CN110136203A - The scaling method and its calibration system of TOF device - Google Patents

The scaling method and its calibration system of TOF device Download PDF

Info

Publication number
CN110136203A
CN110136203A CN201810202909.3A CN201810202909A CN110136203A CN 110136203 A CN110136203 A CN 110136203A CN 201810202909 A CN201810202909 A CN 201810202909A CN 110136203 A CN110136203 A CN 110136203A
Authority
CN
China
Prior art keywords
tof
rgb
module
camera module
image
Prior art date
Legal status (The legal status 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 status listed.)
Granted
Application number
CN201810202909.3A
Other languages
Chinese (zh)
Other versions
CN110136203B (en
Inventor
朱兴霞
张严严
潘锋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang Sunny Optical Intelligent Technology Co Ltd
Original Assignee
Zhejiang Sunny Optical Intelligent Technology Co Ltd
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 Zhejiang Sunny Optical Intelligent Technology Co Ltd filed Critical Zhejiang Sunny Optical Intelligent Technology Co Ltd
Publication of CN110136203A publication Critical patent/CN110136203A/en
Application granted granted Critical
Publication of CN110136203B publication Critical patent/CN110136203B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Studio Devices (AREA)

Abstract

The invention discloses the scaling method of a TOF device and its calibration system, the scaling method is the following steps are included: acquisition at least three TOF images and at least three RGB images；Based on at least three TOF images, monocular demarcates a TOF camera module of the TOF device, parameter outside the TOF intrinsic parameter and TOF to obtain the TOF camera module；Based at least three RGB images, monocular demarcates a RGB camera module of the TOF device, parameter outside the RGB intrinsic parameter and RGB to obtain the RGB camera module；And the binocular calibration TOF device, to obtain the structural parameters of the TOF device.The scaling method of the TOF device provided by the invention can carry out monocular calibration and binocular calibration to the TOF device, and easy to operate.

