each shooting position;
  • a second obtaining subunit 212 configured to acquire a measured value of an angle between an X axis and a magnetic north of the coordinate system coordinate system of the scaling device at each shooting position;
  • the third obtaining sub-unit 213 is configured to acquire a measured value of an angle between a Y-axis and a magnetic north of the coordinate system coordinate system of the scaling device at each shooting position.
  • Comparison unit 241 can be used to estimate the downtilt angle, X of the calibration device coordinate system
  • the measured value of the included angle and the measured value of the angle between the Y-axis and the magnetic north of the calibration device coordinate system are weighted distance calculation, and the calculation result is used as a comparison between the estimated posture information and the measured attitude information result.
  • the comparison unit 241 can compare the estimated posture information with the measurement posture information by the following formula calculation:
  • the T is a comparison result of the estimated posture information and the measured posture information
  • the i represents an i-th shooting position where the scaling device is located, where i is greater than or equal to 1, and is less than or Equal to the n
  • the j represents the jth acquisition of the camera device when the calibration device is located, wherein j is greater than or equal to 1, and less than or equal to the m
  • w 1 , w 2 and w 3 is a preset weight
  • the An estimated value of a downtilt angle of the scaling device when the calibration apparatus performs the jth acquisition by the imaging device at the i-th shooting position a measurement value of a downtilt angle of the calibration device when the scaling device is in an i-th shooting position
  • w 1 , w 2 and w 3 may be configured in a preset manner, for example, w 1 :w 2 :w 3 may be configured to be 100:1:1.
  • FIG. 6 is a schematic structural diagram of another smart device according to an embodiment of the present disclosure. As shown in FIG. 6, the method includes: an obtaining unit 61, a first calculating unit 62, a second calculating unit 63, and a third calculating unit 64, wherein:
  • the acquiring unit 61 is configured to acquire measurement posture information of the calibration device coordinate system in a world coordinate system when the image capturing device acquires an image of the calibration device, wherein the measurement posture information is used to indicate that the calibration device coordinate system is The orientation of the world coordinate system.
  • a first calculating unit 62 configured to calculate a first attitude relationship between the world coordinate system and the sensor coordinate system when the image capturing device acquires an image of the scaling device, and calculate the camera device to collect the calibration device a second attitude relationship between the camera coordinate system and the calibration device coordinate system when the image is imaged.
  • a second calculating unit 63 configured to acquire an initial posture relationship between the camera coordinate system and the sensor coordinate system, and calculate the first posture relationship, the first posture relationship, and the second posture relationship according to the initial posture relationship
  • the calibration device coordinate system estimates the attitude information in the world coordinate system when the camera device acquires an image of the calibration device.
  • the third calculating unit 64 is configured to calculate an attitude relationship between the camera coordinate system and the sensor coordinate system calibration based on the estimated posture information and the measured posture information.
  • the attitude relationship between the camera coordinate system and the sensor coordinate system calibration is calculated by the target formula.
  • the target formula can be as follows:
  • the solution obtained by the above formula is an attitude relationship between the camera system coordinate system and the sensor coordinate system calibration
  • the min() is a minimum value function
  • the i represents the ith shot of the calibration device.
  • a position where i is greater than or equal to 1, and less than or equal to n
  • the n is the number of shooting positions at which the scaling device is photographed
  • j represents the position of the calibration device
  • the imaging device is acquired j times, wherein j is greater than or equal to 1, and less than or equal to m
  • the m is the number of times the camera device collects at each shooting position
  • w 1 , w 2 , and w 3 are presets.
  • w 1 , w 2 and w 3 may be configured in a preset manner, for example, w 1 :w 2 :w 3 may be configured to be 100:1:1.
  • the above formula can be optimized, and the attitude relationship between the camera coordinate system and the sensor coordinate system calibration can be obtained.
  • optimization can be solved by using the Levenberg-Marquardt algorithm.
  • the Levenberg-Marquardt algorithm with different optimization parameters may be used to generate a plurality of candidates (or as a hypothesis) of the camera device coordinate system and the sensor coordinate system, and then according to the camera coordinate system and the sensor.
  • the attitude relationship of the coordinate system calculates the estimated attitude information of the calibration device coordinate system in the world coordinate system, and the estimated attitude information calculated by the attitude relationship between the camera coordinate system and the sensor coordinate system of a candidate When the difference (or the mean square error) of the measured attitude information is the smallest, the attitude relationship between the candidate camera coordinate system and the sensor coordinate system can be used as the attitude relationship between the camera coordinate system and the sensor coordinate system.
  • the parameter to be solved in the above formula is a rotation matrix R, so that the parameter to be solved can be transformed into a triple, so the unknown parameter of the above formula is a set of ternary vectors.
  • the transformation can be a Rodrigues transformation.
  • FIG. 7 is a schematic structural diagram of another smart device according to an embodiment of the present invention.
  • the processor 71 includes a processor 71, a network interface 72, a memory 73, a communication bus 74, and an image capturing device 75.
  • a sensor 76 wherein the communication bus 74 is configured to implement connection communication between the processor 71, the network interface 72, the memory 73, the camera device 74, and the sensor 75, the processor 71 Means for executing a program stored in the memory 73; wherein the package include:
  • the program executed by the processor 71 to calculate the attitude relationship between the camera system 75 coordinate system and the sensor coordinate system calibration based on the estimated posture information and the measurement posture information may include:
  • the attitude relationship between the coordinate system of the imaging device 75 and the sensor coordinate system used when the estimated posture information is calculated is used as the coordinate system of the imaging device 75
  • the attitude relationship with the sensor coordinate system calibration is used as the coordinate system of the imaging device 75 The attitude relationship with the sensor coordinate system calibration.
  • the program executed by the processor 71 may further include:
  • the calibration device coordinate system is in the world coordinate system Estimating the posture information, and performing the step of comparing the estimated posture information with the measurement posture information again.
  • the program executed by the processor 71 to calculate the attitude relationship between the camera system 75 coordinate system and the sensor coordinate system calibration based on the estimated posture information and the measurement posture information may include:
  • the attitude relationship between the coordinate system of the camera device 75 and the calibration of the sensor coordinate system is calculated by the following formula
  • the solution obtained by the above formula is the attitude relationship between the coordinate system of the imaging device 75 and the sensor coordinate system, the min() is a minimum function, and the i represents the ith of the calibration device.
  • a photographing position where i is greater than or equal to 1, and less than or equal to n
  • the n is the number of photographing positions at which the scaling device is photographed
  • j represents each position of the scaling device
  • the imaging device 75 is acquired j-th time, wherein j is greater than or equal to 1, and less than or equal to m, the m is the number of times the imaging device 75 captures each shooting position, w 1 , w 2 , and w 3
  • the An estimated value of the downtilt angle of the scaling device when the z-th acquisition by the imaging device 75 is performed by the scaling device at the i-th shooting position
  • a measurement value of a downtilt angle of the calibration device when the scaling device is in an i-th shooting position An estimated value of an angle between an X-
  • the image capturing device 75 respectively collects images of the scaling device at n shooting positions, and each shooting position is acquired m times at different angles, wherein the n is an integer greater than or equal to 1. , the m is an integer greater than or equal to 1;
  • the program of the measurement posture information of the calibration device coordinate system in the world coordinate system when the acquisition imaging device 75 performs the calibration device by the processor 701 may include:
  • the program for estimating posture information in the world coordinate system of the device coordinate system may include:
  • the calibration device coordinate system is under the world coordinate system Estimated angle of inclination
  • the program executed by the processor 71 may further include:
  • the determining, by the processor 71, calculating, according to the initial posture information, the first posture relationship, and the second posture relationship, the calibration device coordinates when the image capturing device 75 acquires an image of the calibration device may include:
  • the determining, by the processor 71, calculating, according to the initial posture information, the first posture relationship, and the second posture relationship, the calibration device coordinates when the image capturing device 75 acquires an image of the calibration device may include:
  • the determining, by the processor 71, calculating, according to the initial posture information, the first posture relationship, and the second posture relationship, the calibration device coordinates when the image capturing device 75 acquires an image of the calibration device may include:
  • the program executed by the processor 71 to obtain the measurement posture information of the calibration device coordinate system in the world coordinate system when the calibration device is in each shooting position may include:
  • the program executed by the processor 71 to compare the estimated posture information with the measured posture information may include:
  • Estimating the downtilt angle, the estimated value of the angle between the X-axis and the magnetic north of the calibration device coordinate system, and the estimated value of the angle between the Y-axis and the magnetic north of the calibration device coordinate system respectively a measured value of the downtilt angle, a measured value of the angle between the X-axis and the magnetic north of the calibration device coordinate system, and a measured value of the angle between the Y-axis and the magnetic north of the calibration device coordinate system, and The calculation result is used as a comparison result between the estimated posture information and the measured posture information.
  • the program executed by the processor 71 to compare the estimated posture information with the measured posture information may include:
  • the estimated pose information is compared to the measured pose information by a formula:
  • the T is a comparison result of the estimated posture information and the measured posture information
  • the i represents an i-th shooting position where the scaling device is located, where i is greater than or equal to 1, and is less than or Equal to the n
  • the j represents the jth acquisition of the camera device 75 when the calibration device is located, wherein j is greater than or equal to 1, and less than or equal to the m
  • w 1 , w 2 And w 3 are preset weights
  • the smart device may include, but is not limited to, a tablet, a mobile phone, an e-reader, a remote controller, a PC, a notebook computer, an in-vehicle device, a network television, a wearable device, and the like.
  • the calibration device coordinate system is measured posture information in a world coordinate system, wherein the measurement posture information is used to indicate that the calibration device coordinate system is in the Determining the orientation of the world coordinate system; calculating a first attitude relationship between the world coordinate system and the sensor coordinate system when the camera device acquires an image of the calibration device, and calculating an image of the camera device collecting the calibration device a second attitude relationship between the camera coordinate system and the calibration device coordinate system; acquiring an initial attitude relationship between the camera coordinate system and the sensor coordinate system, and according to the initial posture relationship, the first Calculating estimated posture information of the scaling device coordinate system in the world coordinate system when the camera device acquires an image of the scaling device when the image capturing device acquires an image of the scaling device; based on the estimated posture information and The measured attitude information calculates an attitude relationship between the camera coordinate system and the sensor coordinate system calibration. Thereby calibrating the smart device The attitude relationship between the sensor coordinate system and the camera coordinate system.
  • FIG. 8 is a schematic flowchart of a method for calculating an attitude relationship of a smart device according to an embodiment of the present invention. As shown in FIG. 8, the method includes the following steps:
  • the measurement posture information may be the orientation information of the inclination measuring device coordinate system in the world coordinate system as the measurement posture information, or the posture information of the calibration device coordinate system in the world coordinate system may be measured by the attitude measuring instrument.
  • the measurement posture information since the measurement posture information is measured by a special tool for measuring the posture, the measurement posture information can accurately represent the posture information of the calibration device coordinate system in the world coordinate system, or can be understood as the above measurement posture.
  • Information is highly accurate gesture information.
  • the smart device may measure the measurement posture information by using a tool carried by the smart device itself. Alternatively, the measurement status information input by the user may be received, or the measurement status information sent by another device may be received, and the like.
  • step 802 is to calculate a first attitude relationship between the world coordinate system and the sensor coordinate system according to the sensing data used by the sensor.
  • the first attitude relationship between the world coordinate system and the sensor coordinate system is calculated by the acceleration and magnetic flux data collected by the sensor.
  • the above-mentioned calibration device may be an article placed at a specific position, for example, an article placed on a plane parallel to the horizontal plane, or an article placed on a plane 60 degrees from the horizontal plane, or the like.
  • the second attitude relationship between the camera coordinate system and the calibration device coordinate system is calculated by the image of the calibration device acquired by the imaging device.
  • the above calibration device may be a calibration plate.
  • the initial posture relationship may be that the user inputs an attitude relationship between the camera coordinate system and the sensor coordinate system, or may receive an attitude relationship between the camera coordinate system and the sensor coordinate system transmitted by another device.
  • the attitude relationship between the camera coordinate system and the sensor coordinate system herein can be understood as a hypothetical attitude relationship, for example, it can be an ideal attitude relationship between the camera coordinate system and the sensor coordinate system, such as the unit matrix representation. Offset rotation relationship.
  • the attitude relationship between the world coordinate system and the camera coordinate system can be obtained.
  • the second attitude relationship between the camera coordinate system and the calibration device coordinate system is known, the attitude relationship between the world coordinate system and the calibration device coordinate system can be obtained, that is, The target device coordinate system estimates the attitude information in the world coordinate system.
  • the estimated attitude information and the measured attitude information are both spatial attitudes of the coordinate system coordinate system in the world coordinate system
  • the measured attitude information is actually measured, that is, the measured attitude information indicates that the calibration apparatus coordinate system is in the world coordinate.
  • the high-accuracy spatial attitude is obtained, and the estimated attitude information is calculated based on the initial attitude relationship described above.
  • the step 804 can be used as the attitude relationship between the camera coordinate system and the sensor coordinate system when the measurement posture information is the same as or similar to the estimation state information; or when the measurement posture information and the estimation state information exist
  • the initial attitude relationship is specifically adjusted, and then the adjusted attitude relationship is used as an attitude relationship between the camera coordinate system and the sensor coordinate system; or when there is a difference between the measured attitude information and the estimated state information