Description

The scaling method and its calibration system of TOF device
Technical field
The present invention relates to TOF (flight time, Time of Flight) technical fields, more particularly to a TOF device Scaling method and its calibration system.
Background technique
In recent years, with the rapid development of microelectric technique, computer technology, modern communication technology, human society is just It walks with vigorous strides to march toward the information age.Correspondingly, TOF device (such as TOF camera) is also rapidly developed and is popularized therewith, more and more Scene and field need to apply to TOF device, for example, three-dimensional modeling, recognition of face, gesture identification, artificial intelligence, machine view Feel, three-dimensional restores and the numerous areas such as target tracking.
Currently, TOF device in the market generally includes a TOF camera module and a RGB camera module, to pass through the TOF Camera module obtains the depth information and gray level image of a testee, while obtaining the measured object by the RGB camera module The color image of body, so as to such as three-dimensional reduction of subsequent progress, Stereo image matching and three-dimensional reconstruction etc. application.It is worth noting , the TOF camera module of the TOF device is usually that the testee (or measured target) is measured by time-of-flight method Depth information, specifically, time-of-flight method (Time Of Flight, TOF) by measurement one measuring instrument actively issue Pulse signal is from being emitted to received time interval t (being commonly referred to as pulse ranging method) or the round-trip testee of laser is once produced Raw phase difference (phase difference ranging method) is realized for generating depth information to quilt with being converted into the distance for the scenery that is taken The three-dimensional structure of object (or testee detection zone) or the measurement of three-D profile are surveyed, and then obtains the gray scale of the testee Image and depth information.
However, not only needing respectively to take the photograph the TOF of the TOF device due to before being worked using the TOF device As mould group and RGB camera module progress monocular calibration, and it is also required to carry out the TOF device binocular calibration, and the TOF The calibration of equipment is the key that can the TOF device normally work.Mould is being imaged to the TOF camera module and the RGB now Group carries out monocular calibration, and carries out respectively to the binocular calibration of the TOF camera module, this leads to the mark to the TOF device It is very cumbersome to determine process, and inefficiency, needs to expend the more time.Therefore, it is badly in need of a kind of calibration side of TOF device Method and the calibration system for demarcating the TOF device.
Summary of the invention
A purpose of the present invention is that the scaling method and its calibration system of a TOF device are provided, it can be simultaneously to institute It states TOF device and carries out monocular calibration and binocular calibration.
Another object of the present invention is to provide the scaling method of a TOF device and its calibration systems, can be improved institute The efficiency of TOF device calibration is stated, demarcates the time spent by the TOF device to shorten.
Another object of the present invention is to provide the scaling method of a TOF device and its calibration systems, can obtain height The calibration result of precision, to lay a good foundation for subsequent various applications.
Another object of the present invention is to provide the scaling method of a TOF device and its calibration systems, wherein the calibration System can automatically carry out the work for demarcating the TOF device, to save required people during demarcating the TOF device Power and material resources.
Another object of the present invention is to provide the scaling method of a TOF device and its calibration systems, wherein the calibration System can complete the calibration of the TOF device by shirtsleeve operation.
Another object of the present invention is to provide the scaling method of a TOF device and its calibration systems, wherein the calibration System can control the staking-out work of the TOF device by visually operating, difficult with the operation of the simplification calibration system Degree.
Another object of the present invention is to provide the scaling method of a TOF device and its calibration systems, can be right respectively One TOF camera module of the TOF device and a RGB camera module carry out monocular calibration, to obtain the TOF camera shooting mould respectively Group and the respective intrinsic parameter of the RGB camera module and outer parameter.
The present invention is that another object is to provide the scaling method and its calibration system of a TOF device, can be to described One TOF camera module of TOF device and a RGB camera module carry out binocular calibration, to obtain the TOF camera module and described Structural parameters between RGB camera module.
Another object of the present invention is to provide the scaling method of a TOF device and its calibration systems, wherein in order to reach Above-mentioned purpose does not need the structure using expensive material or complexity in the present invention.Therefore, the present invention is successful and effectively mentions For a solution, the scaling method and its calibration system of a simple TOF device are not only provided, while being also added described The scaling method of TOF device and its practicability of calibration system and reliability.
In order to realize above-mentioned at least a goal of the invention or other objects and advantages, the present invention provides the marks of a TOF device Determine method, comprising the following steps:
Acquire at least three TOF images and at least three RGB images；
Based on at least three TOF images, monocular demarcates a TOF camera module of the TOF device, described in obtaining The outer parameter of the TOF intrinsic parameter and TOF of TOF camera module；
Based at least three RGB images, monocular demarcates a RGB camera module of the TOF device, described in obtaining The outer parameter of the RGB intrinsic parameter and RGB of RGB camera module；And
TOF device described in binocular calibration, to obtain the structural parameters of the TOF device.
In one embodiment of this invention, acquisition at least three TOF images and the step of at least three RGB image, are also wrapped Include step:
A target module is configured in the common field of view of the TOF camera module and the RGB camera module；
By the TOF camera module and the RGB camera module, target module described in sync pulse jamming, to acquire a TOF Image and a RGB image；And
Arbitrarily change the position of the target module, and behind the position for changing the target module every time, repeats Described in the sync pulse jamming the step of target module, to acquire at least three TOF images and at least three RGB images.
In one embodiment of this invention, described by the TOF camera module and the RGB camera module, it is synchronous to clap The target module is taken the photograph, to acquire a TOF image and the step of a RGB image, is further comprised the steps of:
By a light source module of the TOF camera module, emit a laser to the target module；With
By a photosensitive control module of the TOF camera module, reception is swashed by reflected this of target module Light, to acquire the TOF image.
In one embodiment of this invention, described by the TOF camera module and the RGB camera module, it is synchronous to clap The target module is taken the photograph, to acquire a TOF image and the step of a RGB image, is further comprised the steps of:
Target module described in light filling.
In one embodiment of this invention, described by the TOF camera module and the RGB camera module, it is synchronous to clap The target module is taken the photograph, to acquire a TOF image and the step of a RGB image, is further comprised the steps of:
The position of the target module is adjusted, so that the target module is completely located at the TOF camera module and institute It states in the common field of view of RGB camera module.
In one embodiment of this invention, the position for arbitrarily changing the target module, and described in each change Behind the position of target module, the step of repeating target module described in the sync pulse jamming, to acquire at least three TOF figure It the step of picture and at least three RGB image, further comprises the steps of:
Change the position of the demarcating module, so that the target module is completely located at the TOF camera module and institute It states in the angular zone in the common field of view of RGB camera module.
In one embodiment of this invention, described that the TOF device is demarcated based at least three TOF images, the monocular One TOF camera module further comprises the steps of: outside the TOF intrinsic parameter and TOF to obtain the TOF camera module the step of parameter
A world coordinate system, a TOF camera module coordinate system, a TOF plane of delineation coordinate system and a TOF pixel is established to sit Mark system；
The TOF pixel coordinate of nonlinear at least four characteristic points on each TOF image is extracted, and acquisition is described at least The world coordinates of four characteristic points；And
Based on the TOF pixel coordinate and the world coordinates, a homography matrix is solved, and according to the homography Matrix finds out the initial TOF intrinsic parameter and the outer parameter of initial TOF of the TOF camera module.
In one embodiment of this invention, described that the TOF device is demarcated based at least three TOF images, the monocular One TOF camera module further comprises the steps of: outside the TOF intrinsic parameter and TOF to obtain the TOF camera module the step of parameter
By the distortion correction model and a data fit object function of the TOF camera module, solves the TOF and take the photograph As the TOF distortion parameter of mould group；With
According to maximum likelihood estimate, optimize the initial TOF intrinsic parameter and the outer parameter of the initial TOF, to obtain State the TOF intrinsic parameter and the outer parameter of the TOF of TOF camera module.
In one embodiment of this invention, described based at least three RGB images, monocular demarcates the TOF device One RGB camera module further comprises the steps of: outside the RGB intrinsic parameter and RGB to obtain the RGB camera module the step of parameter
A world coordinate system, a RGB camera module coordinate system, a RGB image plane coordinate system and a rgb pixel is established to sit Mark system；
The rgb pixel coordinate of nonlinear at least four characteristic points on each RGB image is extracted, and acquisition is described at least The world coordinates of four characteristic points；And
Based on the rgb pixel coordinate and the world coordinates, a homography matrix is solved, and according to the homography square Battle array finds out the Initial R GB intrinsic parameter and the outer parameter of Initial R GB of the RGB camera module.
In one embodiment of this invention, described based at least three RGB images, monocular demarcates the TOF device One RGB camera module further comprises the steps of: outside the RGB intrinsic parameter and RGB to obtain the RGB camera module the step of parameter
By the distortion correction model and a data fit object function of the RGB camera module, finds out the RGB and take the photograph As the RGB distortion correction parameter of mould group；With
According to maximum likelihood estimate, optimize the Initial R GB intrinsic parameter and the outer parameter of the Initial R GB, to obtain State the RGB intrinsic parameter and the outer parameter of the RGB of RGB camera module.
In one embodiment of this invention, TOF device described in the binocular calibration, to obtain the structure of the TOF device The step of parameter, further comprises the steps of:
Down-sampled at least three RGB images, so that the resolution ratio of at least three down-sampled RGB images and described at least three The resolution ratio of TOF image is corresponding；
Based on at least three down-sampled RGB image, monocular demarcates the RGB camera module, to obtain the RGB camera shooting The outer parameter of the down-sampled RGB of mould group；And