  • the estimated posture information of the calibration device coordinate system in the world coordinate system is generated according to the adjusted posture relationship, and the camera coordinate system is calculated according to the estimated posture information and the measured posture information.
  • the attitude relationship with the sensor coordinate system calibration is performed by the initial posture relationship.
  • the foregoing method may be applied to the smart device, and may be implemented by a single smart device independently, or may be implemented by jointly combining multiple devices.
  • the smart device may include, but is not limited to, a tablet, a mobile phone, an e-reader, a remote controller, a PC, a notebook computer, an in-vehicle device, a network television, a wearable device, and the like.
  • the calibration device coordinate system is measured posture information in a world coordinate system, wherein the measurement posture information is used to indicate that the calibration device coordinate system is in the Determining the orientation of the world coordinate system; calculating a first attitude relationship between the world coordinate system and the sensor coordinate system when the camera device acquires an image of the calibration device, and calculating an image of the camera device collecting the calibration device a second attitude relationship between the camera coordinate system and the calibration device coordinate system; acquiring an initial attitude relationship between the camera coordinate system and the sensor coordinate system, and according to the initial posture relationship, the first Calculating estimated posture information of the scaling device coordinate system in the world coordinate system when the camera device acquires an image of the scaling device when the image capturing device acquires an image of the scaling device; based on the estimated posture information and The measured attitude information calculates an attitude relationship between the camera coordinate system and the sensor coordinate system calibration.
  • the attitude relationship between the sensor coordinate system and the camera coordinate system in the calibration smart device can be realized.
  • FIG. 9 is a schematic flowchart diagram of another attitude relationship calculation method according to an embodiment of the present invention. As shown in FIG. 9, the method includes the following steps:
  • the calibration device coordinate is measured posture information of the world coordinate system, wherein the measurement posture information is used to indicate that the calibration device coordinate system is at the world coordinate.
  • the orientation of the system is used to indicate that the calibration device coordinate system is at the world coordinate.
  • step 902 may preprocess the acceleration and magnetic flux data collected by the sensor by using sensor data preprocessing to obtain more accurate sensor data, and then calculate the world coordinate system and the sensor coordinate system by using the sensor data.
  • the first attitude relationship may be preprocess the acceleration and magnetic flux data collected by the sensor by using sensor data preprocessing to obtain more accurate sensor data, and then calculate the world coordinate system and the sensor coordinate system by using the sensor data. The first attitude relationship.
  • the first attitude relationship between the world coordinate system and the sensor coordinate system can be calculated by the following formula:
  • the first attitude relationship between the world coordinate system and the sensor coordinate system can be calculated by the above formula, and specifically, a rotation matrix can be obtained.
  • step 902 is to calculate a second attitude relationship between the camera coordinate system and the calibration device coordinate system by means of photographing by the camera. This is a well-known technical feature and will not be described in detail.
  • the comparison result reaches a preset comparison result for calibrating the attitude relationship
  • the attitude relationship between the camera coordinate system and the sensor coordinate system used when calculating the estimated posture information is used as the camera coordinate system.
  • the attitude relationship with the sensor coordinate system calibration is used as the camera coordinate system.
  • the first attitude relationship between the world coordinate system and the sensor coordinate system, the second attitude relationship between the camera coordinate system and the calibration device coordinate system, the attitude relationship between the camera coordinate system and the sensor coordinate system, and the calibration device coordinate system are The relationship between the four pieces of attitude information in the world coordinate system is constant at the same time, so that a coordinate device coordinate system can be obtained by a hypothetical attitude relationship between the camera coordinate system and the sensor coordinate system.
  • the estimated attitude information in the world coordinate system, when the estimated attitude information is the same as or similar to the measured attitude information, the attitude relationship between the assumed camera coordinate system and the sensor coordinate system may be determined as the camera coordinate system and the sensor coordinates. Department Quasi-posture relationship.
  • the foregoing comparison result for calibrating the attitude relationship may be that the estimated posture information is the same as the measurement posture information, or the similarity between the estimated posture information and the measured posture information is greater than or equal to a specific threshold, or The difference between the estimated attitude information and the measured attitude information is less than or equal to a specific threshold, wherein the specific threshold may be a default or preset threshold of the smart device, for example: 98% or 99%, and the like.
  • the method may further include the following steps:
  • the difference between the estimated posture information and the measurement posture information may be large, and a new attitude relationship may be generated according to the preset rule.
  • the estimated attitude information calculated according to the initial attitude relationship differs greatly from the measured attitude information, it is indicated that the initial attitude relationship is not a high-accuracy attitude relationship between the camera coordinate system and the sensor coordinate system, or the accuracy of the initial attitude relationship. It is relatively low, so it is necessary to generate a new attitude relationship based on the initial attitude relationship.
  • the foregoing generating rule may include:
  • a new attitude relationship is generated based on the above comparison result.
  • the posture relationship between the camera coordinate system and the sensor coordinate system used when calculating the estimated posture information can be greatly adjusted;
  • the difference between the estimated attitude information calculated using the initial attitude relationship and the measured attitude information is large, the initial attitude relationship can be greatly adjusted.
  • the attitude relationship between the camera coordinate system and the sensor coordinate system used when calculating the estimated posture information may be slightly adjusted.
  • the initial attitude relationship can be adjusted to a small extent.
  • the foregoing preset rule may further include:
  • the attitude relationship between the camera coordinate system and the sensor coordinate system is randomly generated within a preset range.
  • the generating rule may further include:
  • the attitude relationship between the camera system coordinate system and the sensor coordinate system is generated by the Levenberg-Marquardt algorithm.
  • the attitude relationship between the camera coordinate system and the sensor coordinate system may be a rotation matrix R, and the rotation matrix may be transformed into a triple, wherein the transformation may be a Rodrigues transformation.
  • step 904 can be performed again.
  • step 906 is repeatedly executed when the comparison result is still not up to the preset comparison result, and of course, step 906 may be to generate a new attitude relationship according to the posture relationship generated when step 906 was last executed. Step 907 is then repeated to form a loop until the comparison result of the estimated posture information and the measured posture information reaches the predetermined comparison result.
  • the shooting position of the above-mentioned calibration device covers the entire three-dimensional space as much as possible.
  • the camera device separately collects images of the calibration device at n shooting positions, and each shooting position is acquired m times at different angles, wherein the n is an integer greater than 1, and the m is greater than 1 The integer.
  • the camera device collects the timing device when the calibration device is horizontally placed on the horizontal surface, the vertical horizontal plane of the calibration device, and the lowering angle of the calibration device are placed at 60 degrees.
  • the calibration device may also be pressed at the above three positions. The calibration device is rotated once every 180 degrees, and the rotated calibration device is used. And each shooting position can be collected multiple times at different angles, for example: images are taken every 30 angles. In this way, six shooting positions can be obtained by the above method, and each shooting position is collected 12 times at different angles, and one image is collected 72 times.
  • step 901 may include:
  • n measurement posture information can be obtained.
  • Step 902 can include:
  • Step 903 can include:
  • the foregoing step 903 may include:
  • the attitude relationship between the camera coordinate system and the sensor coordinate system can be calculated by the above-described downtilt angle, the angle between the X-axis and the magnetic north, and the estimated value of the angle between the Y-axis and the magnetic north, and the above-described measurement and measurement posture information.
  • the above-described measurement and measurement posture information may include the above-described downtilt angle, the angle between the X-axis and the magnetic north, and the measured value of the angle between the Y-axis and the magnetic north.
  • the foregoing method may further include:
  • the calculating, according to the initial posture information, the first attitude relationship, and the second posture relationship, the calibration device coordinate system in the world coordinate system when the camera device collects the calibration device may include:
  • the calculating, according to the initial posture information, the first posture relationship, and the second posture relationship, the X axis and the magnetic north of the coordinate system coordinate system when the camera device collects the calibration device may include:
  • the Y axis and the magnetic north of the calibration device coordinate system when the camera device collects the calibration device according to the initial posture information, the first posture relationship, and the second posture relationship may include:
  • the three-dimensional vector of the calibration device coordinate system X-axis, Y-axis, and Z-axis in the world coordinate system is calculated at the j-th acquisition by the imaging device when the scaling device is in the i-th shooting position.
  • the above-mentioned downtilt angle, the angle between the X-axis and the magnetic north, and the estimated value of the angle between the Y-axis and the magnetic north can be obtained.
  • the step of acquiring the measurement posture information of the calibration device coordinate system in the world coordinate system when the calibration device is obtained at each shooting position may include:
  • step 904 can include: including:
  • the measured value of the downtilt angle, the measured value of the angle between the X-axis and the magnetic north of the calibration device coordinate system, and the measured value of the angle between the Y-axis and the magnetic north of the calibration device coordinate system are calculated by weighted distance and The calculation result is used as a comparison result between the estimated posture information and the measured posture information.
  • 904 can include:
  • the estimated pose information is compared to the measured pose information by a formula:
  • the T is a comparison result of the estimated posture information and the measured posture information
  • the i represents an i-th shooting position where the scaling device is located, where i is greater than or equal to 1, and is less than or Equal to the n
  • the j represents the jth acquisition of the camera device when the calibration device is located, wherein j is greater than or equal to 1, and less than or equal to the m
  • w 1 , w 2 and w 3 is a preset weight
  • the An estimated value of a downtilt angle of the scaling device when the calibration apparatus performs the jth acquisition by the imaging device at the i-th shooting position a measurement value of a downtilt angle of the calibration device when the scaling device is in an i-th shooting position
  • w 1 , w 2 and w 3 may be configured in a preset manner, for example, w 1 :w 2 :w 3 may be configured to be 100:1:1.
  • FIG. 10 is a schematic flowchart diagram of another method for calculating an attitude relationship according to an embodiment of the present invention. As shown in FIG. 10, the method includes the following steps:
  • the acquiring calibration device coordinate system is measured posture information in a world coordinate system when acquiring an image of the calibration device, wherein the measurement posture information is used to indicate that the calibration device coordinate system is in the world coordinate The orientation of the system.
  • the target formula can be as follows:
  • the solution obtained by the above formula is an attitude relationship between the camera system coordinate system and the sensor coordinate system calibration
  • the min() is a minimum value function
  • the i represents the ith shot of the calibration device.
  • a position where i is greater than or equal to 1, and less than or equal to n
  • the n is the number of shooting positions at which the scaling device is photographed
  • j represents the position of the calibration device
  • the imaging device is acquired j times, wherein j is greater than or equal to 1, and less than or equal to m
  • the m is the number of times the camera device collects at each shooting position
  • w 1 , w 2 , and w 3 are presets.
  • w 1 , w 2 and w 3 may be configured in a preset manner, for example, w 1 :w 2 :w 3 may be configured to be 100:1:1.
  • the above formula can be optimized, and the attitude relationship between the camera coordinate system and the sensor coordinate system calibration can be obtained.
  • optimization can be solved by using the Levenberg-Marquardt algorithm.
  • the Levenberg-Marquardt algorithm with different optimization parameters may be used to generate a plurality of candidates (or as a hypothesis) of the camera device coordinate system and the sensor coordinate system, and then according to the camera coordinate system and the sensor.
  • the attitude relationship of the coordinate system calculates the estimated attitude information of the calibration device coordinate system in the world coordinate system, and the estimated attitude information calculated by the attitude relationship between the camera coordinate system and the sensor coordinate system of a candidate When the difference (or the mean square error) of the measured attitude information is the smallest, the attitude relationship between the candidate camera coordinate system and the sensor coordinate system can be used as the attitude relationship between the camera coordinate system and the sensor coordinate system.
  • the parameter to be solved in the above formula is a rotation matrix R, so that the parameter to be solved can be transformed into a triple, so the unknown parameter of the above formula is a set of ternary vectors.
  • the transformation can be a Rodrigues transformation.
  • the camera device separately collects the calibration devices at n shooting positions, and each shooting position is acquired m times at different angles, wherein the n is an integer greater than or equal to 1, The m is an integer greater than or equal to 1;
  • the acquiring posture information of the calibration device coordinate system in the world coordinate system when the acquisition camera device acquires the calibration device may include:
  • the calculating the first attitude relationship between the world coordinate system and the sensor coordinate system when the camera device adopts the calibration device may include:
  • the calculating the second attitude relationship between the camera coordinate system and the calibration device coordinate system when the image capturing device acquires the image of the calibration device may include:
  • the calculating, according to the initial posture information, the first attitude relationship, and the second posture relationship, the calibration device coordinate system in the world coordinate system when the camera device collects the calibration device can include:
  • Calculating the according to the initial posture information, the first attitude relationship, and the second posture relationship An estimated value of an angle between a Y-axis and a magnetic north of the calibration device coordinate system when the imaging device acquires the calibration device.
  • the foregoing method may further include:
  • the calculating, according to the initial posture information, the first attitude relationship, and the second posture relationship, the calibration device coordinate system in the world coordinate system when the camera device collects the calibration device Estimates of the downtilt angle which may include:
  • the calculating, according to the initial posture information, the first posture relationship, and the second posture relationship, the X axis and the magnetic north of the coordinate system coordinate system when the camera device collects the calibration device Estimates of the included angles can include:
  • the Y axis and the magnetic north of the calibration device coordinate system when the camera device collects the calibration device according to the initial posture information, the first posture relationship, and the second posture relationship Estimates of the included angles can include:
  • the obtaining the measured attitude information of the calibration device coordinate system in the world coordinate system when the calibration device is obtained at each shooting position may include:
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

Landscapes

  • Engineering & Computer Science (AREA)
  • Manufacturing & Machinery (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Manipulator (AREA)
  • Studio Devices (AREA)
  • Length Measuring Devices By Optical Means (AREA)

Abstract

一种智能设备的姿态关系计算方法和智能设备,该方法可包括:获取摄像装置采集定标装置的图像时定标装置坐标系在世界坐标系的测量姿态信息(801);计算所述摄像装置采集所述定标装置的图像时世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置采集所述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系(802);获取所述摄像装置坐标系与传感器坐标系的初始姿态关系,并根据所述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息(803);基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系(804)。可以校准传感器坐标系与摄像装置坐标系的姿态关系。

Description

一种智能设备的姿态关系计算方法和智能设备 技术领域
本发明涉及通信领域,尤其涉及一种智能设备的姿态关系计算方法和智能设备。
背景技术
在姿态测量领域带有传感器和摄像装置的智能设备日益普及。在姿态测量的过程中,智能设备的传感器可以用来采集加速度、磁通等传感数据,智能设备的摄像装置可以用来采集图像或者视频等数据,为了对姿态测量中角度计算的精确性,如何校准智能设备中传感器坐标系与摄像装置坐标系的姿态关系是当前亟待解决的问题。
发明内容
本发明提供了一种智能设备的姿态关系计算方法和智能设备,可以校准智能设备中传感器坐标系与摄像装置坐标系的姿态关系。
第一方面,本发明提供一种智能设备,包括:获取单元、第一计算单元、第二计算单元和第三计算单元,其中:
所述获取单元,用于获取摄像装置采集定标装置的图像时所述定标装置坐标系在世界坐标系的测量姿态信息,其中,所述测量姿态信息用于表示所述定标装置坐标系在所述世界坐标系的朝向;
所述第一计算单元,用于计算所述摄像装置采集所述定标装置的图像时所述世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置采集所述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系;
所述第二计算单元,用于获取所述摄像装置坐标系与所述传感器坐标系的初始姿态关系,并根据所述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所 述世界坐标系中的估算姿态信息;
所述第三计算单元,用于基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系。
在第一方面的第一种可能的实现方式中,所述第三计算单元包括:
比较单元,用于将所述估算姿态信息与所述测量姿态信息进行比较;
计算子单元,用于当比较结果达到预先设定的用于校准姿态关系的比较结果时,将计算所述估算姿态信息时使用的所述摄像装置坐标系与传感器坐标系的姿态关系作为所述摄像装置坐标系与传感器坐标系校准的姿态关系。
结合在第一方面的第一种可能的实现方式,在第一方面的第二种可能的实现方式中,所述设备还包括:
生成单元,用于当比较结果未达到所述预先设定的比较结果时,基于预先设定的生成姿态关系的生成规则生成新的所述摄像装置坐标系与传感器坐标系的姿态关系;
第四计算单元,用于根据所述生成的姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息。
结合第一方面,在第一方面的第三种可能的实现方式中,所述第三计算单元包括:
通过如下公式计算所述摄像装置坐标系与传感器坐标系校准的姿态关系
Figure PCTCN2015072461-appb-000001
其中,上述公式求得的解为所述摄像装置坐标系与传感器坐标系校准的姿态关系,所述min()为最小值函数,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于n,所述n为所述定标装置在拍摄时所处的拍摄位置的数量,所述j表示所述定标装置所处各位置时所述摄像装置第j次采集,其中,j大于或者等于1,且小于或者等于m,所述m为每个拍摄位置所述摄像装置采集的次数,w1、w2和w3为预设的权值,所述
Figure PCTCN2015072461-appb-000002
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定 标装置的下倾角的估算值,所述
Figure PCTCN2015072461-appb-000003
表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
Figure PCTCN2015072461-appb-000004
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000005
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
Figure PCTCN2015072461-appb-000006
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000007
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
结合第一方面或者结合第一方面的第一种可能的实现方式或者结合第一方面的第二种可能的实现方式或者结合第一方面的第三种可能的实现方式,在第一方面的第四种可能的实现方式中,所述摄像装置分别采集处于n个拍摄位置的所述定标装置的图像,且每个拍摄位置以不同的角度采集m次,其中,所述n为大于或者等于1的整数,所述m为大于或者等于1的整数;
所述获取单元用于获取所述定标装置在各拍摄位置时所述定标装置坐标系在世界坐标系的测量姿态信息。
结合第一方面的第四种可能的实现方式,在第一方面的第五种可能的实现方式中,所述第二计算单元包括:
第一估算单元,用于根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系的下倾角的估算值;
第二估算单元,用于根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的X轴与磁北的夹角的估算值;
第三估算单元,用于根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的Y轴与磁北的夹角的估算值。
结合第一方面的第五种可能的实现方式,在第一方面的第六种可能的实现方式中,所述设备还包括:
第五计算单元,用于根据所述初始姿态信息、所述第一姿态关系和所述第 二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系与所述世界坐标系的第三姿态关系。
结合第一方面的第六种可能的实现方式,在第一方面的第七种可能的实现方式中,所述第一估算单元用于计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Z轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与重力方向的夹角为作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系在所述世界坐标系的下倾角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
结合第一方面的第六种可能的实现方式,在第一方面的第八种可能的实现方式中,所述第二估算单元用于计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的X轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
结合第一方面的第六种可能的实现方式,在第一方面的第九种可能的实现方式中,所述第三估算单元用于计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Y轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
结合第一方面的第四种可能的实现方式或者结合第一方面的第五种可能的实现方式或者结合第一方面的第六种可能的实现方式或者结合第一方面的第七种可能的实现方式或者结合第一方面的第八种可能的实现方式或者结合第一方面的第九种可能的实现方式,在第一方面的第十种可能的实现方式中,所述获取单元包括:
第一获取子单元,用于获取所述定标装置在各拍摄位置时所述定标装置坐 标系在所述世界坐标系的下倾角的测量值;
第二获取子单元,用于获取所述定标装置在各拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值;
第三获取子单元,用于获取所述定标装置在各拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
结合第一方面的第十种可能的实现方式,在第一方面的第十一种可能的实现方式中,所述比较单元用于将所述下倾角的估算值、所述定标装置坐标系的X轴与磁北的夹角的估算值和所述定标装置坐标系的Y轴与磁北的夹角的估算值分别与所述下倾角的测量值、所述定标装置坐标系的X轴与磁北的夹角的测量值和所述定标装置坐标系的Y轴与磁北的夹角的测量值进行加权距离计算,并将所述计算结果作为所述估算姿态信息与所述测量姿态信息的比较结果。
结合第一方面的第十种可能的实现方式,在第一方面的第十二种可能的实现方式中,所述比较单元用于通过如下公式计算将所述估算姿态信息与所述测量姿态信息进行比较:
Figure PCTCN2015072461-appb-000008
其中,所述T为所述估算姿态信息与所述测量姿态信息的比较结果,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于所述n,所述j表示所述定标装置所处各位置时所述摄像装置第j次采集,其中,j大于或者等于1,且小于或者等于所述m,w1、w2和w3为预设的权值,所述
Figure PCTCN2015072461-appb-000009
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置的下倾角的估算值,所述
Figure PCTCN2015072461-appb-000010
表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
Figure PCTCN2015072461-appb-000011
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000012
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
Figure PCTCN2015072461-appb-000013
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y 轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000014
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
第二方面,本发明提供一种智能设备,包括:处理器、网络接口、存储器、通信总线、摄像装置和传感器,其中,所述通信总线用于实现所述处理器、所述网络接口、所述存储器、所述摄像装置和所述传感器之间连接通信,所述处理器用于执行所述存储器中存储的程序;其中,所述程序包括:
获取所述摄像装置采集定标装置的图像时所述定标装置坐标系在世界坐标系的测量姿态信息,其中,所述测量姿态信息用于表示所述定标装置坐标系在所述世界坐标系的朝向;
计算所述摄像装置采集所述定标装置的图像时所述世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置采集所述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系;
获取所述摄像装置坐标系与所述传感器坐标系的初始姿态关系,并根据所述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息;
基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系。
在第二方面的第一种可能的实现方式中,所述处理器执行的基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系的程序,包括:
将所述估算姿态信息与所述测量姿态信息进行比较;
当比较结果达到预先设定的用于校准姿态关系的比较结果时,将计算所述估算姿态信息时使用的所述摄像装置坐标系与传感器坐标系的姿态关系作为所述摄像装置坐标系与传感器坐标系校准的姿态关系。
结合第二方面的第一种可能的实现方式,在第二方面的第二种可能的实现方式中,所述处理器执行的程序还包括:
当比较结果未达到所述预先设定的比较结果时,基于预先设定的生成姿态关系的生成规则生成新的所述摄像装置坐标系与传感器坐标系的姿态关系;
根据所述生成的姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息,并再次执行所述将所述估算姿态信息与所述测量姿态信息进行比较的步骤。
结合第二方面,在第二方面的第三种可能的实现方式中,所述处理器执行的基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系的程序,包括:
通过如下公式计算所述摄像装置坐标系与传感器坐标系校准的姿态关系
Figure PCTCN2015072461-appb-000015
其中,上述公式求得的解为所述摄像装置坐标系与传感器坐标系校准的姿态关系,所述min()为最小值函数,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于n,所述n为所述定标装置在拍摄时所处的拍摄位置的数量,所述j表示所述定标装置所处各位置时所述摄像装置第j次采集,其中,j大于或者等于1,且小于或者等于m,所述m为每个拍摄位置所述摄像装置采集的次数,w1、w2和w3为预设的权值,所述
Figure PCTCN2015072461-appb-000016
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置的下倾角的估算值,所述
Figure PCTCN2015072461-appb-000017
表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
Figure PCTCN2015072461-appb-000018
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000019
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
Figure PCTCN2015072461-appb-000020
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000021
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
结合第二方面或者第二方面的第一种可能的实现方式或者第二方面的第二种可能的实现方式或者第二方面的第三种可能的实现方式,在第二方面的第四种可能的实现方式中,所述摄像装置分别采集处于n个拍摄位置的所述定标 装置的图像,且每个拍摄位置以不同的角度采集m次,其中,所述n为大于或者等于1的整数,所述m为大于或者等于1的整数;
所述处理器执行的获取摄像装置采集定标装置时所述定标装置坐标系在世界坐标系的测量姿态信息的程序,包括:
获取所述定标装置在各拍摄位置时所述定标装置坐标系在世界坐标系的测量姿态信息。
结合第二方面的第四种可能的实现方式,在第二方面的第五种可能的实现方式中,所述处理器执行的根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息的程序,包括:
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系的下倾角的估算值;
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的X轴与磁北的夹角的估算值;
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的Y轴与磁北的夹角的估算值。
结合第二方面的第五种可能的实现方式,在第二方面的第六种可能的实现方式中,所述处理器执行的程序还包括:
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系与所述世界坐标系的第三姿态关系。
结合第二方面的第六种可能的实现方式,在第二方面的第七种可能的实现方式中,所述处理器执行的根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系的下倾角的估算值的程序,包括:
计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时 的所述第三姿态关系与所述定标装置坐标系的Z轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与重力方向的夹角为作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系在所述世界坐标系的下倾角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
结合第二方面的第六种可能的实现方式,在第二方面的第八种可能的实现方式中,所述处理器执行的根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的X轴与磁北的夹角的估算值的程序,包括:
计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的X轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
结合第二方面的第六种可能的实现方式,在第二方面的第九种可能的实现方式中,所述处理器执行的根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的Y轴与磁北的夹角的估算值的程序,包括:
计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Y轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
结合第二方面的第五种可能的实现方式或者第二方面的第六种可能的实现方式或者第二方面的第七种可能的实现方式或者第二方面的第八种可能的实现方式或者第二方面的第九种可能的实现方式,在第二方面的第十种可能的实现方式中,所述处理器执行的获取所述定标装置在各拍摄位置时所述定标装 置坐标系在世界坐标系的测量姿态信息的程序,包括:
获取所述定标装置在各拍摄位置时所述定标装置坐标系在所述世界坐标系的下倾角的测量值;
获取所述定标装置在各拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值;
获取所述定标装置在各拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
结合第二方面的第十种可能的实现方式,在第二方面的第十一种可能的实现方式中,所述处理器执行的将所述估算姿态信息与所述测量姿态信息进行比较的程序,包括:
将所述下倾角的估算值、所述定标装置坐标系的X轴与磁北的夹角的估算值和所述定标装置坐标系的Y轴与磁北的夹角的估算值分别与所述下倾角的测量值、所述定标装置坐标系的X轴与磁北的夹角的测量值和所述定标装置坐标系的Y轴与磁北的夹角的测量值进行加权距离计算,并将所述计算结果作为所述估算姿态信息与所述测量姿态信息的比较结果。
结合第二方面的第十种可能的实现方式,在第二方面的第十二种可能的实现方式中,所述处理器执行的将所述估算姿态信息与所述测量姿态信息进行比较的程序,包括:
通过如下公式计算将所述估算姿态信息与所述测量姿态信息进行比较:
Figure PCTCN2015072461-appb-000022
其中,所述T为所述估算姿态信息与所述测量姿态信息的比较结果,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于所述n,所述j表示所述定标装置所处各位置时所述摄像装置第j次采集,其中,j大于或者等于1,且小于或者等于所述m,w1、w2和w3为预设的权值,所述
Figure PCTCN2015072461-appb-000023
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置的下倾角的估算值,所述
Figure PCTCN2015072461-appb-000024
表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
Figure PCTCN2015072461-appb-000025
表示所述定标装置在 第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000026
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
Figure PCTCN2015072461-appb-000027
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000028
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
第三方面,本发明提供一种智能设备的姿态关系计算方法,其特征在于,包括:
获取摄像装置采集定标装置的图像时所述定标装置坐标系在世界坐标系的测量姿态信息,其中,所述测量姿态信息用于表示所述定标装置坐标系在所述世界坐标系的朝向;
计算所述摄像装置采集所述定标装置的图像时所述世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置采集所述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系;
获取所述摄像装置坐标系与所述传感器坐标系的初始姿态关系,并根据所述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息;
基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系。
在第三方面的第一种可能的实现方式中,所述基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系,包括:
将所述估算姿态信息与所述测量姿态信息进行比较;
当比较结果达到预先设定的用于校准姿态关系的比较结果时,将计算所述估算姿态信息时使用的所述摄像装置坐标系与传感器坐标系的姿态关系作为所述摄像装置坐标系与传感器坐标系校准的姿态关系。
结合第三方面的第一种可能的实现方式,在第三方面的第二种可能的实现方式中,所述方法还包括:
当比较结果未达到所述预先设定的比较结果时,基于预先设定的生成姿态关系的生成规则生成新的所述摄像装置坐标系与传感器坐标系的姿态关系;
根据所述生成的姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息,并再次执行所述将所述估算姿态信息与所述测量姿态信息进行比较的步骤。
结合第三方面,在第三方面的第三种可能的实现方式中,所述基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系,包括:
通过如下公式计算所述摄像装置坐标系与传感器坐标系校准的姿态关系
Figure PCTCN2015072461-appb-000029
其中,上述公式求得的解为所述摄像装置坐标系与传感器坐标系校准的姿态关系,所述min()为最小值函数,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于n,所述n为所述定标装置在拍摄时所处的拍摄位置的数量,所述j表示所述定标装置所处各位置时所述摄像装置第j次采集,其中,j大于或者等于1,且小于或者等于m,所述m为每个拍摄位置所述摄像装置采集的次数,w1、w2和w3为预设的权值,所述
Figure PCTCN2015072461-appb-000030
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置的下倾角的估算值,所述
Figure PCTCN2015072461-appb-000031
表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
Figure PCTCN2015072461-appb-000032