Based on parameter outside parameter outside the TOF and the down-sampled RGB, the structural parameters of the TOF device are solved.
In one embodiment of this invention, down-sampled at least three RGB images, so that at least three is down-sampled The resolution ratio of RGB image step corresponding with the resolution ratio of at least three TOF images, further comprises the steps of:
Equal proportion scales the RGB image；With
Cut the RGB image.
In one embodiment of this invention, in down-sampled at least three RGB images, so that at least three is down-sampled In the resolution ratio of RGB image step corresponding with the resolution ratio of at least three TOF images, the down-sampled RGB image packet Include the complete image of the demarcating module.
According to another aspect of the present invention, invention further provides a calibration systems, for demarcating a TOF device, In the TOF device include a TOF camera module and a RGB camera module, comprising:
One target module, wherein the target module is configured in the common of the TOF camera module and the RGB camera module In field of view；With
One control module, wherein the control module includes:
One interactive module, wherein the interactive module is set to be communicatively coupled with the TOF device, it should with acquisition At least three RGB images captured by least three TOF images captured by TOF camera module and the RGB camera module；With
One demarcating module, wherein the demarcating module is communicatively coupled with the interactive module, and the calibration mold Block is demarcated according to a scaling method, wherein the scaling method comprising steps of
Based on at least three TOF images, monocular demarcates the TOF camera module, to obtain the TOF of the TOF camera module Intrinsic parameter and the outer parameter of TOF；
Based at least three RGB images, monocular demarcates the RGB camera module, to obtain the RGB of the RGB camera module Intrinsic parameter and the outer parameter of RGB；And
Binocular calibration TOF device, to obtain the structural parameters of the TOF device.
In one embodiment of this invention, which includes a light source module and a photosensitive control module, wherein The light source module is set to emit a laser towards the target module, which is provided to receive and handles By the reflected laser of the target module, to obtain at least three TOF images by the TOF camera module.
It in one embodiment of this invention, further include a supplementary lighting module, wherein the supplementary lighting module is set with light filling institute State target module.
In one embodiment of this invention, the supplementary lighting module is a flat surface light source, and the supplementary lighting module is stacked In a back side of the target module.
It in one embodiment of this invention, further include a positioning module, wherein the positioning module is set to adjust The position of target module is stated, so that the target module is completely located at the TOF camera module and the RGB camera module is total In the same field of view.
In one embodiment of this invention, the positioning module is set purposefully to change the position of the demarcating module It sets, so that the target module is completely located at the visual field area of the TOF camera module and the RGB camera module jointly In angular zone in domain.
In one embodiment of this invention, the demarcating module according to the scaling method in, it is described be based on institute At least three TOF images are stated, monocular demarcates the TOF camera module, joins outside the TOF intrinsic parameter and TOF to obtain the TOF camera module Several steps, further comprises the steps of:
A world coordinate system, a TOF camera module coordinate system, a TOF plane of delineation coordinate system and a TOF pixel is established to sit Mark system；
The TOF pixel coordinate of nonlinear at least four characteristic points on each TOF image is extracted, and acquisition is described at least The world coordinates of four characteristic points；And
Based on the TOF pixel coordinate and the world coordinates, a homography matrix is solved, and according to the homography Matrix finds out the initial TOF intrinsic parameter and the outer parameter of initial TOF of the TOF camera module.
In one embodiment of this invention, the demarcating module according to the scaling method in, it is described be based on institute At least three TOF images are stated, monocular demarcates the TOF camera module, joins outside the TOF intrinsic parameter and TOF to obtain the TOF camera module Several steps, further comprises the steps of:
By the distortion correction model and a data fit object function of the TOF camera module, solves the TOF and take the photograph As the TOF distortion parameter of mould group；With
According to maximum likelihood estimate, optimize the initial TOF intrinsic parameter and the outer parameter of the initial TOF, to obtain State the TOF intrinsic parameter and the outer parameter of the TOF of TOF camera module.
In one embodiment of this invention, the demarcating module according to the scaling method in, it is described be based on institute At least three RGB images are stated, monocular demarcates the RGB camera module, joins outside the RGB intrinsic parameter and RGB to obtain the RGB camera module Several steps, further comprises the steps of:
A world coordinate system, a RGB camera module coordinate system, a RGB image plane coordinate system and a rgb pixel is established to sit Mark system；
Extract at least rgb pixel coordinate of four characteristic points, and the acquisition at least four characteristic points on each RGB image World coordinates；And
Based on the rgb pixel coordinate and the world coordinates, a homography matrix is solved, and according to the homography square Battle array finds out the Initial R GB intrinsic parameter and the outer parameter of Initial R GB of the RGB camera module.
In one embodiment of this invention, the demarcating module according to the scaling method in, it is described be based on institute At least three RGB images are stated, monocular demarcates the RGB camera module, joins outside the RGB intrinsic parameter and RGB to obtain the RGB camera module Several steps, further comprises the steps of:
By the distortion correction model and a data fit object function of the RGB camera module, solves the RGB and take the photograph As the RGB distortion parameter of mould group；With
According to maximum likelihood estimate, optimize the Initial R GB intrinsic parameter and the outer parameter of the Initial R GB, to obtain State the RGB intrinsic parameter and the outer parameter of the RGB of RGB camera module.
In one embodiment of this invention, the demarcating module according to the scaling method in, the Bi-objective It the step of fixed TOF device, structural parameters to obtain the TOF device, further comprises the steps of:
Down-sampled at least three RGB images, so that the resolution ratio of at least three down-sampled RGB images and described at least three The resolution ratio of TOF image is corresponding；
Based on at least three down-sampled RGB image, monocular demarcates the RGB camera module, to obtain the RGB camera module The outer parameter of down-sampled RGB；And
Based on parameter outside parameter outside the TOF and the down-sampled RGB, the structural parameters of the TOF device are solved.
In one embodiment of this invention, the demarcating module according to the scaling method in, it is described down-sampled At least three RGB images, so that the resolution of the resolution ratio of at least three down-sampled RGB images and at least three TOF images The corresponding step of rate, further comprises the steps of:
Equal proportion scales the RGB image；With
Cut the RGB image.
In one embodiment of this invention, at least three down-sampled RGB image includes the complete of the demarcating module Image.
It in one embodiment of this invention, further include a light-adjusting module, wherein the light-adjusting module is set according to institute The brightness of the image of target module described in TOF image is stated to automatically adjust the described of target module described in the TOF image The brightness of image.
By the understanding to subsequent description and attached drawing, further aim of the present invention and advantage will be fully demonstrated.
These and other objects of the invention, feature and advantage, by following detailed descriptions, drawings and claims are obtained To fully demonstrate.
Detailed description of the invention
Fig. 1 is the stereoscopic schematic diagram of a TOF device.
Fig. 2 is the flow diagram of the scaling method of a TOF device according to a preferred embodiment of the present invention.
Fig. 3 is an acquisition step of the scaling method of the TOF device of above-mentioned preferred embodiment according to the present invention Flow diagram.
Fig. 4 is the monocular calibration step of the scaling method of the TOF device of above-mentioned preferred embodiment according to the present invention Rapid flow diagram.
Fig. 5 is another monocular calibration of the scaling method of the TOF device of above-mentioned preferred embodiment according to the present invention The flow diagram of step.
Fig. 6 is the binocular calibration step of the scaling method of the TOF device of above-mentioned preferred embodiment according to the present invention Rapid flow diagram.
Fig. 7 is the schematic illustration of the scaling method of the TOF device of above-mentioned preferred embodiment according to the present invention, is shown The calibration principle of a TOF camera module is gone out.
Fig. 8 is the schematic illustration of the scaling method of the TOF device of above-mentioned preferred embodiment according to the present invention, is shown The calibration principle of a RGB camera module is gone out.
Fig. 9 is the stereoscopic schematic diagram of a target module of a calibration system of above-mentioned preferred embodiment according to the present invention.
Figure 10 is that the calibration system of above-mentioned preferred embodiment according to the present invention acquires the demarcating module at one The status diagram for the image set.
Figure 11 is that the calibration system of above-mentioned preferred embodiment according to the present invention acquires the demarcating module another The status diagram of the image of position.
Figure 12 be above-mentioned preferred embodiment according to the present invention the demarcating module at the position calibration system It unites TOF image schematic diagram collected.
Figure 13 be above-mentioned preferred embodiment according to the present invention the demarcating module at the position calibration system It unites RGB image schematic diagram collected.
Figure 14 be above-mentioned preferred embodiment according to the present invention the demarcating module at the another location mark Determine system TOF image schematic diagram collected.
Figure 15 be above-mentioned preferred embodiment according to the present invention the demarcating module at the another location mark Determine system RGB image schematic diagram collected.
Figure 16 shows in the scaling method of above-mentioned preferred embodiment according to the present invention and carries out to the RGB image Down-sampled status diagram.
Figure 17 is the block diagram representation of the calibration system of above-mentioned preferred embodiment according to the present invention.
Specific embodiment
It is described below for disclosing the present invention so that those skilled in the art can be realized the present invention.It is excellent in being described below Embodiment is selected to be only used as illustrating, it may occur to persons skilled in the art that other obvious modifications.It defines in the following description Basic principle of the invention can be applied to other embodiments, deformation scheme, improvement project, equivalent program and do not carry on the back Other technologies scheme from the spirit and scope of the present invention.