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000033
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
Figure PCTCN2015072461-appb-000034
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000035
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
结合第三方面的上述任一种可能的实现方式,在第三方面的第四种可能的 实现方式中,所述摄像装置分别采集处于n个拍摄位置的所述定标装置的图像,且每个拍摄位置以不同的角度采集m次,其中,所述n为大于或者等于1的整数,所述m为大于或者等于1的整数;
所述获取摄像装置采集定标装置时所述定标装置坐标系在世界坐标系的测量姿态信息,包括:
获取所述定标装置在各拍摄位置时所述定标装置坐标系在世界坐标系的测量姿态信息。
结合第三方面的第四种可能的实现方式,在第三方面的第五种可能的实现方式中,所述根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息,包括:
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系的下倾角的估算值;
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的X轴与磁北的夹角的估算值;
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的Y轴与磁北的夹角的估算值。
结合第三方面的第五种可能的实现方式,在第三方面的第六种可能的实现方式中,所述方法还包括:
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系与所述世界坐标系的第三姿态关系。
结合第三方面的第六种可能的实现方式,在第三方面的第七种可能的实现方式中,所述根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系的下倾角的估算值,包括:
计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Z轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与重力方向的夹角为作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系在所述世界坐标系的下倾角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
结合第三方面的第六种可能的实现方式,在第三方面的第八种可能的实现方式中,所述根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的X轴与磁北的夹角的估算值,包括:
计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的X轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
结合第三方面的第六种可能的实现方式,在第三方面的第九种可能的实现方式中,所述根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的Y轴与磁北的夹角的估算值,包括:
计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Y轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
结合第三方面的第五种可能的实现方式或者第三方面的第六种可能的实现方式或者第三方面的第七种可能的实现方式或者第三方面的第八种可能的实现方式或者第三方面的第九种可能的实现方式,在第三方面的第十种可能的 实现方式中,所述获取所述定标装置在各拍摄位置时所述定标装置坐标系在世界坐标系的测量姿态信息,包括:
获取所述定标装置在各拍摄位置时所述定标装置坐标系在所述世界坐标系的下倾角的测量值;
获取所述定标装置在各拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值;
获取所述定标装置在各拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
结合第三方面的第十种可能的实现方式,在第三方面的第十一种可能的实现方式中,所述将所述估算姿态信息与所述测量姿态信息进行比较,包括:
将所述下倾角的估算值、所述定标装置坐标系的X轴与磁北的夹角的估算值和所述定标装置坐标系的Y轴与磁北的夹角的估算值分别与所述下倾角的测量值、所述定标装置坐标系的X轴与磁北的夹角的测量值和所述定标装置坐标系的Y轴与磁北的夹角的测量值进行加权距离计算,并将所述计算结果作为所述估算姿态信息与所述测量姿态信息的比较结果。
结合第三方面的第十种可能的实现方式,在第三方面的第十二种可能的实现方式中,所述将所述估算姿态信息与所述测量姿态信息进行比较,包括:
通过如下公式计算将所述估算姿态信息与所述测量姿态信息进行比较:
Figure PCTCN2015072461-appb-000036
其中,所述T为所述估算姿态信息与所述测量姿态信息的比较结果,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于所述n,所述j表示所述定标装置所处各位置时所述摄像装置第j次采集,其中,j大于或者等于1,且小于或者等于所述m,w1、w2和w3为预设的权值,所述
Figure PCTCN2015072461-appb-000037
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置的下倾角的估算值,所述
Figure PCTCN2015072461-appb-000038
表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
Figure PCTCN2015072461-appb-000039
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X 轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000040
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
Figure PCTCN2015072461-appb-000041
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000042
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
上述技术方案中,获取摄像装置采集定标装置的图像时所述定标装置坐标系在世界坐标系的测量姿态信息,其中,所述测量姿态信息用于表示所述定标装置坐标系在所述世界坐标系的朝向;计算所述摄像装置采集所述定标装置的图像时所述世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置采集所述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系;获取所述摄像装置坐标系与所述传感器坐标系的初始姿态关系,并根据所述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息;基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系。从而可以实现校准智能设备中传感器坐标系与摄像装置坐标系的姿态关系。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1是本发明实施例提供的一种智能设备的结构示意图;
图2是本发明实施例提供的另一种智能设备的结构示意图;
图3是本发明实施例提供的另一种智能设备的结构示意图;
图4是本发明实施例提供的另一种智能设备的结构示意图;
图5是本发明实施例提供的另一种智能设备的结构示意图;
图6是本发明实施例提供的另一种智能设备的结构示意图;
图7是本发明实施例提供的另一种智能设备的结构示意图;
图8是本发明实施例提供的一种智能设备的姿态关系计算方法的流程示意图;
图9是本发明实施例提供的另一种智能设备的姿态关系计算方法的流程示意图;
图10是本发明实施例提供的另一种智能设备的姿态关系计算方法的流程示意图。
具体实施方式
下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。
请参阅图1,图1是本发明实施例提供的一种智能设备的结构示意图,如图1所示,包括:获取单元11、第一计算单元12、第二计算单元13和第三计算单元14,其中:
获取单元11,用于获取摄像装置采集定标装置的图像时所述定标装置坐标系在世界坐标系的测量姿态信息,其中,所述测量姿态信息用于表示所述定标装置坐标系在所述世界坐标系的朝向。
可选的,上述测量姿态信息可以是倾角仪测量定标装置坐标系在世界坐标系的朝向信息作为上述测量姿态信息,或者可以通过姿态测量仪测量定标装置坐标系在世界坐标系的姿态信息作为上述测量姿态信息。另外,由于上述测量姿态信息是通过测量姿态的专门的工具进行测量的,这样上述测量姿态信息可以很精准地表示出定标装置坐标系在世界坐标系的姿态信息,或者可理解为上述测量姿态信息是高精准的姿态信息。另外,智能设备可以是通过智能设备自身携带的工具测量出上述测量姿态信息。或者可以是接收用户输入的上述测量状态信息,或者可以是接收其他设备发送的上述测量状态信息等等,此处不作限定。
第一计算单元12,用于计算所述摄像装置采集所述定标装置的图像时所述世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置采集所 述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系。
可选的,第一计算单元12可以是根据上述传感器采用的传感数据计算世界坐标系与传感器坐标系的第一姿态关系。例如:通过传感器采集的加速度和磁通数据计算世界坐标系与传感器坐标系的第一姿态关系。
可选的,上述定标装置可以是放置在某一特定位置的物品,例如:放置在与水平面平行的平面上的物品,或者放置在与水平面成60度的平面上的物品等。通过摄像装置采集到的上述定标装置的图像计算摄像装置坐标系与所述定标装置坐标系的第二姿态关系。另外,上述定标装置可以是定标板。
第二计算单元13,用于获取所述摄像装置坐标系与所述传感器坐标系的初始姿态关系,并根据所述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息。
可选的,上述初始姿态关系可以是用户输入摄像装置坐标系与传感器坐标系之间的姿态关系,或者可以是接收其他设备发送的摄像装置坐标系与传感器坐标系之间的姿态关系。另外,这里的摄像装置坐标系与传感器坐标系之间的姿态关系可以理解为假设的姿态关系,例如:可以是摄像装置坐标系与传感器坐标系之间的理想姿态关系,如单位矩阵表示的无偏移旋转关系。
当世界坐标系与传感器坐标系的第一姿态关系和所述摄像装置坐标系与传感器坐标系的初始姿态关系已知后,就可以得到世界坐标系与所述摄像装置坐标系之间的姿态关系,且由于所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系已知后,就可以得到世界坐标系与所述定标装置坐标系之间的姿态关系,即可以得到定标装置坐标系在所述世界坐标系中的估算姿态信息。
第三计算单元14,用于基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系。
由于上述估算姿态信息和测量姿态信息都是反应定标装置坐标系在世界坐标系中的空间姿态,而测量姿态信息是实际测量出来的,即该测量姿态信息表示定标装置坐标系在世界坐标系中高准确度的空间姿态,而估算姿态信息是基于上述初始姿态关系计算得到的。这样第三计算单元14就可以在当测量姿 态信息与估算状态信息相同或者相似时,就可以将上述初始姿态关系作为摄像装置坐标系与传感器坐标系校准的姿态关系;或者当测量姿态信息与估算状态信息存在一定差别时,将上述初始姿态关系进行特定调整后,再将调整后的姿态关系作为摄像装置坐标系与传感器坐标系校准的姿态关系;或者当测量姿态信息与估算状态信息存在一定差别时,将上述初始姿态关系进行调整后,再根据调整后的姿态关系生成定标装置坐标系在所述世界坐标系中的估算姿态信息,再根据该估算姿态信息与测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系。
可选的,上述能设备可以包括但不限于:平板电脑、手机、电子阅读器、遥控器、个人计算机(Personal Computer,PC)、笔记本电脑、车载设备、网络电视、可穿戴设备等用户设备(user equipment,UE)。
上述技术方案中,获取摄像装置采集定标装置的图像时所述定标装置坐标系在世界坐标系的测量姿态信息,其中,所述测量姿态信息用于表示所述定标装置坐标系在所述世界坐标系的朝向;计算所述摄像装置采集所述定标装置的图像时所述世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置采集所述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系;获取所述摄像装置坐标系与所述传感器坐标系的初始姿态关系,并根据所述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息;基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系。从而可以实现校准智能设备中传感器坐标系与摄像装置坐标系的姿态关系。
请参阅图2,图2是本发明实施例提供的另一种智能设备的结构示意图,如图2所示,包括:获取单元21、第一计算单元22、第二计算单元23和第三计算单元24,第三计算单元24包括:比较单元241和计算子单元242,其中:
获取单元21,用于获取摄像装置采集定标装置的图像时所述定标装置坐标系在世界坐标系的测量姿态信息,其中,所述测量姿态信息用于表示所述定标装置坐标系在所述世界坐标系的朝向。
第一计算单元22,用于计算所述摄像装置采集所述定标装置的图像时所述世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置采集所述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系。
可选的,第一计算单元22可以是采用传感器数据预处理的方式对传感器采集的加速度和磁通预数据进行处理,以得到更加精确的传感器数据,之后再通过传感器数据计算世界坐标系与传感器坐标系的第一姿态关系。
具体可以通过如下公式计算出世界坐标系与传感器坐标系的第一姿态关系:
Figure PCTCN2015072461-appb-000043
其中,所述
Figure PCTCN2015072461-appb-000044
为世界坐标系与传感器坐标系的第一姿态关系,其中,ψ,θ,φ分别表示偏航角(yaw)、俯仰角(pitch)和滚转角(roll)角度,Rx(φ)Ry(θ)Rz(ψ)可以通过如下公式表示
Figure PCTCN2015072461-appb-000045
Figure PCTCN2015072461-appb-000046
Figure PCTCN2015072461-appb-000047
这样通过上述公式就可以计算世界坐标系与传感器坐标系的第一姿态关系,具体可以是得到一个旋转矩阵。
可选的,第一计算单元22可以是通过摄像装置拍照的方式计算出所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系。此处为公知的技术特征,不作详细说明。
第二计算单元23,用于获取所述摄像装置坐标系与所述传感器坐标系的初始姿态关系,并根据所述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所 述世界坐标系中的估算姿态信息。
比较单元241,用于将所述估算姿态信息与所述测量姿态信息进行比较。
计算子单元242,用于当比较结果达到预先设定的用于校准姿态关系的比较结果时,将计算所述估算姿态信息时使用的所述摄像装置坐标系与传感器坐标系的姿态关系作为所述摄像装置坐标系与传感器坐标系校准的姿态关系。
由于世界坐标系与传感器坐标系的第一姿态关系、摄像装置坐标系与所述定标装置坐标系的第二姿态关系、摄像装置坐标系与传感器坐标系的姿态关系以及定标装置坐标系在所述世界坐标系中的姿态信息这四者之间的关系在同一时刻是不变的,这样通过一个假设的摄像装置坐标系与传感器坐标系的姿态关系就可以得到一个定标装置坐标系在所述世界坐标系中的估算姿态信息,当该估算姿态信息与测量姿态信息相同或者相似时,就可以确定上述假设的摄像装置坐标系与传感器坐标系的姿态关系为摄像装置坐标系与传感器坐标系校准的姿态关系。
可选的,上述预先设定的用于校准姿态关系的比较结果可以是估算姿态信息与所述测量姿态信息相同,或者估算姿态信息与所述测量姿态信息的相似度大于或者等于特定阈值,或者估算姿态信息与所述测量姿态信息之间的差异小于或者等于特定阈值,其中,上述特定阈值可以是智能设备默认或者预先设定的阈值,例如:98%或者99%等。
可选的,如图3所示,上述设备还可以包括:
生成单元25,用于当比较结果未达到所述预先设定的比较结果时,基于预先设定的生成姿态关系的生成规则生成新的所述摄像装置坐标系与传感器坐标系的姿态关系;
第四计算单元26,用于根据所述生成的姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息。
上述比较结果未达到所述预先设定的比较结果可以是表示上述估算姿态信息与测量姿态信息的差别较大,这时就可以根据预设规则生成新的姿态关系。例如:根据上述初始姿态关系计算的估算姿态信息与测量姿态信息差别较大时,就说明上述初始姿态关系并非摄像装置坐标系与传感器坐标系高准确度 的姿态关系,或者上述初始姿态关系的精度比较低,从而需要基于该初始姿态关系生成新的姿态关系。
可选的,上述生成规则可以包括:
根据上述比较结果生成新的姿态关系。
例如:当比较结果表示估算姿态信息与测量姿态信息的差异较大时,就可以将计算所述估算姿态信息时使用的所述摄像装置坐标系与传感器坐标系的姿态关系进行大幅度的调整;例如:当使用上述初始姿态关系计算的估算姿态信息与测量姿态信息的差异较大时,就可以将该初始姿态关系进行大幅度的调整。
当比较结果表示估算姿态信息与测量姿态信息的差异较小时,可以将计算所述估算姿态信息时使用的所述摄像装置坐标系与传感器坐标系的姿态关系进行小幅度的调整。例如:当使用上述初始姿态关系计算的估算姿态信息与测量姿态信息的差异较小时,就可以将该初始姿态关系进行小幅度的调整。
可选的,上述预设规则还可以包括:
在预设范围内随机生成所述摄像装置坐标系与传感器坐标系的姿态关系。
可选的,上述生成规则还可以包括:
通过Levenberg-Marquardt算法生成所述摄像装置坐标系与传感器坐标系的姿态关系。
可选的,上述摄像装置坐标系与传感器坐标系的姿态关系具体可以为一个旋转矩阵R,该旋转矩阵可以变换为一个三元组,其中,该变换可以是用Rodrigues变换。
另外,当第四计算单元26计算完估算姿态信息后,计算子单元242就可以使用第四计算单元26计算的估算姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系
可选的,本实施例中为了得到更加准确和精确的效果,上述定标装置的拍摄位置尽可能覆盖整个三维空间。例如:摄像装置分别采集处于n个拍摄位置的所述定标装置的图像,且每个拍摄位置以不同的角度采集m次,其中,所述n为大于1的整数,所述m为大于1的整数。例如:摄像装置对定标装置水平放置在水平面上、定标装置垂直水平面放置和定标装置下倾角60度放置 时定时装置进行采集,另外,在上述三个位置上定标装置还可以是按每隔180度旋转一次定标装置,再对旋转后的定标装置进行采用。且每个拍摄位置可以不同的角度采集多次,例如:每隔30角度采集一次图像。这样通过上述方式就可以得到6个拍摄位置,以及每个拍摄位置以不同的角度采集12次,一个采集72张图像。
该实施方式中,获取单元21可以用于获取所述定标装置在各拍摄位置时所述定标装置坐标系在世界坐标系的测量姿态信息。
该实施方式中,如图4所示,第二计算单元23可以包括:
第一估算单元231,用于根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系的下倾角的估算值;
第二估算单元232,用于根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的X轴与磁北的夹角的估算值;
第三估算单元233,用于根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的Y轴与磁北的夹角的估算值。
该实施方式中可以通过上述下倾角、X轴与磁北的夹角和Y轴与磁北的夹角的估算值,以及上述测量测量姿态信息计算摄像装置坐标系与传感器坐标系的姿态关系。另外,上述测量测量姿态信息可以包括上述下倾角、X轴与磁北的夹角和Y轴与磁北的夹角的测量值。
可选的,上述设备还可以包括:
第五计算单元27,用于根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系与所述世界坐标系的第三姿态关系。
例如:可以通过如下公式计算上述第三姿态关系:
Figure PCTCN2015072461-appb-000048
其中,所述
Figure PCTCN2015072461-appb-000049
为所述定标装置在第i个拍摄位置时所述摄像装置进 行的第j次采集时定标装置与世界坐标系的第三姿态关系,
Figure PCTCN2015072461-appb-000050
为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时世界坐标系与传感器坐标系的第一姿态关系,
Figure PCTCN2015072461-appb-000051
为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时摄像装置坐标系与传感器坐标系的姿态关系,该姿态关系可以为假设的姿态关系,例如:上述初始姿态关系,
Figure PCTCN2015072461-appb-000052
为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时摄像装置坐标系与所述定标装置坐标系的第二姿态关系,其中,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
可选的,第一估算单元231可以用于计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Z轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与重力方向的夹角为作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系在所述世界坐标系的下倾角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
可选的,第二估算单元232可以用于计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的X轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
可选的,第三估算单元231可以用于计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Y轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
上述对下倾角、X轴与磁北的夹角和Y轴与磁北的夹角的估算值可以通过如下公式计算:
Figure PCTCN2015072461-appb-000053
Figure PCTCN2015072461-appb-000054
Figure PCTCN2015072461-appb-000055
Figure PCTCN2015072461-appb-000056
Figure PCTCN2015072461-appb-000057
Figure PCTCN2015072461-appb-000058
Figure PCTCN2015072461-appb-000059
其中,上述
Figure PCTCN2015072461-appb-000060
分别表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时计算得到定标装置坐标系X轴、Y轴和Z轴在世界坐标系下的3维向量。
Figure PCTCN2015072461-appb-000061
Figure PCTCN2015072461-appb-000062
分别表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时定标装置坐标系的X轴、Y轴和Z轴在定标装置坐标系的3维向量,
Figure PCTCN2015072461-appb-000063
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时定标装置坐标系在所述世界坐标系的下倾角的估算值,
Figure PCTCN2015072461-appb-000064