It will be understood by those skilled in the art that in exposure of the invention, term " longitudinal direction ", " transverse direction ", "upper", The orientation of the instructions such as "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom" "inner", "outside" or position are closed System is to be based on the orientation or positional relationship shown in the drawings, and is merely for convenience of description of the present invention and simplification of the description, without referring to Show or imply that signified device or element must have a particular orientation, be constructed and operated in a specific orientation, therefore above-mentioned art Language is not considered as limiting the invention.
In the present invention, term " one " is interpreted as " one or more " in claim and specification, i.e., in a reality Example is applied, the quantity of an element can be one, and in a further embodiment, the quantity of the element can be multiple.Unless Clearly illustrate in exposure of the invention the element quantity only one, otherwise term " one " can not be interpreted as unique or single One, term " one " should not be understood as the limitation to quantity.
Camera calibration typically refers to establish video camera geometry imaging model, describes world coordinate system (i.e. space coordinates) In scene point with the corresponding relationship between its picture point on the image plane.In other words, camera calibration is substantially determining One process of the intrinsic parameter of the video camera and outer parameter, wherein the intrinsic parameter of video camera is that video camera is intrinsic, and with position Confidence ceases unrelated parameter, generally includes focal length, lens distortion, image center etc.；And the outer parameter of video camera is video camera Coordinate system is relative to the three-dimensional position of a certain world coordinate system and the relationship in direction.Accurately solving these parameters not only can be improved The precision of three dimensional signal space, and can lay a good foundation for subsequent Stereo image matching with three-dimensional reconstruction.
However, TOF device (such as TOF camera or TOF machine vision etc.) generally includes a TOF camera module and one RGB camera module, wherein the TOF camera module is used to shoot a testee, to obtain the depth information of the testee With gray level image (i.e. TOF image), the RGB camera module is for shooting the testee, to obtain the coloured silk of the testee Chromatic graph picture (i.e. RGB image), and then depth information and color image by handling the testee, to obtain the testee The color image with depth information, to reinforce and expand the practical application of the TOF device.Therefore, in order to ensure institute The normal use of TOF device is stated, and improves the service precision of the TOF device, so need to carry out the TOF device complete The calibration in face.That is, not only individually to be demarcated (i.e. single goal to the TOF camera module of the TOF device It is fixed), with obtain the TOF camera module intrinsic parameter and outer parameter, will also be to the RGB camera module of the TOF device Individually demarcated (i.e. monocular calibration), with obtain the TOF camera module intrinsic parameter and outer parameter, and inevitably Also binocular calibration is carried out to the TOF device, to obtain the structural parameters of the TOF device, i.e., the described TOF device it is described Positional relationship between TOF camera module and the RGB camera module.
More specifically, as shown in Figure 1, the TOF device 10 includes a TOF camera module 11 and a RGB camera module 12. The TOF camera module 11 includes a light source module 111 and a photosensitive control module 112, and the light source module 111 is set It sets to generate the laser with preset wavelength to a measured target, the photosensitive control module 112 is provided to receive and locates The laser of reason measured target reflection, to obtain the depth information and gray level image of the measured target.The RGB camera module 12 can shoot the measured target, to obtain the color image of the measured target.It will be appreciated by those skilled in the art that of the invention The laser that the light source module 111 of the TOF camera module 11 issues (transmitting) can be infrared laser.Preferably, described Light source module 111 is implemented as a vertical cavity surface laser emitter (abbreviation VCSEL).
It is worth noting that, the measured target is implemented as a target mould during demarcating TOF device 10 Block 20, and the target module 20 has nonlinear at least four characteristic points P, with by the target module 20 to described The TOF camera module 11 and the RGB camera module 12 of TOF device 10 are demarcated, with reference to attached drawing 9 and Figure 10.It is preferred that Ground, as shown in figure 9, the target module 20 is carried out and is a flat surface gridiron pattern target 21, wherein the plane gridiron pattern target 7 × 10 chequered with black and white grids (the i.e. described plane gridiron pattern target 21 has 7 row grids and 10 column grids) are provided on 21, And the common angle point of adjacent black and white grid is implemented as the characteristic point P, and therefore, the target module 20 has 6*9 Characteristic point P.It will be appreciated by those skilled in the art that the line number of the black and white grid on heretofore described plane gridiron pattern target 21 Exemplary only with columns, the specific number and size of black and white grid are selected and are adjusted according to specific needs, This does not do further limitation.
With reference to shown in Fig. 2 to Figure 17 of attached drawing, the calibration of a TOF device of preferred embodiment according to the present invention is shown Method and its calibration system.Preferred embodiment according to the present invention, as shown in Fig. 2, the scaling method of the TOF device 10 The following steps are included:
S1: acquisition at least three TOF images 100 and at least three RGB images 200.Specifically, as shown in figure 3, the step S1 It further includes steps of
S11: one target module 20 of the configuration visual field common in the TOF camera module 11 and the RGB camera module 12 In region.Specifically, as shown in Figure 10 and Figure 11, the TOF device 10 is opened, is imaged in the TOF of the TOF device 10 In mould group 11 and the common field of view of the RGB camera module 12, the target module 20 is placed without restrictions, so that institute It states TOF camera module 11 and the RGB camera module 12 can while shoot the target module 20, in order to obtain Obtain the image 200 of the target module 20.
S12: by the TOF camera module 11 and the RGB camera module 12, target module 20 described in sync pulse jamming, To acquire a TOF image 110 and a RGB image 120.Specifically, when the TOF camera module 11 is shot, the TOF camera shooting The light source module 111 of mould group 11 is set to emit laser towards the target module 20, and the laser encounters the mark Plate module 20 will be reflected back, and the photosensitive control module 112 of the TOF camera module 11 is provided to receive and handles By the reflected laser of the target module 20, to obtain (the i.e. described mark of described image 200 of the target module 20 The gray level image of plate module 20), to obtain the TOF image 110 captured by the TOF camera module 11, that is to say, that Due to the 11 automatic light source module 111 of TOF camera module, the TOF camera module 11 to the target module 20 into When row shooting, do not need by any external light source, it will be able to obtain the TOF image 110.At the same time, the TOF device The 10 RGB camera module 12 shoots the target module 20, to obtain the described image of the target module 20 200 (color images of the i.e. described target module 20), and then obtain the RGB image captured by the RGB camera module 12 120。
It is worth noting that, as shown in Figure 10 and Figure 11, in order to enable the TOF image 110 and the RGB image 120 Comprising all characteristic point P in the target module 20, to improve the precision of follow-up calibration result to the maximum extent, because This, when being shot using the TOF device to the target module 20, should target module 20 described in appropriate adjustment position (including the distance between the target module 20 and the TOF device 10, direction and angle) is set, to ensure the target mould Block 20 is completely in the TOF camera module 11 and the common field of view of the RGB camera module, and then is ensured described TOF image 110 includes the complete image 200 of the target module 20, while the RGB image 120 also includes the target mould The complete image 200 of block 20.
It is noted that since the laser intensity that the light source module 111 of the TOF camera module 11 is emitted is logical Chang Buhui is too high, therefore the TOF camera module 11 can only could shoot and obtain in the position close to the target module 20 The TOF image 110 of high quality.However, in order to widen the phase between the TOF camera module 11 and the target module 20 To positional relationship, light filling can be carried out to the target module 20, to increase the brightness of the target module 20, so that institute The target module 20 can shot apart from 20 remote position of target module by stating TOF camera module 11, and can be obtained The TOF image 110 of high quality, by subsequent calibration and usage.
S13: arbitrarily changing the position of the target module 20, and behind the position for changing the target module 20 every time, The step S12 is repeated, to acquire at least three TOF images 110 and at least three RGB images 120.It should manage Solution, since Zhang Zhengyou images the inside and outside parameter that calibration technique only needs three width images that can solve video camera, at this In the step S1 of invention, need to obtain RGB image described in TOF image described at least three width and at least three width simultaneously, to divide It is other that monocular calibration is carried out to the TOF camera module and the RGB camera module.Certainly, in the step S13, Ke Yiwu Limit changes the position of the target module 20 secondaryly, with acquisition without TOF image 110 described in clipping and the RGB image 120, with Improve the precision for finally solving obtained inside and outside parameter.However, in order to obtain high-precision inside and outside parameter simultaneously, and will not be big Width increases the difficulty of staking-out work, in the step S1 of the invention, preferably obtains TOF image 110 described in 10~20 width With RGB image 120 described in 10~20 width corresponding with the TOF image.
It is worth noting that, as shown in fig. 10 and fig. 12, obtaining the TOF by the scaling method to further increase The precision of the inside and outside parameter of camera module 11, in the step S1, it should purposefully adjust the position of the target module 20 It sets, so that the target module 20 is completely located at common described of the TOF camera module 11 and the RGB camera module 12 In field of view, so that it is guaranteed that the target module 20 is respectively in the photosensitive control module 112 of the TOF camera module 11 Sensitive chip on each region in be imaged, with obtain to greatest extent the sensitive chip of the TOF camera module 11 at The complete information of picture, and then improve the stated accuracy of the TOF camera module 11.That is, by purposefully adjusting institute The position of target module 20 is stated, so that in the different TOF images 110, the described image of the target module 20 200 In the different zones of the TOF image 110.
Preferably, as shown in Figure 11 and Figure 14, there should be the target module 20 in all TOF images 110 Described image 200 be located at the TOF image 110 angular zone image.It is highly preferred that in all TOF images 110 At least there is TOF image 110 described in four width, so that the image 200 of the target module 20 is located at the TOF image 110 Four angular zones.
Similarly, such as Figure 10 and Figure 13 in order to further increase by the scaling method obtain as described in RGB camera module 12 Inside and outside parameter precision, in the step S1, it should the position for purposefully adjusting the target module 20 so that described Target module 20 is completely located in the field of view of the TOF camera module 11 and the RGB camera module 12 jointly, So that it is guaranteed that the target module 20 is imaged in each region on the sensitive chip of the RGB camera module 12 respectively, with The complete information of the sensitive chip imaging of the RGB camera module 12 is obtained to greatest extent, and then improves the RGB camera shooting The stated accuracy of mould group 12.That is, by the position for purposefully adjusting the target module 20, so that in different institutes It states in RGB image 120, the described image 200 of the target module 20 is located at the different zones of the RGB image 120.It should manage The field of view of the TOF camera module 11 is usually completely covered in solution, the field of view of the RGB camera module 12, therefore only It is necessary to ensure that the target module 20 always in the field of view in the TOF camera module 11.