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,
Figure PCTCN2015072461-appb-000065
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,上述
Figure PCTCN2015072461-appb-000066
表示
Figure PCTCN2015072461-appb-000067
与重力方向的夹角,
Figure PCTCN2015072461-appb-000068
表示
Figure PCTCN2015072461-appb-000069
投影到水平面上和磁北方向的夹角,
Figure PCTCN2015072461-appb-000070
表示
Figure PCTCN2015072461-appb-000071
投影到水平面上和磁北方向的夹角。
通过上述公式就可以得到上述下倾角、X轴与磁北的夹角和Y轴与磁北的夹角的估算值。
可选的,如图5所示,获取单元21可以包括:
第一获取子单元211,用于获取所述定标装置在各拍摄位置时所述定标装置坐标系在所述世界坐标系的下倾角的测量值;
第二获取子单元212,用于获取所述定标装置在各拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值;
第三获取子单元213,用于获取所述定标装置在各拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
比较单元241可以用于将所述下倾角的估算值、所述定标装置坐标系的X 轴与磁北的夹角的估算值和所述定标装置坐标系的Y轴与磁北的夹角的估算值分别与所述下倾角的测量值、所述定标装置坐标系的X轴与磁北的夹角的测量值和所述定标装置坐标系的Y轴与磁北的夹角的测量值进行加权距离计算,并将所述计算结果作为所述估算姿态信息与所述测量姿态信息的比较结果。
例如:比较单元241可以通过如下公式计算将所述估算姿态信息与所述测量姿态信息进行比较:
Figure PCTCN2015072461-appb-000072
其中,所述T为所述估算姿态信息与所述测量姿态信息的比较结果,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于所述n,所述j表示所述定标装置所处各位置时所述摄像装置第j次采集,其中,j大于或者等于1,且小于或者等于所述m,w1、w2和w3为预设的权值,所述
Figure PCTCN2015072461-appb-000073
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置的下倾角的估算值,所述
Figure PCTCN2015072461-appb-000074
表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
Figure PCTCN2015072461-appb-000075
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000076
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
Figure PCTCN2015072461-appb-000077
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000078
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
其中,上述w1、w2和w3可以为预设配置的,例如:将w1:w2:w3可以配置为100:1:1。
本实施例,在图1所示的实施例的基础上增加了多种可选的实施方式,且都可以实现校准智能设备中传感器坐标系与摄像装置坐标系的姿态关系。
请参阅图6,图6是本发明实施例提供的另一种智能设备的结构示意图, 如图6所示,包括:获取单元61、第一计算单元62、第二计算单元63和第三计算单元64,其中:
获取单元61,用于获取摄像装置采集定标装置的图像时所述定标装置坐标系在世界坐标系的测量姿态信息,其中,所述测量姿态信息用于表示所述定标装置坐标系在所述世界坐标系的朝向。
第一计算单元62,用于计算所述摄像装置采集所述定标装置的图像时所述世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置采集所述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系。
第二计算单元63,用于获取所述摄像装置坐标系与所述传感器坐标系的初始姿态关系,并根据所述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息。
第三计算单元64,用于基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系。
通过目标公式计算所述摄像装置坐标系与传感器坐标系校准的姿态关系。
其中,目标公式可以如下:
Figure PCTCN2015072461-appb-000079
其中,上述公式求得的解为所述摄像装置坐标系与传感器坐标系校准的姿态关系,所述min()为最小值函数,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于n,所述n为所述定标装置在拍摄时所处的拍摄位置的数量,所述j表示所述定标装置所处各位置时所述摄像装置第j次采集,其中,j大于或者等于1,且小于或者等于m,所述m为每个拍摄位置所述摄像装置采集的次数,w1、w2和w3为预设的权值,所述
Figure PCTCN2015072461-appb-000080
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置的下倾角的估算值,所述
Figure PCTCN2015072461-appb-000081
表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
Figure PCTCN2015072461-appb-000082
表示所述定标装置在第i个拍摄位置时 所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000083
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
Figure PCTCN2015072461-appb-000084
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000085
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
其中,上述w1、w2和w3可以为预设配置的,例如:将w1:w2:w3可以配置为100:1:1。
可选的,可以是对上述公式进行优化求解,就可以得到摄像装置坐标系与传感器坐标系校准的姿态关系。例如:可以通过用Levenberg-Marquardt算法进行优化求解。具体可以是通过带有不同优化参数的Levenberg-Marquardt算法生成多个候选(或者理解为假设)的所述摄像装置坐标系与传感器坐标系的姿态关系,再根据该所述摄像装置坐标系与传感器坐标系的姿态关系计算所述定标装置坐标系在所述世界坐标系中的估算姿态信息,当某一个候选的所述摄像装置坐标系与传感器坐标系的姿态关系计算出的上述估算姿态信息与测量姿态信息的差异(或者均方误差)最小时,该候选的所述摄像装置坐标系与传感器坐标系的姿态关系就可以作为所述摄像装置坐标系与传感器坐标系校准的姿态关系。
其中,上述公式中的待求解参数为一个旋转矩阵R,这样可以将待求解参数变换为一个三元组,故上述公式的未知参数是一组三元向量。其中,该变换可以是用Rodrigues变换。
本实施例,在图1所示的实施例的基础上增加了多种可选的实施方式,且都可以实现校准智能设备中传感器坐标系与摄像装置坐标系的姿态关系。
请参阅图7,图7是本发明实施例提供的另一种智能设备的结构示意图,如图7所示,包括:处理器71、网络接口72、存储器73、通信总线74、摄像装置75和传感器76,其中,所述通信总线74用于实现所述处理器71、所述网络接口72、所述存储器73、所述摄像装置74和所述传感器75之间连接通信,所述处理器71用于执行所述存储器73中存储的程序;其中,所述程序包 括:
获取所述摄像装置75采集定标装置的图像时所述定标装置坐标系在世界坐标系的测量姿态信息,其中,所述测量姿态信息用于表示所述定标装置坐标系在所述世界坐标系的朝向;
计算所述摄像装置75采集所述定标装置的图像时所述世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置75采集所述定标装置的图像时所述摄像装置75坐标系与所述定标装置坐标系的第二姿态关系;
获取所述摄像装置75坐标系与所述传感器坐标系的初始姿态关系,并根据所述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置75采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息;
基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置75坐标系与传感器坐标系校准的姿态关系。
可选的,处理器71执行的基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置75坐标系与传感器坐标系校准的姿态关系的程序,可以包括:
将所述估算姿态信息与所述测量姿态信息进行比较;
当比较结果达到预先设定的用于校准姿态关系的比较结果时,将计算所述估算姿态信息时使用的所述摄像装置75坐标系与传感器坐标系的姿态关系作为所述摄像装置75坐标系与传感器坐标系校准的姿态关系。
可选的,处理器71执行的程序还可以包括:
当比较结果未达到所述预先设定的比较结果时,基于预先设定的生成姿态关系的生成规则生成新的所述摄像装置75坐标系与传感器坐标系的姿态关系;
根据所述生成的姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置75采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息,并再次执行所述将所述估算姿态信息与所述测量姿态信息进行比较的步骤。
可选的,处理器71执行的基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置75坐标系与传感器坐标系校准的姿态关系的程序,可以包括:
通过如下公式计算所述摄像装置75坐标系与传感器坐标系校准的姿态关系
Figure PCTCN2015072461-appb-000086
其中,上述公式求得的解为所述摄像装置75坐标系与传感器坐标系校准的姿态关系,所述min()为最小值函数,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于n,所述n为所述定标装置在拍摄时所处的拍摄位置的数量,所述j表示所述定标装置所处各位置时所述摄像装置75第j次采集,其中,j大于或者等于1,且小于或者等于m,所述m为每个拍摄位置所述摄像装置75采集的次数,w1、w2和w3为预设的权值,所述
Figure PCTCN2015072461-appb-000087
表示所述定标装置在第i个拍摄位置时所述摄像装置75进行的第j次采集时所述定标装置的下倾角的估算值,所述
Figure PCTCN2015072461-appb-000088
表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
Figure PCTCN2015072461-appb-000089
表示所述定标装置在第i个拍摄位置时所述摄像装置75进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000090
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
Figure PCTCN2015072461-appb-000091
表示所述定标装置在第i个拍摄位置时所述摄像装置75进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000092
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
可选的,所述摄像装置75分别采集处于n个拍摄位置的所述定标装置的图像,且每个拍摄位置以不同的角度采集m次,其中,所述n为大于或者等于1的整数,所述m为大于或者等于1的整数;
处理器701执行的获取摄像装置75采集定标装置时所述定标装置坐标系在世界坐标系的测量姿态信息的程序,可以包括:
获取所述定标装置在各拍摄位置时所述定标装置坐标系在世界坐标系的测量姿态信息。
可选的,处理器71执行的根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置75采集所述定标装置的图像时所述定标 装置坐标系在所述世界坐标系中的估算姿态信息的程序,可以包括:
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置75采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系的下倾角的估算值;
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置75采集所述定标装置的图像时所述定标装置坐标系的X轴与磁北的夹角的估算值;
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置75采集所述定标装置的图像时所述定标装置坐标系的Y轴与磁北的夹角的估算值。
可选的,处理器71执行的程序还可以包括:
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置75采集所述定标装置的图像时所述定标装置坐标系与所述世界坐标系的第三姿态关系。
可选的,处理器71执行的根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置75采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系的下倾角的估算值的程序,可以包括:
计算所述定标装置在第i个拍摄位置时所述摄像装置75进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Z轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与重力方向的夹角为作为所述定标装置在第i个拍摄位置时所述摄像装置75进行的第j次采集时所述定标装置坐标系在所述世界坐标系的下倾角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
可选的,处理器71执行的根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置75采集所述定标装置的图像时所述定标装置坐标系的X轴与磁北的夹角的估算值的程序,可以包括:
计算所述定标装置在第i个拍摄位置时所述摄像装置75进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的X轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个 拍摄位置时所述摄像装置75进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
可选的,处理器71执行的根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置75采集所述定标装置的图像时所述定标装置坐标系的Y轴与磁北的夹角的估算值的程序,可以包括:
计算所述定标装置在第i个拍摄位置时所述摄像装置75进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Y轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置75进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
可选的,处理器71执行的获取所述定标装置在各拍摄位置时所述定标装置坐标系在世界坐标系的测量姿态信息的程序,可以包括:
获取所述定标装置在各拍摄位置时所述定标装置坐标系在所述世界坐标系的下倾角的测量值;
获取所述定标装置在各拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值;
获取所述定标装置在各拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
可选的,处理器71执行的将所述估算姿态信息与所述测量姿态信息进行比较的程序,可以包括:
将所述下倾角的估算值、所述定标装置坐标系的X轴与磁北的夹角的估算值和所述定标装置坐标系的Y轴与磁北的夹角的估算值分别与所述下倾角的测量值、所述定标装置坐标系的X轴与磁北的夹角的测量值和所述定标装置坐标系的Y轴与磁北的夹角的测量值进行加权距离计算,并将所述计算结果作为所述估算姿态信息与所述测量姿态信息的比较结果。
可选的,处理器71执行的将所述估算姿态信息与所述测量姿态信息进行比较的程序,可以包括:
通过如下公式计算将所述估算姿态信息与所述测量姿态信息进行比较:
Figure PCTCN2015072461-appb-000093
其中,所述T为所述估算姿态信息与所述测量姿态信息的比较结果,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于所述n,所述j表示所述定标装置所处各位置时所述摄像装置75第j次采集,其中,j大于或者等于1,且小于或者等于所述m,w1、w2和w3为预设的权值,所述
Figure PCTCN2015072461-appb-000094
表示所述定标装置在第i个拍摄位置时所述摄像装置75进行的第j次采集时所述定标装置的下倾角的估算值,所述
Figure PCTCN2015072461-appb-000095
表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
Figure PCTCN2015072461-appb-000096
表示所述定标装置在第i个拍摄位置时所述摄像装置75进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000097
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
Figure PCTCN2015072461-appb-000098
表示所述定标装置在第i个拍摄位置时所述摄像装置75进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000099
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
可选的,上述智能设备可以包括但不限于:平板电脑、手机、电子阅读器、遥控器、PC、笔记本电脑、车载设备、网络电视、可穿戴设备等UE。
上述技术方案中,获取摄像装置采集定标装置的图像时所述定标装置坐标系在世界坐标系的测量姿态信息,其中,所述测量姿态信息用于表示所述定标装置坐标系在所述世界坐标系的朝向;计算所述摄像装置采集所述定标装置的图像时所述世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置采集所述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系;获取所述摄像装置坐标系与所述传感器坐标系的初始姿态关系,并根据所述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息;基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系。从而可以实现校准智能设备中 传感器坐标系与摄像装置坐标系的姿态关系。
请参阅图8,图8是本发明实施例提供的一种智能设备的姿态关系计算方法的流程示意图,如图8所示,包括以下步骤:
801、获取摄像装置采集定标装置的图像时所述定标装置坐标系在世界坐标系的测量姿态信息,其中,所述测量姿态信息用于表示所述定标装置坐标系在所述世界坐标系的朝向。
可选的,上述测量姿态信息可以是倾角仪测量定标装置坐标系在世界坐标系的朝向信息作为上述测量姿态信息,或者可以通过姿态测量仪测量定标装置坐标系在世界坐标系的姿态信息作为上述测量姿态信息。另外,由于上述测量姿态信息是通过测量姿态的专门的工具进行测量的,这样上述测量姿态信息可以很精准地表示出定标装置坐标系在世界坐标系的姿态信息,或者可理解为上述测量姿态信息是高精准的姿态信息。另外,智能设备可以是通过智能设备自身携带的工具测量出上述测量姿态信息。或者可以是接收用户输入的上述测量状态信息,或者可以是接收其他设备发送的上述测量状态信息等等,此处不作限定。
802、计算所述摄像装置采集所述定标装置的图像时世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置采集所述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系。
可选的,步骤802可以是根据上述传感器采用的传感数据计算世界坐标系与传感器坐标系的第一姿态关系。例如:通过传感器采集的加速度和磁通数据计算世界坐标系与传感器坐标系的第一姿态关系。
可选的,上述定标装置可以是放置在某一特定位置的物品,例如:放置在与水平面平行的平面上的物品,或者放置在与水平面成60度的平面上的物品等。通过摄像装置采集到的上述定标装置的图像计算摄像装置坐标系与所述定标装置坐标系的第二姿态关系。另外,上述定标装置可以是定标板。
803、获取所述摄像装置坐标系与传感器坐标系的初始姿态关系,并根据所述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算 姿态信息。
可选的,上述初始姿态关系可以是用户输入摄像装置坐标系与传感器坐标系之间的姿态关系,或者可以是接收其他设备发送的摄像装置坐标系与传感器坐标系之间的姿态关系。另外,这里的摄像装置坐标系与传感器坐标系之间的姿态关系可以理解为假设的姿态关系,例如:可以是摄像装置坐标系与传感器坐标系之间的理想姿态关系,如单位矩阵表示的无偏移旋转关系。
当世界坐标系与传感器坐标系的第一姿态关系和所述摄像装置坐标系与传感器坐标系的初始姿态关系已知后,就可以得到世界坐标系与所述摄像装置坐标系之间的姿态关系,且由于所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系已知后,就可以得到世界坐标系与所述定标装置坐标系之间的姿态关系,即可以得到定标装置坐标系在所述世界坐标系中的估算姿态信息。
804、基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系。
由于上述估算姿态信息和测量姿态信息都是反应定标装置坐标系在世界坐标系中的空间姿态,而测量姿态信息是实际测量出来的,即该测量姿态信息表示定标装置坐标系在世界坐标系中高准确度的空间姿态,而估算姿态信息是基于上述初始姿态关系计算得到的。这样步骤804就可以在当测量姿态信息与估算状态信息相同或者相似时,就可以将上述初始姿态关系作为摄像装置坐标系与传感器坐标系校准的姿态关系;或者当测量姿态信息与估算状态信息存在一定差别时,将上述初始姿态关系进行特定调整后,再将调整后的姿态关系作为摄像装置坐标系与传感器坐标系校准的姿态关系;或者当测量姿态信息与估算状态信息存在一定差别时,将上述初始姿态关系进行调整后,再根据调整后的姿态关系生成定标装置坐标系在所述世界坐标系中的估算姿态信息,再根据该估算姿态信息与测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系。
可选的,上述方法可以应用于智能设备,具体可以是单个智能设备独立实现,或者可以是多个设备联合一起实现。其中,该智能设备可以包括但不限于:平板电脑、手机、电子阅读器、遥控器、PC、笔记本电脑、车载设备、网络电视、可穿戴设备等UE。
上述技术方案中,获取摄像装置采集定标装置的图像时所述定标装置坐标系在世界坐标系的测量姿态信息,其中,所述测量姿态信息用于表示所述定标装置坐标系在所述世界坐标系的朝向;计算所述摄像装置采集所述定标装置的图像时所述世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置采集所述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系;获取所述摄像装置坐标系与所述传感器坐标系的初始姿态关系,并根据所述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息;基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系。从而可以实现校准智能设备中传感器坐标系与摄像装置坐标系的姿态关系。
请参阅图9,图9是本发明实施例提供的另一种姿态关系计算方法的流程示意图,如图9所示,包括以下步骤:
901、获取摄像装置采集定标装置的图像时所述定标装置坐标系在世界坐标系的测量姿态信息,其中,所述测量姿态信息用于表示所述定标装置坐标系在所述世界坐标系的朝向。
902、计算所述摄像装置采集所述定标装置的图像时世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置采集所述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系。
可选的,步骤902可以是采用传感器数据预处理的方式对传感器采集的加速度和磁通数据进行预处理,以得到更加精确的传感器数据,之后再通过传感器数据计算世界坐标系与传感器坐标系的第一姿态关系。
具体可以通过如下公式计算出世界坐标系与传感器坐标系的第一姿态关系:
Figure PCTCN2015072461-appb-000100
其中,所述
Figure PCTCN2015072461-appb-000101
为世界坐标系与传感器坐标系的第一姿态关系,其中,ψ,θ,φ分别表示偏航角(yaw)、俯仰角(pitch)和滚转角(roll)角度, Rx(φ)Ry(θ)Rz(ψ)可以通过如下公式表示
Figure PCTCN2015072461-appb-000102
Figure PCTCN2015072461-appb-000103
Figure PCTCN2015072461-appb-000104
这样通过上述公式就可以计算世界坐标系与传感器坐标系的第一姿态关系,具体可以是得到一个旋转矩阵。
可选的,步骤902可以是通过摄像装置拍照的方式计算出所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系。此处为公知的技术特征,不作详细说明。
903、获取所述摄像装置坐标系与传感器坐标系的初始姿态关系,并根据所述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息。
904、将所述估算姿态信息与所述测量姿态信息进行比较。
905、当比较结果达到预先设定的用于校准姿态关系的比较结果时,将计算所述估算姿态信息时使用的所述摄像装置坐标系与传感器坐标系的姿态关系作为所述摄像装置坐标系与传感器坐标系校准的姿态关系。
由于世界坐标系与传感器坐标系的第一姿态关系、摄像装置坐标系与所述定标装置坐标系的第二姿态关系、摄像装置坐标系与传感器坐标系的姿态关系以及定标装置坐标系在所述世界坐标系中的姿态信息这四者之间的关系在同一时刻是不变的,这样通过一个假设的摄像装置坐标系与传感器坐标系的姿态关系就可以得到一个定标装置坐标系在所述世界坐标系中的估算姿态信息,当该估算姿态信息与测量姿态信息相同或者相似时,就可以确定上述假设的摄像装置坐标系与传感器坐标系的姿态关系为摄像装置坐标系与传感器坐标系校 准的姿态关系。
可选的,上述预先设定的用于校准姿态关系的比较结果可以是估算姿态信息与所述测量姿态信息相同,或者估算姿态信息与所述测量姿态信息的相似度大于或者等于特定阈值,或者估算姿态信息与所述测量姿态信息之间的差异小于或者等于特定阈值,其中,上述特定阈值可以是智能设备默认或者预先设定的阈值,例如:98%或者99%等。
可选的,所述方法还可以包括如下步骤:
906、当比较结果未达到所述预先设定的比较结果时,基于预先设定的生成姿态关系的生成规则生成新的所述摄像装置坐标系与传感器坐标系的姿态关系。
上述比较结果未达到所述预先设定的比较结果可以是表示上述估算姿态信息与测量姿态信息的差别较大,这时就可以根据预设规则生成新的姿态关系。例如:根据上述初始姿态关系计算的估算姿态信息与测量姿态信息差别较大时,就说明上述初始姿态关系并非摄像装置坐标系与传感器坐标系高准确度的姿态关系,或者上述初始姿态关系的精度比较低,从而需要基于该初始姿态关系生成新的姿态关系。
可选的,上述生成规则可以包括:
根据上述比较结果生成新的姿态关系。
例如:当比较结果表示估算姿态信息与测量姿态信息的差异较大时,就可以将计算所述估算姿态信息时使用的所述摄像装置坐标系与传感器坐标系的姿态关系进行大幅度的调整;例如:当使用上述初始姿态关系计算的估算姿态信息与测量姿态信息的差异较大时,就可以将该初始姿态关系进行大幅度的调整。
当比较结果表示估算姿态信息与测量姿态信息的差异较小时,可以将计算所述估算姿态信息时使用的所述摄像装置坐标系与传感器坐标系的姿态关系进行小幅度的调整。例如:当使用上述初始姿态关系计算的估算姿态信息与测量姿态信息的差异较小时,就可以将该初始姿态关系进行小幅度的调整。
可选的,上述预设规则还可以包括:
在预设范围内随机生成所述摄像装置坐标系与传感器坐标系的姿态关系。
可选的,上述生成规则还可以包括:
通过Levenberg-Marquardt算法生成所述摄像装置坐标系与传感器坐标系的姿态关系。
可选的,上述摄像装置坐标系与传感器坐标系的姿态关系具体可以为一个旋转矩阵R,该旋转矩阵可以变换为一个三元组,其中,该变换可以是用Rodrigues变换。
907、根据所述生成的姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息。
执行完步骤907之后,就可以再次执行步骤904。
另外,当再次执行步骤904时,比较结果还是未达到上述预先设定的比较结果时重复执行步骤906,当然执行步骤906可以是根据上一次执行步骤906时生成的姿态关系生成新的姿态关系,之后再重复执行步骤907,这样形成一种循环,直到估算姿态信息与测量姿态信息的比较结果达到所述预先设定的比较结果。
可选的,本实施例中为了得到更加准确和精确的效果,上述定标装置的拍摄位置尽可能覆盖整个三维空间。例如:摄像装置分别采集处于n个拍摄位置的所述定标装置的图像,且每个拍摄位置以不同的角度采集m次,其中,所述n为大于1的整数,所述m为大于1的整数。例如:摄像装置对定标装置水平放置在水平面上、定标装置垂直水平面放置和定标装置下倾角60度放置时定时装置进行采集,另外,在上述三个位置上定标装置还可以是按每隔180度旋转一次定标装置,再对旋转后的定标装置进行采用。且每个拍摄位置可以不同的角度采集多次,例如:每隔30角度采集一次图像。这样通过上述方式就可以得到6个拍摄位置,以及每个拍摄位置以不同的角度采集12次,一个采集72张图像。
该实施方式中,步骤901可以包括:
获取所述定标装置在各拍摄位置时所述定标装置坐标系在世界坐标系的测量姿态信息。即可以得到n个测量姿态信息。
步骤902可以包括:
计算所述摄像装置每次采集所述定标装置时世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置每次采集所述定标装置时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系;
步骤903可以包括:
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置每次采集所述定标装置时所述定标装置坐标系在所述世界坐标系中的估算姿态信息。
可选的,上述步骤903可以包括:
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置时所述定标装置坐标系在所述世界坐标系的下倾角的估算值;
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置时所述定标装置坐标系的X轴与磁北的夹角的估算值;
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置时所述定标装置坐标系的Y轴与磁北的夹角的估算值。
该实施方式中可以通过上述下倾角、X轴与磁北的夹角和Y轴与磁北的夹角的估算值,以及上述测量测量姿态信息计算摄像装置坐标系与传感器坐标系的姿态关系。另外,上述测量测量姿态信息可以包括上述下倾角、X轴与磁北的夹角和Y轴与磁北的夹角的测量值。
可选的,上述方法还可以包括:
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置时所述定标装置坐标系与所述世界坐标系的第三姿态关系。
例如:可以通过如下公式计算上述第三姿态关系:
Figure PCTCN2015072461-appb-000105
其中,所述
Figure PCTCN2015072461-appb-000106
为所述定标装置在第i个拍摄位置时所述摄像装置进 行的第j次采集时定标装置与世界坐标系的第三姿态关系,
Figure PCTCN2015072461-appb-000107
为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时世界坐标系与传感器坐标系的第一姿态关系,
Figure PCTCN2015072461-appb-000108
为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时摄像装置坐标系与传感器坐标系的姿态关系,该姿态关系可以为假设的姿态关系,例如:上述初始姿态关系,
Figure PCTCN2015072461-appb-000109
为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时摄像装置坐标系与所述定标装置坐标系的第二姿态关系,其中,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
可选的,上述根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置时所述定标装置坐标系在所述世界坐标系的下倾角的估算值的步骤,可以包括:
计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Z轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与重力方向的夹角为作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系在所述世界坐标系的下倾角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
可选的,上述根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置时所述定标装置坐标系的X轴与磁北的夹角的估算值的步骤,可以包括:
计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的X轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
可选的,上述根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置时所述定标装置坐标系的Y轴与磁北的夹角的估算值的步骤,可以包括:
计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Y轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
上述对下倾角、X轴与磁北的夹角和Y轴与磁北的夹角的估算值可以通过如下公式计算:
Figure PCTCN2015072461-appb-000110
Figure PCTCN2015072461-appb-000111
Figure PCTCN2015072461-appb-000112
Figure PCTCN2015072461-appb-000113
Figure PCTCN2015072461-appb-000114
Figure PCTCN2015072461-appb-000115
Figure PCTCN2015072461-appb-000116
其中,上述
Figure PCTCN2015072461-appb-000117
分别表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时计算得到定标装置坐标系X轴、Y轴和Z轴在世界坐标系下的3维向量。
Figure PCTCN2015072461-appb-000118
Figure PCTCN2015072461-appb-000119
分别表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时定标装置坐标系的X轴、Y轴和Z轴在定标装置坐标系的3维向量,
Figure PCTCN2015072461-appb-000120
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时定标装置坐标系在所述世界坐标系的下倾角的估算值,
Figure PCTCN2015072461-appb-000121
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,
Figure PCTCN2015072461-appb-000122
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,上述
Figure PCTCN2015072461-appb-000123
表示
Figure PCTCN2015072461-appb-000124
与重力方向的夹角,
Figure PCTCN2015072461-appb-000125
表示
Figure PCTCN2015072461-appb-000126
投影到水平面上和磁北方向的夹角,
Figure PCTCN2015072461-appb-000127
表示
Figure PCTCN2015072461-appb-000128
投影到水平面上和磁北方向的夹角。
通过上述公式就可以得到上述下倾角、X轴与磁北的夹角和Y轴与磁北的夹角的估算值。
可选的,上述获取所述定标装置在各拍摄位置时所述定标装置坐标系在世界坐标系的测量姿态信息的步骤,可以包括:
获取所述定标装置在各拍摄位置时所述定标装置坐标系在所述世界坐标系的下倾角的测量值;
获取所述定标装置在各拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值;
获取所述定标装置在各拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
可选的,步骤904可以包括:包括:
将所述下倾角的估算值、所述定标装置坐标系的X轴与磁北的夹角的估算值和所述定标装置坐标系的Y轴与磁北的夹角的估算值分别与所述下倾角的测量值、所述定标装置坐标系的X轴与磁北的夹角的测量值和所述定标装置坐标系的Y轴与磁北的夹角的测量值进行加权距离计算并将所述计算结果作为所述估算姿态信息与所述测量姿态信息的比较结果。
例如:904可以包括:包括:
通过如下公式计算将所述估算姿态信息与所述测量姿态信息进行比较:
Figure PCTCN2015072461-appb-000129
其中,所述T为所述估算姿态信息与所述测量姿态信息的比较结果,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于所述n,所述j表示所述定标装置所处各位置时所述摄像装置第j次采集,其中,j大于或者等于1,且小于或者等于所述m,w1、w2和w3为预设的权值,所述
Figure PCTCN2015072461-appb-000130
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置的下倾角的估算值,所述
Figure PCTCN2015072461-appb-000131
表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
Figure PCTCN2015072461-appb-000132
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000133
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
Figure PCTCN2015072461-appb-000134
表示所述定标装置在 第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000135
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
其中,上述w1、w2和w3可以为预设配置的,例如:将w1:w2:w3可以配置为100:1:1。
本实施例,在图8所示的实施例的基础上增加了多种可选的实施方式,且都可以实现校准智能设备中传感器坐标系与摄像装置坐标系的姿态关系。
请参阅图10,图10是本发明实施例提供的另一种姿态关系计算方法的流程示意图,如图10所示,包括以下步骤:
1001、获取摄像装置采集定标装置的图像时所述定标装置坐标系在世界坐标系的测量姿态信息,其中,所述测量姿态信息用于表示所述定标装置坐标系在所述世界坐标系的朝向。
1002、计算所述摄像装置采集所述定标装置的图像时世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置采集所述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系。
1003、获取所述摄像装置坐标系与传感器坐标系的初始姿态关系,并根据所述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息。
1004、通过目标公式计算所述摄像装置坐标系与传感器坐标系校准的姿态关系。
其中,目标公式可以如下:
Figure PCTCN2015072461-appb-000136
其中,上述公式求得的解为所述摄像装置坐标系与传感器坐标系校准的姿态关系,所述min()为最小值函数,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于n,所述n为所述定标装置在 拍摄时所处的拍摄位置的数量,所述j表示所述定标装置所处各位置时所述摄像装置第j次采集,其中,j大于或者等于1,且小于或者等于m,所述m为每个拍摄位置所述摄像装置采集的次数,w1、w2和w3为预设的权值,所述
Figure PCTCN2015072461-appb-000137
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置的下倾角的估算值,所述
Figure PCTCN2015072461-appb-000138
表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
Figure PCTCN2015072461-appb-000139
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000140
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
Figure PCTCN2015072461-appb-000141
表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述
Figure PCTCN2015072461-appb-000142
表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
其中,上述w1、w2和w3可以为预设配置的,例如:将w1:w2:w3可以配置为100:1:1。
可选的,可以是对上述公式进行优化求解,就可以得到摄像装置坐标系与传感器坐标系校准的姿态关系。例如:可以通过用Levenberg-Marquardt算法进行优化求解。具体可以是通过带有不同优化参数的Levenberg-Marquardt算法生成多个候选(或者理解为假设)的所述摄像装置坐标系与传感器坐标系的姿态关系,再根据该所述摄像装置坐标系与传感器坐标系的姿态关系计算所述定标装置坐标系在所述世界坐标系中的估算姿态信息,当某一个候选的所述摄像装置坐标系与传感器坐标系的姿态关系计算出的上述估算姿态信息与测量姿态信息的差异(或者均方误差)最小时,该候选的所述摄像装置坐标系与传感器坐标系的姿态关系就可以作为所述摄像装置坐标系与传感器坐标系校准的姿态关系。
其中,上述公式中的待求解参数为一个旋转矩阵R,这样可以将待求解参数变换为一个三元组,故上述公式的未知参数是一组三元向量。其中,该变换可以是用Rodrigues变换。
可选的,所述摄像装置分别采集处于n个拍摄位置的所述定标装置,且每个拍摄位置以不同的角度采集m次,其中,所述n为大于或者等于1的整数, 所述m为大于或者等于1的整数;
所述获取摄像装置采集定标装置时所述定标装置坐标系在世界坐标系的测量姿态信息,可以包括:
获取所述定标装置在各拍摄位置时所述定标装置坐标系在世界坐标系的测量姿态信息;
所述计算所述摄像装置采用所述定标装置时世界坐标系与传感器坐标系的第一姿态关系,可以包括:
计算所述摄像装置每次采集所述定标装置时世界坐标系与传感器坐标系的第一姿态关系;
所述计算所述摄像装置采集所述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系,可以包括:
计算所述摄像装置每次采集所述定标装置时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系;
所述根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息,可以包括:
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置每次采集所述定标装置时所述定标装置坐标系在所述世界坐标系中的估算姿态信息。
可选的,上述根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置时所述定标装置坐标系在所述世界坐标系中的估算姿态信息,可以包括:
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置时所述定标装置坐标系在所述世界坐标系的下倾角的估算值;
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置时所述定标装置坐标系的X轴与磁北的夹角的估算值;
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述 摄像装置采集所述定标装置时所述定标装置坐标系的Y轴与磁北的夹角的估算值。
可选的,上述方法还可以包括:
根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置时所述定标装置坐标系与所述世界坐标系的第三姿态关系。
可选的,上述根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置时所述定标装置坐标系在所述世界坐标系的下倾角的估算值,可以包括:
计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Z轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与重力方向的夹角为作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系在所述世界坐标系的下倾角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
可选的,上述根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置时所述定标装置坐标系的X轴与磁北的夹角的估算值,可以包括:
计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的X轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
可选的,上述根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置时所述定标装置坐标系的Y轴与磁北的夹角的估算值,可以包括:
计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Y轴在所述定标装置坐标系中 的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
可选的,上述获取所述定标装置在各拍摄位置时所述定标装置坐标系在世界坐标系的测量姿态信息,可以包括:
获取所述定标装置在各拍摄位置时所述定标装置坐标系在所述世界坐标系的下倾角的测量值;
获取所述定标装置在各拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值;
获取所述定标装置在各拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
需要说明的是,本实施例中的介绍的多种实施方式都可以参考图2所示的实施例中介绍的实施方式,此处不作重复说明。
本实施例,在图8所示的实施例的基础上增加了多种可选的实施方式,且都可以实现校准智能设备中传感器坐标系与摄像装置坐标系的姿态关系。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程,是可以通过计算机程序来指令相关的硬件来完成,所述的程序可存储于一计算机可读取存储介质中,该程序在执行时,可包括如上述各方法的实施例的流程。其中,所述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory,ROM)或随机存取存储器(Random Access Memory,简称RAM)等。
以上所揭露的仅为本发明较佳实施例而已,当然不能以此来限定本发明之权利范围,因此依本发明权利要求所作的等同变化,仍属本发明所涵盖的范围。