Preferably, as shown in Figure 11 and Figure 15, there should be the target module 20 in all RGB images 120 Described image 200 be located at the RGB image 120 angular zone image.It is highly preferred that in all RGB images 120 At least there is RGB image 120 described in four width, so that the image 200 of the target module 20 is located at the RGB image 120 Four angular zones.
S2: based on at least three TOF images 110, monocular demarcates the TOF camera module 11 of the TOF device 10, Parameter outside TOF intrinsic parameter and TOF to obtain the TOF camera module 11.Calibration technique is imaged according to Zhang Zhengyou, such as Fig. 4 institute Show, the step S2 of the scaling method described in the preferred embodiment of the invention is further included steps of
S21: a world coordinate system, a TOF camera module coordinate system, a TOF plane of delineation coordinate system and a TOF are established Pixel coordinate system.Citing ground, as shown in figures 7 and 9, establishes the world coordinate system according to the position of the target module 20 Ow-XwYwZW, wherein origin (Ow) be the target module 20 an angle point, XwAxis and YwAxis respectively with the target module 20 Row and column it is parallel；The TOF camera module coordinate system O is established according to the placement position of the TOF camera module 11Tc- XTcYTcZTc, wherein origin (OTc) be the TOF camera module 11 optical center, XTcAxis and YTcAxis respectively with the TOF image Row and column is parallel；The TOF plane of delineation coordinate system O is established according to the placement position of the TOF camera module 11Ti-xTyT, Middle origin (OTi) it is the optical axis of the TOF camera module 11 and the intersection point of TOF imaging plane, xTAxis and yTAxis respectively with XTcAxis and YTcAxis is parallel and is directed toward opposite；The TOF pixel coordinate system O is established according to the TOF imageT-UTVT, wherein origin (OT) position In on an angle of TOF picture element matrix, UTAxis and VTAxis is respectively parallel to xTAxis and yTAxis and be directed toward it is identical.It is worth noting that, Characteristic point P in the target module 20 is in the world coordinate system Ow-XwYwZWOn world coordinates PW=(Xw,Yw,Zw), with Accordingly, the characteristic point P is in the TOF camera module coordinate system OTc-XTcYTcZTcOn TOF module coordinate PTc=(XTc, YTc,ZTc)；The image of the characteristic point P is in the TOF plane of delineation coordinate system OTi-xTyTOn TOF image coordinate PTi=(xT, yT)；The image of the characteristic point P is in the TOF pixel coordinate system OT-UTVTOn TOF pixel coordinate PT=(UT,VT)。
S22: the TOF pixel coordinate of nonlinear at least four characteristic points P on each TOF image 110 is extracted, and is obtained The world coordinates of at least four characteristic points P.Specifically, according to Zhang Zhengyou calibration technique, TOF image described in every width need to only be extracted The TOF pixel coordinate P that the characteristic point P of four non-linear relations is fastened in the TOF pixel coordinate on 110T, and calculate described World coordinates P of the characteristic point P in the world coordinate systemw, it will be able to the TOF camera module 11 is found out by homography calculating Inside and outside parameter.In order to improve the obtained TOF camera module 11 inside and outside parameter precision, the extracted feature The quantity of point P is The more the better, therefore, in the preferred embodiment of the invention, preferably extracts in the TOF image and owns Characteristic point P.
S23: it is based on the TOF pixel coordinate and the world coordinates, solves a homography matrix, and according to the list Answering property matrix finds out the initial TOF intrinsic parameter and the outer parameter of initial TOF of the TOF camera module 11.Specifically, it is assumed that institute In the case where TOF camera module is stated there is no distorting, according to the TOF picture of the characteristic point P on TOF image described in every width Plain coordinate PTWith the world coordinates PwHomography calculating is carried out, to find out the homography matrix H of the TOF camera module 11T= sMT[RT,TT], and then according to homography matrix HT, acquire the initial TOF intrinsic parameter and initial TOF of the TOF camera module 11 Outer parameter.The formula that wherein homography calculates is as follows:
PT=HTPw
In formula: homography matrix HT=sMT[RT,TT], wherein s is scale factor；MTFor the TOF camera module 11 TOF Intrinsic Matrix；[RT,TT] be the TOF camera module 11 the outer parameter matrix of TOF, wherein RTMould is imaged for the TOF Group 11 arrives the spin matrix of the world coordinate system, TTThe translation vector of the world coordinate system is arrived for the TOF camera module 11 Amount.
It is worth noting that, solving the homography matrix HTDuring, it is assumed that the TOF camera module is not present Distortion, that is to say, that, it is assumed that the TOF distortion parameter γ in the TOF intrinsic parameter of the TOF camera module 11T=0, with letter Change calculating process, thereby reduces solution difficulty.
However, due to assuming the TOF camera module in the step S23, there is no distortion, therefore obtain the TOF internal reference TOF distortion parameter in number is zero not meet actual conditions, therefore the initial intrinsic parameter of the TOF cannot function as final institute The final argument for stating TOF camera module needs to solve the TOF camera module.
S24: by the distortion correction model and a data fit object function of the TOF camera module 11, institute is found out State the TOF distortion parameter γ of TOF camera module 11T.Specifically, according to the characteristic point P on the TOF image 110 in institute State TOF plane of delineation coordinate system OTi-xTyTOn TOF actual coordinate (xTi,yTi), using described in the TOF camera module 11 Distortion model calculates the TOF standardization coordinate (X of the characteristic point PTi,YTi), it finally will be described in this two groups of coordinate values substitutions The data fit object function of TOF camera module 11 carries out data fitting, final to calculate the corresponding TOF distortion ginseng Number, so that the TOF distortion parameter in the initial intrinsic parameter of the TOF is not zero.It should be appreciated that working as the data fit object function Obtained value is smaller, it was demonstrated that data are fitted better, and correspondingly, the TOF distortion parameter acquired is also more accurate.In addition, By the method for this successive ignition, direct solution nonlinear distortion varying model is not only avoided, but also calculating is switched to solve non- Linear least square problem makes the smallest distortion parameter value γ of target function value to finally obtainT=[kT1,kT2,kT3,pT1, pT2], which not only simplifies calculating, also reduce solution difficulty.
It is worth noting that, the distortion correction model of the TOF camera module is as follows:
In formula:For the TOF actual coordinate of the characteristic point P, (XTi,YTi) it is the feature The TOF standardization coordinate of point P.
The data fit object function of the TOF camera module is as follows:
S25: according to maximum likelihood estimate, optimize the initial TOF intrinsic parameter and the outer parameter of the initial TOF, to obtain Obtain the TOF intrinsic parameter and the outer parameter of the TOF of the TOF camera module 11.In the preferred embodiment of the invention, The TOF intrinsic parameter include it is optimised after TOF distortion parameter γT, the outer parameter of the TOF includes that the TOF camera module is sat Transformation relation between mark system and the world coordinate system, i.e. TOF spin matrix RTWith TOF translation vector TT。
S3: based at least three RGB images 120, monocular demarcates a RGB camera module 12 of the TOF device 10, with Obtain the RGB intrinsic parameter and the outer parameter of RGB of the RGB camera module 12.It is worth noting that, monocular demarcates the RGB camera shooting The method and steps of mould group 12 is identical as the monocular calibration TOF camera module 11, specifically, as shown in figure 5, of the invention The step S3 of scaling method described in the preferred embodiment is further included steps of
S31: the world coordinate system, a RGB camera module coordinate system, a RGB image plane coordinate system and one are established Rgb pixel coordinate system.Citing ground, as shown in Figure 8 and Figure 9, according to the orientation of the target module 20 of the target unit 20 Establish the world coordinate system Ow-XwYwZW, wherein origin (Ow) be the target module 20 an angle point, XwAxis and YwAxis point It is not parallel with the row and column of the target module 20；The RGB camera shooting is established according to the placement position of the RGB camera module 12 Module coordinate system ORc-XRcYRcZRc, wherein origin ORcFor the optical center of the RGB camera module 12, XRcAxis and YRcAxis respectively with institute The row and column for stating RGB image is parallel；The RGB image plane coordinates is established according to the placement position of the RGB camera module 12 It is ORi-xRyR, wherein origin (ORi) it is the optical axis of the RGB camera module 12 and the intersection point of RGB imaging plane, xRAxis and yRAxis Respectively with XRcAxis and YRcAxis is parallel and is directed toward opposite；The rgb pixel coordinate system O is established according to the RGB imageR-URVR, Middle origin (OR) on an angle of the picture element matrix of RGB image, URAxis and VRAxis is respectively parallel to xRAxis and yRAxis and direction It is identical.It is worth noting that, the characteristic point P in the target module 20 is in the world coordinate system Ow-XwYwZWOn the world sit Mark PW=(Xw,Yw,Zw), correspondingly, the characteristic point P is in the RGB camera module coordinate system ORc-XRcYRcZRcOn RGB module coordinate PRc=(XRc,YRc,ZRc)；The image of the characteristic point P is in the RGB image plane coordinate system ORi-xRyROn RGB image coordinate PRi=(xR,yR)；The image of the characteristic point P is in the rgb pixel coordinate system OR-URVROn RGB picture Plain coordinate PR=(UR,VR)。
S32: the rgb pixel coordinate of nonlinear at least four characteristic points P on each RGB image 120 is extracted, and is obtained The world coordinates of at least four characteristic points P.Specifically, according to Zhang Zhengyou calibration technique, RGB image described in every width need to only be extracted Rgb pixel coordinate of the characteristic point P of four non-linear relations on the rgb pixel coordinate system on 120, and calculate the spy Point P is levied in the world coordinates P of the world coordinate systemw, it will be able to the RGB camera module 12 is found out by homography calculating Inside and outside parameter.In order to improve the obtained RGB camera module 12 inside and outside parameter precision, the extracted characteristic point The quantity of P is The more the better, therefore, in the preferred embodiment of the invention, preferably extracts institute in the RGB image 120 Some characteristic point P.
S33: being based on the rgb pixel coordinate and the world coordinates, solves a homography matrix, and answer according to the list Property matrix, find out the Initial R GB intrinsic parameter and the outer parameter of Initial R GB of the RGB camera module 12.Specifically, it is assumed that described In the case that RGB camera module is there is no distorting, according to the rgb pixel of the characteristic point P on RGB image 120 described in every width Coordinate and world coordinates carry out homography calculating, to find out the homography matrix H of the RGB camera module 12R=sMR[RR, TR], and then according to homography matrix HR, acquire the Initial R GB intrinsic parameter and the outer parameter of Initial R GB of the RGB camera module 12. Specifically, the formula that homography calculates is as follows:
PR=HRPw
In formula: homography matrix HR=sMR[RR,TR], wherein s is scale factor；MRFor the RGB camera module 12 RGB Intrinsic Matrix；[RR,TR] be the RGB camera module 12 the outer parameter matrix of RGB, wherein RRMould is imaged for the RGB Group 12 arrives the spin matrix of the world coordinate system, TRThe translation vector of the world coordinate system is arrived for the RGB camera module 12 Amount.
It is worth noting that, solving the homography matrix HTDuring, it is assumed that the RGB camera module is not present Distortion, that is to say, that, it is assumed that the RGB distortion parameter γ in the RGB intrinsic parameter of the RGB camera module 12R=0, with letter Change calculating process, thereby reduces solution difficulty.
However, due to assuming the RGB camera module 12 in the step S33, there is no distortion, therefore obtain in the RGB RGB distortion parameter in parameter is zero, does not meet actual conditions, therefore the initial intrinsic parameter of the RGB cannot function as final institute The final argument for stating RGB camera module 12 needs to solve the distortion parameter of the RGB camera module 12.
S34: by the distortion correction model and a data fit object function of the RGB camera module 12, institute is found out State the RGB distortion parameter γ of RGB camera module 12R。
Specifically, according to a certain characteristic point P on the RGB image 120 in the RGB image plane coordinate system ORi-xRyRActual coordinate (xRi,yRi), the characteristic point P is calculated using the distortion model of the RGB camera module 12 Standardization coordinate (XRi,YRi), the data that this two groups of coordinate values substitute into the RGB camera module 12 are finally fitted mesh Scalar functions carry out data fitting, final to calculate the corresponding RGB distortion parameter γR, so that in the initial intrinsic parameter of the RGB The RGB distortion parameter is not zero.It should be appreciated that when the value that the data fit object function obtains is smaller, it was demonstrated that data are quasi- Close better, correspondingly, the RGB distortion parameter γ acquiredRAlso more accurate.In addition, passing through the side of this successive ignition Method not only avoids direct solution nonlinear distortion varying model, but also calculating is switched to solve nonlinear least square problem, with Finally obtaining makes the smallest RGB distortion parameter value γ of target function valueR=[kR1,kR2,kR3,pR1,pR2], which not only simplifies It calculates, also reduces solution difficulty.
It is worth noting that, the distortion correction model of the RGB camera module is as follows:
In formula:For the actual coordinate of the characteristic point P, (XRi,YRi) it is the characteristic point The standardization coordinate of P.
The data fit object function of the RGB camera module is as follows:
S35: according to maximum likelihood estimate, optimize the Initial R GB intrinsic parameter and the outer parameter of the Initial R GB, to obtain Obtain the RGB intrinsic parameter and the outer parameter of the RGB of the RGB camera module 12.In the preferred embodiment of the invention, In the RGB intrinsic parameter include it is optimised after the RGB distortion parameter γR.The outer parameter of the RGB includes the RGB camera shooting Transformation relation between module coordinate system and the world coordinate system, i.e. RGB spin matrix RRWith RGB translation vector TR。
It is worth noting that, the step S2 and the step S3 are in no particular order in the scaling method of the TOF device Order, that is to say, that the step S2 can be first carried out, execute the step S3 afterwards；The step S3 can also be first carried out, after Execute the step S2；It may also be performed simultaneously the step S2 and the step S3.
S4: TOF device 10 described in binocular calibration, to obtain the structural parameters of the TOF device 10.Specifically, Bi-objective Surely be that basis is demarcated as with monocular, during being demarcated according to monocular outside the TOF obtained outside parameter and the RGB parameter come Acquire the positional relationship between the TOF camera module 11 and the RGB camera module 12, that is, the institute of the TOF device State structural parameters.Stated differently, since the positional relationship packet between the TOF camera module 11 and the RGB camera module 12 Spin matrix R and translation vector T between the two is included, therefore the structural parameters of the TOF device 10 include the TOF camera shooting mould Spin matrix R and translation vector T between group 11 and the RGB camera module 12.
It is worth noting that, in the TOF device 10 described in binocular calibration, need to keep the TOF image 110 and described RGB image 120 it is equal in magnitude, i.e., the resolution ratio (i.e. pixel) of the described TOF image 110 and the RGB image 120 will keep phase Together.However, it is generally the case that the resolution ratio of the RGB image 120 of the TOF device 10 is greater than the TOF image 110 Resolution ratio, therefore, before carrying out binocular calibration, first have to carry out the RGB image 120 it is down-sampled, with obtain with it is described The identical RGB image of the resolution ratio of TOF image 110.As shown in fig. 6, in the preferred embodiment of the invention, the step S4 is further included steps of
S41: down-sampled at least three RGB images 120, so that the resolution ratio of at least three down-sampled RGB images 120 ' It is corresponding with the resolution ratio of at least three TOF images 110.Citing ground, as shown in Figure 13 and Figure 16, the step S41 is into one Step the following steps are included:
Equal proportion scales the RGB image 120, so that the width W ' of the RGB image after scaling is equal to the TOF image 110 width W；With
The RGB image 120 is cut, so that the length L ' of the down-sampled RGB image 120 ' is equal to the TOF image 110 length L.In other words, by the width W ' and length L ' of the down-sampled down-sampled RGB image 120 ' obtained point Not Deng Yu the TOF image 110 width W and length L.
Citing ground, the resolution ratio of the TOF camera module 11 of the TOF device 10 TOF image 110 obtained Resolution ratio for 224*172, the RGB camera module 12 of the TOF device 10 RGB image 120 obtained is 1344*760, therefore before carrying out binocular calibration need to carry out the RGB image 120 down-sampled.Specifically, first equal proportion contracting Put the RGB image 120, with the reduced width of the RGB image 120 to 172, the resolution ratio of the RGB image obtained at this time For 304*172, then 40 pixels of cutting each to RGB image preferably left and right, to obtain the down-sampled RGB image 120 ', The resolution ratio of the down-sampled RGB image 120 ' described at this time is 224*172, that is to say, that by down-sampled to obtain and the TOF The RGB image 120 of 110 same size of image.It will be appreciated by those skilled in the art that can also first to the RGB image 120 into Row is cut, and then carries out scaled down to the RGB image after cutting again, identical as the TOF image 110 big equally to obtain Small RGB image 120, and then realize the purpose of the down-sampled RGB image 120.
Significantly, since the step 41 it is down-sampled during need to cut out the RGB image 120 It cuts, and all or part of image 200 of the target module 20 in the RGB image 120 may be cropped, to make The image 200 for obtaining the target module 20 is imperfect in the down-sampled RGB image 120 ', therefore in order to enable after cutting RGB image still includes the complete image 200 of the target module 20, shoots the target module in the RGB camera module 12 When 20, it is necessary to which the position for keeping the target module 20 to be imaged on the RGB image is not in the RGB image 120 Cropped region is needed, to guarantee the complete of the image 200 of the target module 20 described in the down-sampled RGB image 120 ' Whole property, to prevent the precision for causing to reduce TOF device described in binocular calibration because down-sampled.
S42: based on at least three down-sampled RGB images 120 ', monocular demarcates the RGB camera module 12, to obtain The outer parameter of the down-sampled RGB of the RGB camera module 12.It is worth noting that, the monocular of the step S42 demarcates the RGB The process of camera module 12 is identical with the step S3, and only image based on monocular calibration is different, the step Image based on monocular calibration is the RGB image 120 in rapid S3, and is schemed based on monocular calibration in the step S42 As being the down-sampled RGB image 120 ', parameter outside the down-sampled RGB finally to obtain the RGB camera module 12, Described in the outer parameter of down-sampled RGB include it is down-sampled after the RGB camera module coordinate system and the world coordinate system between Transformation relation, i.e., down-sampled RGB spin matrix RRJWith down-sampled RGB translation vector TRJ。
S43: based on the outer parameter of the TOF and the outer parameter of the down-sampled RGB, the knot of the TOF device 10 is solved Structure parameter.Specifically, the purpose of TOF device 10 described in binocular calibration is to calculate TOF camera module described in the TOF device 10 Relative positional relationship between 11 and the RGB camera module 12 seeks the TOF camera module 11 and RGB camera shooting Spin matrix R and translation vector T between mould group 12.In other words, the structural parameters of the TOF device 10 include described The spin matrix R and translation vector T, wherein the formula for calculating the spin matrix R and the translation vector T is as follows:
T=TT-R-TTRJ
In formula: RRJAnd TRJIt is found out by the monocular calibration in the step S42, the RTAnd TTPass through the step S24 It finds out.
According to another aspect of the present invention, in order to preferably realize above-mentioned scaling method, easily to the TOF device It is demarcated, the present invention further provides a calibration systems, described for being demarcated according to the scaling method of above-mentioned TOF device TOF device 10.Preferred embodiment according to the present invention, as shown in figure 17, the calibration system include a target module 20 With a control module 30.The target module 20 is movably disposed in the TOF camera module 11 of the TOF device 10 With the RGB camera module 12 in common field of view, enable the target module 20 by the institute of the TOF device 10 TOF camera module 11 and 12 sync pulse jamming of RGB camera module are stated, to obtain the image 200 of the target module 20, thus By the TOF camera module 11 and 12 synchronous acquisition TOF image 110 of the RGB camera module and RGB image 120.The control Molding block 30 is set to be communicatively coupled with the TOF device 10, and the control module 30 is enabled to acquire the TOF Image 110 and the RGB image 120, and the control module 30 can be based on the TOF image 110 and the RGB image 120 pairs of TOF devices 10 are demarcated, to obtain the TOF intrinsic parameter of the TOF camera module 11 of the TOF device 10 With parameter outside TOF, the outer parameter of the RGB intrinsic parameter and RGB of the RGB camera module 12 of the TOF device 10 and described The structural parameters of TOF device 10.It will be appreciated by those skilled in the art that the control module 30 can be, but not limited to include an intelligence Energy terminal, such as laptop, desktop computer, smart phone, Ipad or tablet computer etc. can also include various auxiliary Caliberating device is helped, to realize optimal scaling scheme.Preferably, the control module 30 can by USB data line with it is described TOF device 10 connects.Certainly, the side that the control module 30 can also be wirelessly connected by such as WIFI, infrared, bluetooth etc. Formula is connect with the TOF device 10.
Specifically, preferred embodiment according to the present invention, the control module 30 include an interactive module 31 and one Demarcating module 32.The interactive module 31 can be communicatively coupled with the TOF device 10, to acquire the TOF camera module 11 At least three RGB images 120 captured by the captured at least three TOF images 110 and the RGB camera module 12, That is the interactive module 31 is arranged to perform the step S1 in the scaling method of the TOF device, obtained with realizing Obtain the purpose of at least three TOF images 110 and at least three RGB images 120.It should be appreciated that citing ground, the interactive mould Block 31 obtains RGB image captured by TOF image 110 captured by the TOF camera module 11 and the RGB camera module 12 After 120, the interactive module 31 can also TOF image 110 and the RGB image 120 described in simultaneous display, so that user is straight It connects and observes the image 200 of the target module 20 position in the TOF image 110 and the RGB image 120 respectively, and And according to position of the image 200 of the target module 20 in the TOF image 110 and the RGB image 120 to adjust The position of target module 20 is stated, so that it is guaranteed that the TOF image 110 and the RGB image 120 include the target module 20 Complete image 200.Preferably, the control module 30 can store the TOF image 110 and the RGB image 120 Come, in case using.
It is noted that as shown in figure 16, the interactive module 31 of the control module 30 can also be based on described The size of TOF image divides the RGB image 120, and the RGB image 120 is divided into an at least clipping region 1201 With a non-clipping region 1202, wherein the resolution ratio of the RGB image 120 in the non-clipping region 1202 and the TOF image 110 resolution ratio linearly, is schemed with will pass through the down-sampled module 353 for the RGB shown in the non-clipping region As zooming to the image with the resolution ratio same size of the TOF image, i.e., the described down-sampled RGB image.It is worth noting that, In order to avoid falling the image cropping of the target module 20 in down-sampled RGB image, in the side for changing the target module 20 When position, therefore, to assure that the image of the target module 20 is in always in the non-clipping region.
It should be appreciated that the interactive module 31 of the control module 30 can also transmit a control to the TOF device 10 Signal, and by the control signal, the control module 30 can control the opening and closing of the TOF device, thus It realizes and remotely controls the opening and closing of the TOF device, with step needed for simplifying the calibration TOF device.
Preferred embodiment according to the present invention, the demarcating module 32 of the calibration system and the interactive module 31 are communicatively coupled, to be based on the TOF image 110 and the RGB image 120, to demarcate the TOF camera module respectively 11 and the RGB camera module, to obtain the inside and outside parameter of the TOF camera module 11, inside and outside the RGB camera module 12 The structural parameters of parameter and the TOF device 10.In other words, the demarcating module 32 is arranged to perform described The step S2, the step S3 and the step S4 in the scaling method of TOF device, that is to say, that the calibration mold Block 32 can be based on at least three TOF images 110, and monocular demarcates the TOF camera module 11, to obtain the TOF camera module 11 TOF intrinsic parameter and the outer parameter of TOF；Based at least three RGB images 120, monocular demarcates the RGB camera module 12, to obtain Obtain the RGB intrinsic parameter and the outer parameter of RGB of the RGB camera module 12；And the binocular calibration TOF device 10, it is set with obtaining the TOF Standby 10 structural parameters.
It is worth noting that, as shown in figure 17, the control module 30 further comprises a positioning module 33, wherein described Positioning module 33 is set to change the position of the target module 20, to shoot the target module for being in different location 20, enable the visual field that the target module 20 is common in the TOF camera module 11 and the RGB camera module 12 It is moved in region, to change the image 200 of the target module 20 in the TOF image 110 and the RGB image 120 Position, and then the TOF image 110 and the RGB image 120 are acquired again by the interactive module 31, until described Until the home position of target module 20 is all collected, and ensure at least three width TOF images 110 and at least three width RGB figure As 120.Preferably, the positioning module 33 may be implemented as a remote controlled Universal working carriage, so that user can Long-range control remotely changes arbitrarily to change the orientation and angle of the Universal working carriage and is arranged at the universal work The orientation and angle of the target module 20 of platform.In other words, the calibration system can be improved the TOF device calibration Efficiency demarcates the time spent by the TOF device to shorten.It will be appreciated by those skilled in the art that the positioning module 33 Also it may be implemented as a moveable target bracket, to install the target module 20 to the target bracket, into And the position of the target module 20 can be changed by way of artificially changing the orientation and angle of the target bracket.
As shown in Fig. 9 and Figure 17, the control module 30 further comprises a supplementary lighting module 34, wherein the supplementary lighting module 34 are set with target module 20 described in light filling.Citing ground, the supplementary lighting module 34 are set to emit and image mould with the TOF The identical laser of laser that is emitted of the light source module 111 of group 11, to enhance the brightness of the target module 20, so as to When the TOF camera module 11 and the target module 20 are apart from each other, the TOF camera module 11 still is able to take clear The target module 20 image, to obtain the TOF image of high quality.It should be understood that when there are the supplementary lighting modules When 34 pairs of target modules 20 carry out light filling, the light source module 111 of the TOF camera module 11 can be closed, to prevent The light source module 111 occur in the target module 20 part it is reflective, so that the TOF camera module 11 be avoided to be clapped The TOF image on do not have highlighted place, and then obtain the TOF image 110 of high quality.Certainly, at this time can also using by The target module 20 made of light absorbent swashs to prevent the target module 20 because what the reflection light source module 11 was emitted Light and the quality for reducing the TOF image.
In addition, as shown in figure 17, in order to further increase the quality of the TOF image, the control module 30 is further Including a light-adjusting module 35, wherein the light-adjusting module 35 is set with the TOF according to the display module 32 The brightness of the image 200 of target module 20 described in image 110 automatically adjusts the target module of the TOF image 110 The brightness of 20 image 200 enables the interactive module 31 to acquire best matter to improve the quality of the TOF image 110 The TOF image 110 of amount, helps to improve the stated accuracy that the calibration system demarcates the TOF device 10.
Preferred embodiment according to the present invention, as shown in figure 9, the target module 20 of the calibration system is preferred Ground, which is carried out, is a flat surface gridiron pattern target 21, wherein the plane gridiron pattern target 21 has chequered with black and white grid, so that Four angle points of each grid are the characteristic point P of the target module 20.Preferably, the plane gridiron pattern target 21 by Such as plastics etc. translucent material or semi transparent material are made, and the supplementary lighting module 34 is implemented as a planar light source 341, wherein The planar light source 341 is arranged at a back side of the plane gridiron pattern target 21, and the planar light source 341 is stacked In the plane gridiron pattern target 21, the laser that the planar light source 341 is issued flows uniformly across the plane chess Disk case marker plate 21, and then received by the photosensitive control module 112 of the TOF camera module 11, so that the TOF takes the photograph As mould group 11 can take the TOF image of high quality.It is highly preferred that the laser that the planar light source 341 is emitted Wavelength is 850nm.
The step of citing ground, the calibration system demarcates TOF device 10, is as follows:
Firstly, opening the calibration system and the TOF device 10, one target module 20 of configuration images mould in the TOF In group and the common field of view of the RGB camera module, so that the TOF camera module 11 of the TOF device 10 and institute The target module 20 can be taken by stating RGB camera module 12, with the image 200 of target module 20 described in synchronization gain, To acquire corresponding TOF image and RGB image, wherein the TOF image and the RGB image include the target module 20 Image.
Then, the interactive module 31 of the control module 30 can receive the TOF image 110 and the RGB image 120, and the TOF image 110 and 120 simultaneous display of the RGB image can be come out.
Then, judge the target module 20 in the TOF image 110 and the RGB image 120 image whether It is complete, if it is, by TOF image 110 and the RGB image 120 described in 31 synchronous acquisition of interactive module, and will The TOF image 110 and the RGB image 120 save；If it is not, then first pass through the positioning module 33 change it is described The position of target module 20 to change position of the target module 20 relative to the TOF device 10, and then adjusts the mark Position of the image 200 of plate module 20 in the TOF image 110 and the RGB image 120, so that the target module 20 Image can be completely revealed in the TOF image and the RGB image, then same by the interactive module 31 Step acquires the TOF image 110 and the RGB120 image, so that it is guaranteed that in the TOF image 110 and the RGB image 120 It include the complete image of the target module 20.
Later, repeatedly change the position of the target module 20 by the positioning module 33, and change institute every time Behind the orientation for stating target module 20, repeats and acquire the TOF image 110 and described above by the interactive module 31 The step of RGB image 120, to obtain at least three width TOF images and corresponding at least three width RGB images.
Subsequently, based on at least three TOF images 110, by the demarcating module 32, monocular demarcates the TOF camera shooting Mould group 11, to obtain the TOF intrinsic parameter M of the TOF camera module 11TWith parameter (R outside TOFT, TT), for using；Based on institute At least three RGB images are stated, by the demarcating module 32, monocular demarcates the RGB camera module 12, takes the photograph to obtain the RGB As the RGB intrinsic parameter M of mould group 12RWith parameter (R outside RGBR, TR), for using.
Next, each RGB image 120 is carried out by the demarcating module 32 it is down-sampled, with obtain with it is described The identical down-sampled RGB image 120 ' of the resolution ratio of TOF image 110.
Then, it is based on the down-sampled RGB image 120 ', by the demarcating module 32, monocular demarcates the RGB again Camera module 12, parameter (R outside the down-sampled RGB to obtain the RGB camera module 12RJ, TRJ)。
Finally, based on the outer parameter (R of the TOFT, TT) and the outer parameter (R of the down-sampled RGBRJ, TRJ), by the mark Cover half block 32, TOF device described in binocular calibration, to obtain the structural parameters (R, T) of the TOF device, i.e., the described TOF device 10 Described in positional relationship between TOF camera module 11 and the RGB camera module 12, to complete to demarcate the TOF device 10 Task.
In conclusion not needing the structure using expensive material or complexity in the present invention.Therefore, the present invention success and One solution is effectively provided, not only provides the scaling method and its calibration system of a simple TOF device, while also increasing The scaling method of the TOF device and its practicability of calibration system and reliability.
It should be understood by those skilled in the art that foregoing description and the embodiment of the present invention shown in the drawings are only used as illustrating And it is not intended to limit the present invention.The purpose of the present invention has been fully and effectively achieved.Function and structural principle of the invention exists It shows and illustrates in embodiment, under without departing from the principle, embodiments of the present invention can have any deformation or modification.