Claims (39)

  1. 一种智能设备,其特征在于,包括:获取单元、第一计算单元、第二计算单元和第三计算单元,其中:
    所述获取单元,用于获取摄像装置采集定标装置的图像时所述定标装置坐标系在世界坐标系的测量姿态信息,其中,所述测量姿态信息用于表示所述定标装置坐标系在所述世界坐标系的朝向;
    所述第一计算单元,用于计算所述摄像装置采集所述定标装置的图像时所述世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置采集所述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系;
    所述第二计算单元,用于获取所述摄像装置坐标系与所述传感器坐标系的初始姿态关系,并根据所述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息;
    所述第三计算单元,用于基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系。
  2. 如权利要求1所述的设备,其特征在于,所述第三计算单元包括:
    比较单元,用于将所述估算姿态信息与所述测量姿态信息进行比较;
    计算子单元,用于当比较结果达到预先设定的用于校准姿态关系的比较结果时,将计算所述估算姿态信息时使用的所述摄像装置坐标系与传感器坐标系的姿态关系作为所述摄像装置坐标系与传感器坐标系校准的姿态关系。
  3. 如权利要求2所述的设备,其特征在于,所述设备还包括:
    生成单元,用于当比较结果未达到所述预先设定的比较结果时,基于预先设定的生成姿态关系的生成规则生成新的所述摄像装置坐标系与传感器坐标系的姿态关系;
    第四计算单元,用于根据所述生成的姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐 标系在所述世界坐标系中的估算姿态信息。
  4. 如权利要求1所述的设备,其特征在于,所述第三计算单元包括:
    通过如下公式计算所述摄像装置坐标系与传感器坐标系校准的姿态关系
    Figure PCTCN2015072461-appb-100001
    其中,上述公式求得的解为所述摄像装置坐标系与传感器坐标系校准的姿态关系,所述min()为最小值函数,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于n,所述n为所述定标装置在拍摄时所处的拍摄位置的数量,所述j表示所述定标装置所处各位置时所述摄像装置第j次采集,其中,j大于或者等于1,且小于或者等于m,所述m为每个拍摄位置所述摄像装置采集的次数,w1、w2和w3为预设的权值,所述
    Figure PCTCN2015072461-appb-100002
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置的下倾角的估算值,所述
    Figure PCTCN2015072461-appb-100003
    表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
    Figure PCTCN2015072461-appb-100004
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述
    Figure PCTCN2015072461-appb-100005
    表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
    Figure PCTCN2015072461-appb-100006
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述
    Figure PCTCN2015072461-appb-100007
    表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
  5. 如权利要求1-4中任一项所述的设备,其特征在于,所述摄像装置分别采集处于n个拍摄位置的所述定标装置的图像,且每个拍摄位置以不同的角度采集m次,其中,所述n为大于或者等于1的整数,所述m为大于或者等于1的整数;
    所述获取单元用于获取所述定标装置在各拍摄位置时所述定标装置坐标系在世界坐标系的测量姿态信息。
  6. 如权利要求5所述的设备,其特征在于,所述第二计算单元包括:
    第一估算单元,用于根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系的下倾角的估算值;
    第二估算单元,用于根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的X轴与磁北的夹角的估算值;
    第三估算单元,用于根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的Y轴与磁北的夹角的估算值。
  7. 如权利要求6所述的设备,其特征在于,所述设备还包括:
    第五计算单元,用于根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系与所述世界坐标系的第三姿态关系。
  8. 如权利要求7所述的设备,其特征在于,所述第一估算单元用于计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Z轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与重力方向的夹角为作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系在所述世界坐标系的下倾角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
  9. 如权利要求7所述的设备,其特征在于,所述第二估算单元用于计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的X轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时 所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
  10. 如权利要求7所述的设备,其特征在于,所述第三估算单元用于计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Y轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
  11. 如权利要求5-10中任一项所述的设备,其特征在于,所述获取单元包括:
    第一获取子单元,用于获取所述定标装置在各拍摄位置时所述定标装置坐标系在所述世界坐标系的下倾角的测量值;
    第二获取子单元,用于获取所述定标装置在各拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值;
    第三获取子单元,用于获取所述定标装置在各拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
  12. 如权利要求11所述的设备,其特征在于,所述比较单元用于将所述下倾角的估算值、所述定标装置坐标系的X轴与磁北的夹角的估算值和所述定标装置坐标系的Y轴与磁北的夹角的估算值分别与所述下倾角的测量值、所述定标装置坐标系的X轴与磁北的夹角的测量值和所述定标装置坐标系的Y轴与磁北的夹角的测量值进行加权距离计算,并将所述计算结果作为所述估算姿态信息与所述测量姿态信息的比较结果。
  13. 如权利要求11所述的设备,其特征在于,所述比较单元用于通过如 下公式计算将所述估算姿态信息与所述测量姿态信息进行比较:
    Figure PCTCN2015072461-appb-100008
    其中,所述T为所述估算姿态信息与所述测量姿态信息的比较结果,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于所述n,所述j表示所述定标装置所处各位置时所述摄像装置第j次采集,其中,j大于或者等于1,且小于或者等于所述m,w1、w2和w3为预设的权值,所述
    Figure PCTCN2015072461-appb-100009
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置的下倾角的估算值,所述
    Figure PCTCN2015072461-appb-100010
    表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
    Figure PCTCN2015072461-appb-100011
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述
    Figure PCTCN2015072461-appb-100012
    表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
    Figure PCTCN2015072461-appb-100013
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述
    Figure PCTCN2015072461-appb-100014
    表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
  14. 一种智能设备,其特征在于,包括:处理器、网络接口、存储器、通信总线、摄像装置和传感器,其中,所述通信总线用于实现所述处理器、所述网络接口、所述存储器、所述摄像装置和所述传感器之间连接通信,所述处理器用于执行所述存储器中存储的程序;其中,所述程序包括:
    获取所述摄像装置采集定标装置的图像时所述定标装置坐标系在世界坐标系的测量姿态信息,其中,所述测量姿态信息用于表示所述定标装置坐标系在所述世界坐标系的朝向;
    计算所述摄像装置采集所述定标装置的图像时所述世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置采集所述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系;
    获取所述摄像装置坐标系与所述传感器坐标系的初始姿态关系,并根据所 述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息;
    基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系。
  15. 如权利要求14所述的设备,其特征在于,所述处理器执行的基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系的程序,包括:
    将所述估算姿态信息与所述测量姿态信息进行比较;
    当比较结果达到预先设定的用于校准姿态关系的比较结果时,将计算所述估算姿态信息时使用的所述摄像装置坐标系与传感器坐标系的姿态关系作为所述摄像装置坐标系与传感器坐标系校准的姿态关系。
  16. 如权利要求15所述的设备,其特征在于,所述处理器执行的程序还包括:
    当比较结果未达到所述预先设定的比较结果时,基于预先设定的生成姿态关系的生成规则生成新的所述摄像装置坐标系与传感器坐标系的姿态关系;
    根据所述生成的姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息,并再次执行所述将所述估算姿态信息与所述测量姿态信息进行比较的步骤。
  17. 如权利要求14所述的设备,其特征在于,所述处理器执行的基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系的程序,包括:
    通过如下公式计算所述摄像装置坐标系与传感器坐标系校准的姿态关系
    Figure PCTCN2015072461-appb-100015
    其中,上述公式求得的解为所述摄像装置坐标系与传感器坐标系校准的姿态关系,所述min()为最小值函数,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于n,所述n为所述定标装置在拍摄时所处的拍摄位置的数量,所述j表示所述定标装置所处各位置时所述摄像装置第j次采集,其中,j大于或者等于1,且小于或者等于m,所述m为每个拍摄位置所述摄像装置采集的次数,w1、w2和w3为预设的权值,所述
    Figure PCTCN2015072461-appb-100016
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置的下倾角的估算值,所述
    Figure PCTCN2015072461-appb-100017
    表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
    Figure PCTCN2015072461-appb-100018
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述
    Figure PCTCN2015072461-appb-100019
    表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
    Figure PCTCN2015072461-appb-100020
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述
    Figure PCTCN2015072461-appb-100021
    表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
  18. 如权利要求14-17中任一项所述的设备,其特征在于,所述摄像装置分别采集处于n个拍摄位置的所述定标装置的图像,且每个拍摄位置以不同的角度采集m次,其中,所述n为大于或者等于1的整数,所述m为大于或者等于1的整数;
    所述处理器执行的获取摄像装置采集定标装置时所述定标装置坐标系在世界坐标系的测量姿态信息的程序,包括:
    获取所述定标装置在各拍摄位置时所述定标装置坐标系在世界坐标系的测量姿态信息。
  19. 如权利要求18所述的方法,其特征在于,所述处理器执行的根据所 述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息的程序,包括:
    根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系的下倾角的估算值;
    根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的X轴与磁北的夹角的估算值;
    根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的Y轴与磁北的夹角的估算值。
  20. 如权利要求19所述的设备,其特征在于,所述处理器执行的程序还包括:
    根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系与所述世界坐标系的第三姿态关系。
  21. 如权利要求20所述的设备,其特征在于,所述处理器执行的根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系的下倾角的估算值的程序,包括:
    计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Z轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与重力方向的夹角为作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系在所述世界坐标系的下倾角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
  22. 如权利要求20所述的程序,其特征在于,所述处理器执行的根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的X轴与磁北的夹角的估算值的程序,包括:
    计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的X轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
  23. 如权利要求20所述的设备,其特征在于,所述处理器执行的根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的Y轴与磁北的夹角的估算值的程序,包括:
    计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Y轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
  24. 如权利要求18-23中任一项所述的设备,其特征在于,所述处理器执行的获取所述定标装置在各拍摄位置时所述定标装置坐标系在世界坐标系的测量姿态信息的程序,包括:
    获取所述定标装置在各拍摄位置时所述定标装置坐标系在所述世界坐标系的下倾角的测量值;
    获取所述定标装置在各拍摄位置时所述定标装置坐标系的X轴与磁北的 夹角的测量值;
    获取所述定标装置在各拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
  25. 如权利要求24所述的设备,其特征在于,所述处理器执行的将所述估算姿态信息与所述测量姿态信息进行比较的程序,包括:
    将所述下倾角的估算值、所述定标装置坐标系的X轴与磁北的夹角的估算值和所述定标装置坐标系的Y轴与磁北的夹角的估算值分别与所述下倾角的测量值、所述定标装置坐标系的X轴与磁北的夹角的测量值和所述定标装置坐标系的Y轴与磁北的夹角的测量值进行加权距离计算,并将所述计算结果作为所述估算姿态信息与所述测量姿态信息的比较结果。
  26. 如权利要求24所述的设备,其特征在于,所述处理器执行的将所述估算姿态信息与所述测量姿态信息进行比较的程序,包括:
    通过如下公式计算将所述估算姿态信息与所述测量姿态信息进行比较:
    Figure PCTCN2015072461-appb-100022
    其中,所述T为所述估算姿态信息与所述测量姿态信息的比较结果,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于所述n,所述j表示所述定标装置所处各位置时所述摄像装置第j次采集,其中,j大于或者等于1,且小于或者等于所述m,w1、w2和w3为预设的权值,所述
    Figure PCTCN2015072461-appb-100023
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置的下倾角的估算值,所述
    Figure PCTCN2015072461-appb-100024
    表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
    Figure PCTCN2015072461-appb-100025
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述
    Figure PCTCN2015072461-appb-100026
    表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
    Figure PCTCN2015072461-appb-100027
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y 轴与磁北的夹角的估算值,所述
    Figure PCTCN2015072461-appb-100028
    表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
  27. 一种智能设备中的姿态关系计算方法,其特征在于,包括:
    获取摄像装置采集定标装置的图像时所述定标装置坐标系在世界坐标系的测量姿态信息,其中,所述测量姿态信息用于表示所述定标装置坐标系在所述世界坐标系的朝向;
    计算所述摄像装置采集所述定标装置的图像时所述世界坐标系与传感器坐标系的第一姿态关系,以及计算所述摄像装置采集所述定标装置的图像时所述摄像装置坐标系与所述定标装置坐标系的第二姿态关系;
    获取所述摄像装置坐标系与所述传感器坐标系的初始姿态关系,并根据所述初始姿态关系、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息;
    基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系。
  28. 如权利要求27所述的方法,其特征在于,所述基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系,包括:
    将所述估算姿态信息与所述测量姿态信息进行比较;
    当比较结果达到预先设定的用于校准姿态关系的比较结果时,将计算所述估算姿态信息时使用的所述摄像装置坐标系与传感器坐标系的姿态关系作为所述摄像装置坐标系与传感器坐标系校准的姿态关系。
  29. 如权利要求28所述的方法,其特征在于,所述方法还包括:
    当比较结果未达到所述预先设定的比较结果时,基于预先设定的生成姿态关系的生成规则生成新的所述摄像装置坐标系与传感器坐标系的姿态关系;
    根据所述生成的姿态关系、所述第一姿态关系和所述第二姿态关系计算所 述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息,并再次执行所述将所述估算姿态信息与所述测量姿态信息进行比较的步骤。
  30. 如权利要求27所述的方法,其特征在于,所述基于所述估算姿态信息和所述测量姿态信息计算所述摄像装置坐标系与传感器坐标系校准的姿态关系,包括:
    通过如下公式计算所述摄像装置坐标系与传感器坐标系校准的姿态关系
    Figure PCTCN2015072461-appb-100029
    其中,上述公式求得的解为所述摄像装置坐标系与传感器坐标系校准的姿态关系,所述min()为最小值函数,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于n,所述n为所述定标装置在拍摄时所处的拍摄位置的数量,所述j表示所述定标装置所处各位置时所述摄像装置第j次采集,其中,j大于或者等于1,且小于或者等于m,所述m为每个拍摄位置所述摄像装置采集的次数,w1、w2和w3为预设的权值,所述
    Figure PCTCN2015072461-appb-100030
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置的下倾角的估算值,所述
    Figure PCTCN2015072461-appb-100031
    表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
    Figure PCTCN2015072461-appb-100032
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述
    Figure PCTCN2015072461-appb-100033
    表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
    Figure PCTCN2015072461-appb-100034
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述
    Figure PCTCN2015072461-appb-100035
    表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
  31. 如权利要求27-30中任一项所述的方法,其特征在于,所述摄像装置采集处于n个拍摄位置的所述定标装置的图像,且每个拍摄位置以不同的角度 采集m次,其中,所述n为大于或者等于1的整数,所述m为大于或者等于1的整数;
    所述获取摄像装置采集定标装置时所述定标装置坐标系在世界坐标系的测量姿态信息,包括:
    获取所述定标装置在各拍摄位置时所述定标装置坐标系在世界坐标系的测量姿态信息。
  32. 如权利要求31所述的方法,其特征在于,所述根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系中的估算姿态信息,包括:
    根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系的下倾角的估算值;
    根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的X轴与磁北的夹角的估算值;
    根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的Y轴与磁北的夹角的估算值。
  33. 如权利要求32所述的方法,其特征在于,所述方法还包括:
    根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系与所述世界坐标系的第三姿态关系。
  34. 如权利要求33所述的方法,其特征在于,所述根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系在所述世界坐标系的下倾角的估算值,包括:
    计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时 的所述第三姿态关系与所述定标装置坐标系的Z轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与重力方向的夹角为作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系在所述世界坐标系的下倾角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
  35. 如权利要求33所述的方法,其特征在于,所述根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的X轴与磁北的夹角的估算值,包括:
    计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的X轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
  36. 如权利要求33所述的方法,其特征在于,所述根据所述初始姿态信息、所述第一姿态关系和所述第二姿态关系计算所述摄像装置采集所述定标装置的图像时所述定标装置坐标系的Y轴与磁北的夹角的估算值,包括:
    计算所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时的所述第三姿态关系与所述定标装置坐标系的Y轴在所述定标装置坐标系中的向量的乘积,并将所述乘积与磁北方向的夹角作为所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述i大于或者等于1,且小于或者等于所述n,所述j大于或者等于1,且小于或者等于所述m。
  37. 如权利要求31-36中任一项所述的方法,其特征在于,所述获取所述定标装置在各拍摄位置时所述定标装置坐标系在世界坐标系的测量姿态信息,包括:
    获取所述定标装置在各拍摄位置时所述定标装置坐标系在所述世界坐标系的下倾角的测量值;
    获取所述定标装置在各拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值;
    获取所述定标装置在各拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
  38. 如权利要求27所述的方法,其特征在于,所述将所述估算姿态信息与所述测量姿态信息进行比较,包括:
    将所述下倾角的估算值、所述定标装置坐标系的X轴与磁北的夹角的估算值和所述定标装置坐标系的Y轴与磁北的夹角的估算值分别与所述下倾角的测量值、所述定标装置坐标系的X轴与磁北的夹角的测量值和所述定标装置坐标系的Y轴与磁北的夹角的测量值进行加权距离计算,并将所述计算结果作为所述估算姿态信息与所述测量姿态信息的比较结果。
  39. 如权利要求27所述的方法,其特征在于,所述将所述估算姿态信息与所述测量姿态信息进行比较,包括:
    通过如下公式计算将所述估算姿态信息与所述测量姿态信息进行比较:
    Figure PCTCN2015072461-appb-100036
    其中,所述T为所述估算姿态信息与所述测量姿态信息的比较结果,所述i表示所述定标装置所处的第i个拍摄位置,其中,i大于或者等于1,且小于或者等于所述n,所述j表示所述定标装置所处各位置时所述摄像装置第j次采集,其中,j大于或者等于1,且小于或者等于所述m,w1、w2和w3为预设的权值,所述
    Figure PCTCN2015072461-appb-100037
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置的下倾角的估算值,所述
    Figure PCTCN2015072461-appb-100038
    表示所述定标装置在第i个拍摄位置时所述定标装置的下倾角的测量值,所述
    Figure PCTCN2015072461-appb-100039
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的X 轴与磁北的夹角的估算值,所述
    Figure PCTCN2015072461-appb-100040
    表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的X轴与磁北的夹角的测量值,所述
    Figure PCTCN2015072461-appb-100041
    表示所述定标装置在第i个拍摄位置时所述摄像装置进行的第j次采集时所述定标装置坐标系的Y轴与磁北的夹角的估算值,所述
    Figure PCTCN2015072461-appb-100042
    表示所述定标装置在第i个拍摄位置时所述定标装置坐标系的Y轴与磁北的夹角的测量值。
PCT/CN2015/072461 2015-02-07 2015-02-07 一种智能设备的姿态关系计算方法和智能设备 WO2016123813A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201580010064.2A CN106461414B (zh) 2015-02-07 2015-02-07 一种智能设备的姿态关系计算方法和智能设备
PCT/CN2015/072461 WO2016123813A1 (zh) 2015-02-07 2015-02-07 一种智能设备的姿态关系计算方法和智能设备

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2015/072461 WO2016123813A1 (zh) 2015-02-07 2015-02-07 一种智能设备的姿态关系计算方法和智能设备

Publications (1)

Publication Number Publication Date
WO2016123813A1 true WO2016123813A1 (zh) 2016-08-11

Family

ID=56563352

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/072461 WO2016123813A1 (zh) 2015-02-07 2015-02-07 一种智能设备的姿态关系计算方法和智能设备

Country Status (2)

Country Link
CN (1) CN106461414B (zh)
WO (1) WO2016123813A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110036259A (zh) * 2017-09-30 2019-07-19 华为技术有限公司 一种姿态矩阵的计算方法及设备
CN115953477A (zh) * 2022-12-20 2023-04-11 苏州方石科技有限公司 坐标数据处理方法和装置

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106931960B (zh) * 2017-03-13 2020-05-05 武汉准我飞科技有限公司 一种姿态数据获取方法及装置
CN109814714B (zh) * 2019-01-21 2020-11-20 北京诺亦腾科技有限公司 运动传感器的安装姿态确定方法、装置以及存储介质
CN112733773B (zh) * 2021-01-18 2024-07-16 上海商汤智能科技有限公司 一种对象检测方法、装置、计算机设备和存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080228422A1 (en) * 2007-03-13 2008-09-18 Canon Kabushiki Kaisha Information processing apparatus and information processing method
CN101539397A (zh) * 2009-04-17 2009-09-23 中国人民解放军国防科学技术大学 物体三维姿态的精密光学测量方法
CN102162738A (zh) * 2010-12-08 2011-08-24 中国科学院自动化研究所 摄像头与惯性传感器组合定位定姿系统的标定方法
CN104021554A (zh) * 2014-04-23 2014-09-03 北京大学深圳研究生院 基于部分传感器信息的相机-惯性传感器标定方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103514450B (zh) * 2012-06-29 2017-02-15 华为技术有限公司 一种图像特征提取方法和图像校正方法以及设备
US9699375B2 (en) * 2013-04-05 2017-07-04 Nokia Technology Oy Method and apparatus for determining camera location information and/or camera pose information according to a global coordinate system
CN103335634B (zh) * 2013-06-24 2015-10-28 南京航空航天大学 一种用于部件安装姿态校准的视觉靶板装置及标定方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080228422A1 (en) * 2007-03-13 2008-09-18 Canon Kabushiki Kaisha Information processing apparatus and information processing method
CN101539397A (zh) * 2009-04-17 2009-09-23 中国人民解放军国防科学技术大学 物体三维姿态的精密光学测量方法
CN102162738A (zh) * 2010-12-08 2011-08-24 中国科学院自动化研究所 摄像头与惯性传感器组合定位定姿系统的标定方法
CN104021554A (zh) * 2014-04-23 2014-09-03 北京大学深圳研究生院 基于部分传感器信息的相机-惯性传感器标定方法

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110036259A (zh) * 2017-09-30 2019-07-19 华为技术有限公司 一种姿态矩阵的计算方法及设备
CN110036259B (zh) * 2017-09-30 2021-03-30 华为技术有限公司 一种姿态矩阵的计算方法及设备
US11392668B2 (en) 2017-09-30 2022-07-19 Huawei Technologies Co., Ltd. Attitude matrix calculation method and device
CN115953477A (zh) * 2022-12-20 2023-04-11 苏州方石科技有限公司 坐标数据处理方法和装置
CN115953477B (zh) * 2022-12-20 2024-05-14 苏州方石科技有限公司 坐标数据处理方法和装置

Also Published As

Publication number Publication date
CN106461414A (zh) 2017-02-22
CN106461414B (zh) 2019-06-28

Similar Documents

Publication Publication Date Title
US20210233275A1 (en) Monocular vision tracking method, apparatus and non-transitory computer-readable storage medium
US20210190497A1 (en) Simultaneous location and mapping (slam) using dual event cameras
CN107747941B (zh) 一种双目视觉定位方法、装置及系统
CN106446815B (zh) 一种同时定位与地图构建方法
JP6622503B2 (ja) カメラモデルパラメータ推定装置及びそのプログラム
JP5746477B2 (ja) モデル生成装置、3次元計測装置、それらの制御方法及びプログラム
WO2016123813A1 (zh) 一种智能设备的姿态关系计算方法和智能设备
WO2021043213A1 (zh) 标定方法、装置、航拍设备和存储介质
WO2016199605A1 (ja) 画像処理装置および方法、並びにプログラム
CN107577451B (zh) 多Kinect人体骨架坐标变换方法及处理设备、可读存储介质
WO2016146001A1 (zh) 一种三维建模的方法及装置
WO2013111229A1 (ja) カメラ校正装置、カメラ校正方法およびカメラ校正用プログラム
CN102472609A (zh) 位置和姿势校准方法及设备
CN110660098B (zh) 基于单目视觉的定位方法和装置
JP6584208B2 (ja) 情報処理装置、情報処理方法、プログラム
WO2022247548A1 (zh) 定位方法和装置、电子设备及存储介质
CN108780577A (zh) 图像处理方法和设备
WO2019080046A1 (zh) 惯性测量单元的漂移标定方法、设备及无人飞行器
CN112509036B (zh) 位姿估计网络训练及定位方法、装置、设备、存储介质
JP6662382B2 (ja) 情報処理装置および方法、並びにプログラム
JP2016162425A (ja) 骨格姿勢推定装置、方法、及びプログラム
JP2010145219A (ja) 運動推定装置及びプログラム
US9245343B1 (en) Real-time image geo-registration processing
CN112330726A (zh) 一种图像处理方法及装置
Park et al. Noise sensitivity analysis of statistically consistent optimal structure from motion

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15880769

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15880769

Country of ref document: EP

Kind code of ref document: A1