Claims (27)

1. the scaling method of a TOF device, which comprises the following steps:
Acquire at least three TOF images and at least three RGB images；
Based on at least three TOF images, monocular is demarcated a TOF camera module of the TOF device, is taken the photograph with obtaining the TOF As mould group TOF intrinsic parameter and TOF outside parameter；
Based at least three RGB images, monocular demarcates a RGB camera module of the TOF device, is taken the photograph with obtaining the RGB As mould group RGB intrinsic parameter and RGB outside parameter；And
TOF device described in binocular calibration, to obtain the structural parameters of the TOF device.
2. the scaling method of TOF device as described in claim 1, wherein the acquisition at least three TOF images and at least three The step of RGB image, further comprises the steps of:
A target module is configured in the common field of view of the TOF camera module and the RGB camera module；
By the TOF camera module and the RGB camera module, target module described in sync pulse jamming, to acquire a TOF image With a RGB image；And
Arbitrarily change the position of the target module, and behind the position for changing the target module every time, repeats described Described in sync pulse jamming the step of target module, to acquire at least three TOF images and at least three RGB images.
3. the scaling method of TOF device as claimed in claim 2, wherein described by the TOF camera module and described RGB camera module, target module described in sync pulse jamming the step of to acquire the TOF image and the RGB image, further include Step:
By a light source module of the TOF camera module, emit a laser to the target module；With
By a photosensitive control module of the TOF camera module, receive by the reflected laser of the target module, with Acquire the TOF image.
4. the scaling method of TOF device as claimed in claim 2, wherein described by the TOF camera module and described RGB camera module, target module described in sync pulse jamming the step of to acquire the TOF image and the RGB image, further include Step:
Target module described in light filling.
5. the scaling method of TOF device as claimed in claim 2, wherein described by the TOF camera module and described RGB camera module, target module described in sync pulse jamming the step of to acquire the TOF image and the RGB image, further include Step:
The position of the target module is adjusted, so that the target module is completely located at the TOF camera module and the RGB In the common field of view of camera module.
6. the scaling method of TOF device as claimed in claim 2, wherein the position for arbitrarily changing the target module, And behind the position for changing the target module every time, the step of repeating target module described in the sync pulse jamming, to adopt It the step of collecting at least three TOF images and at least three RGB image, further comprises the steps of:
Change the position of the demarcating module, so that the target module is completely located at the TOF camera module and the RGB In angular zone in the common field of view of camera module.
7. such as the scaling method of TOF device as claimed in any one of claims 1 to 6, wherein described based on at least three TOF figure Picture, monocular demarcate the TOF camera module of the TOF device, to obtain the TOF intrinsic parameter and TOF of the TOF camera module The step of outer parameter, further comprises the steps of:
Establish a world coordinate system, a TOF camera module coordinate system, a TOF plane of delineation coordinate system and a TOF pixel coordinate System；
The TOF pixel coordinate of nonlinear at least four characteristic points on each TOF image is extracted, and it is special to obtain described at least four Levy the world coordinates of point；And
Based on the TOF pixel coordinate and the world coordinates, a homography matrix is solved, and according to the homography square Battle array finds out the initial TOF intrinsic parameter and the outer parameter of initial TOF of the TOF camera module.
8. the scaling method of TOF device as claimed in claim 7, wherein described based on at least three TOF images, monocular The TOF camera module of the TOF device is demarcated, parameter outside the TOF intrinsic parameter and TOF to obtain the TOF camera module The step of, it further comprises the steps of:
By the distortion correction model and a data fit object function of the TOF camera module, the TOF camera shooting mould is solved The TOF distortion parameter of group；With
According to maximum likelihood estimate, optimize the initial TOF intrinsic parameter and the outer parameter of the initial TOF, to obtain the TOF The TOF intrinsic parameter of camera module and the outer parameter of the TOF.
9. such as the scaling method of TOF device as claimed in any one of claims 1 to 6, wherein described based on at least three RGB figure Picture, monocular demarcate the RGB camera module of the TOF device, to obtain the RGB intrinsic parameter and RGB of the RGB camera module The step of outer parameter, further comprises the steps of:
Establish a world coordinate system, a RGB camera module coordinate system, a RGB image plane coordinate system and a rgb pixel coordinate System；
The rgb pixel coordinate of nonlinear at least four characteristic points on each RGB image is extracted, and it is special to obtain described at least four Levy the world coordinates of point；And
Based on the rgb pixel coordinate and the world coordinates, a homography matrix is solved, and according to the homography matrix, Find out the Initial R GB intrinsic parameter and the outer parameter of Initial R GB of the RGB camera module.
10. the scaling method of TOF device as claimed in claim 9, wherein described based at least three RGB images, monocular The RGB camera module of the TOF device is demarcated, parameter outside the RGB intrinsic parameter and RGB to obtain the RGB camera module The step of, it further comprises the steps of:
By the distortion correction model and a data fit object function of the RGB camera module, the RGB camera shooting mould is found out The RGB distortion parameter of group；With
According to maximum likelihood estimate, optimize the Initial R GB intrinsic parameter and the outer parameter of the Initial R GB, to obtain the RGB The RGB intrinsic parameter of camera module and the outer parameter of the RGB.
11. such as the scaling method of TOF device as claimed in any one of claims 1 to 6, wherein TOF described in the binocular calibration is set It is standby, the step of structural parameters to obtain the TOF device, further comprise the steps of:
Down-sampled at least three RGB images, so that the resolution ratio of at least three down-sampled RGB images and at least three TOF The resolution ratio of image is corresponding；
Based on at least three down-sampled RGB image, monocular demarcates the RGB camera module, to obtain the RGB camera module The outer parameter of down-sampled RGB；And
Based on parameter outside parameter outside the TOF and the down-sampled RGB, the structural parameters of the TOF device are solved.
12. the scaling method of TOF device as claimed in claim 11, wherein down-sampled at least three RGB images, So that the resolution ratio of at least three down-sampled RGB images step corresponding with the resolution ratio of at least three TOF images, is also wrapped Include step:
Equal proportion scales the RGB image；With
Cut the RGB image.
13. the scaling method of TOF device as claimed in claim 11, wherein in the down-sampled at least three RGB figure Picture, so that the resolution ratio of at least three down-sampled RGB images step corresponding with the resolution ratio of at least three TOF images In, the down-sampled RGB image includes the complete image of the demarcating module.
14. a calibration system, for demarcating a TOF device, wherein the TOF device includes that a TOF camera module and a RGB image mould Group characterized by comprising
One target module, wherein the target module is configured in the common visual field of the TOF camera module He the RGB camera module In region；With
One control module, wherein the control module includes:
One interactive module, wherein the interactive module is set to be communicatively coupled with the TOF device,
To acquire at least three RGB captured by least three TOF images captured by the TOF camera module and the RGB camera module Image；With
One demarcating module, wherein the demarcating module is communicatively coupled with the interactive module, and the demarcating module is pressed Demarcated according to a scaling method, wherein the scaling method comprising steps of
Based on at least three TOF images, monocular demarcates the TOF camera module, to obtain the TOF internal reference of the TOF camera module The outer parameter of several and TOF；
Based at least three RGB images, monocular demarcates the RGB camera module, to obtain the RGB internal reference of the RGB camera module The outer parameter of several and RGB；And
Binocular calibration TOF device, to obtain the structural parameters of the TOF device.
15. calibration system as claimed in claim 14, wherein the TOF camera module includes a light source module and a photosensitive control Molding block, wherein the light source module is set to emit a laser towards the target module, which is set To receive and process by the reflected laser of the target module, to obtain described at least three by the TOF camera module TOF image.
16. calibration system as claimed in claim 14 further includes a supplementary lighting module, wherein the supplementary lighting module is set to mend Target module described in light.
17. calibration system as claimed in claim 16, wherein the supplementary lighting module is a flat surface light source, and the light filling Module is stacked and placed on a back side of the target module.
18. calibration system as claimed in claim 14 further includes a positioning module, wherein the positioning module is set to adjust The position of the whole target module, so that the target module is completely located at the TOF camera module and RGB camera shooting mould In the common field of view of group.
19. calibration system as claimed in claim 18, wherein the positioning module is set purposefully to change the mark The position of cover half block, so as to be completely located at the TOF camera module and the RGB camera module common for the target module In angular zone in the field of view.
20. the calibration system as described in any in claim 14-19, wherein the demarcating module according to the mark Determine in method, described based on at least three TOF images, monocular demarcates the TOF camera module, to obtain the TOF camera module TOF intrinsic parameter and the step of TOF outer parameter, further comprise the steps of:
Establish a world coordinate system, a TOF camera module coordinate system, a TOF plane of delineation coordinate system and a TOF pixel coordinate System；
The TOF pixel coordinate of nonlinear at least four characteristic points on each TOF image is extracted, and it is special to obtain described at least four Levy the world coordinates of point；And
Based on the TOF pixel coordinate and the world coordinates, a homography matrix is solved, and according to the homography square Battle array finds out the initial TOF intrinsic parameter and the outer parameter of initial TOF of the TOF camera module.
21. calibration system as claimed in claim 20, wherein the demarcating module according to the scaling method in, At least three TOF images based on described in, monocular demarcates the TOF camera module, to obtain the TOF internal reference of the TOF camera module Outside several and TOF the step of parameter, further comprise the steps of:
By the distortion correction model and a data fit object function of the TOF camera module, the TOF camera module is found out TOF distortion parameter；With
According to maximum likelihood estimate, optimizes the initial TOF intrinsic parameter and the outer parameter of the initial TOF, taken the photograph with obtaining the TOF As parameter outside the TOF intrinsic parameter of mould group and the TOF.
22. the calibration system as described in any in claim 14-19, wherein the demarcating module according to the mark Determine in method, described based at least three RGB images, monocular demarcates the RGB camera module, to obtain the RGB camera module RGB intrinsic parameter and the step of RGB outer parameter, further comprise the steps of:
Establish a world coordinate system, a RGB camera module coordinate system, a RGB image plane coordinate system and a rgb pixel coordinate System；
Extract at least rgb pixel coordinate of four characteristic points, and the generation of the acquisition at least four characteristic points on each RGB image Boundary's coordinate；And
Based on the rgb pixel coordinate and the world coordinates, a homography matrix is solved, and according to the homography matrix, Find out the Initial R GB intrinsic parameter and the outer parameter of Initial R GB of the RGB camera module.
23. calibration system as claimed in claim 22, wherein the demarcating module according to the scaling method in, At least three RGB images based on described in, monocular demarcates the RGB camera module, to obtain the RGB internal reference of the RGB camera module Outside several and RGB the step of parameter, further comprise the steps of:
By the distortion correction model and a data fit object function of the RGB camera module, the RGB camera shooting mould is found out The RGB distortion parameter of group；With
According to maximum likelihood estimate, optimize the Initial R GB intrinsic parameter and the outer parameter of the Initial R GB, to obtain the RGB The RGB intrinsic parameter of camera module and the outer parameter of the RGB.
24. the calibration system as described in any in claim 14-19, wherein the demarcating module according to the mark Determine in method, the binocular calibration TOF device, the step of structural parameters to obtain the TOF device, further comprises the steps of:
Down-sampled at least three RGB images, so that the resolution ratio of at least three down-sampled RGB images and at least three TOF The resolution ratio of image is corresponding；
Based on at least three down-sampled RGB image, monocular demarcates the RGB camera module, to obtain the drop of the RGB camera module Sample the outer parameter of RGB；And
Based on parameter outside parameter outside the TOF and the down-sampled RGB, the structural parameters of the TOF device are solved.
25. calibration system as claimed in claim 24, wherein the demarcating module according to the scaling method in, Down-sampled at least three RGB images, so that the resolution ratio of at least three down-sampled RGB images and at least three TOF The corresponding step of the resolution ratio of image, further comprises the steps of:
Equal proportion scales the RGB image；With
Cut the RGB image.
26. calibration system as claimed in claim 24, wherein at least three down-sampled RGB image includes the calibration The complete image of module.
27. calibration system as claimed in claim 14 further includes a light-adjusting module, wherein the light-adjusting module is set with root Target module described in the TOF image is automatically adjusted according to the brightness of the image of target module described in the TOF image The brightness of described image.
CN201810202909.3A 2018-02-08 2018-03-13 Calibration method and calibration system of TOF equipment Active CN110136203B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN2018101269390 2018-02-08
CN201810126939 2018-02-08

Publications (2)

Publication Number Publication Date
CN110136203A true CN110136203A (en) 2019-08-16
CN110136203B CN110136203B (en) 2022-02-18

Family

ID=67568293

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810202909.3A Active CN110136203B (en) 2018-02-08 2018-03-13 Calibration method and calibration system of TOF equipment

Country Status (1)

Country Link
CN (1) CN110136203B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112415492A (en) * 2019-08-22 2021-02-26 Enscape有限公司 Inspection apparatus of camera module with TOF function
CN112581542A (en) * 2020-12-24 2021-03-30 北京百度网讯科技有限公司 Method, device and equipment for evaluating automatic driving monocular calibration algorithm
CN113689391A (en) * 2021-08-16 2021-11-23 炬佑智能科技(苏州)有限公司 ToF device installation parameter acquisition method and system and ToF device

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106772431A (en) * 2017-01-23 2017-05-31 杭州蓝芯科技有限公司 A kind of Depth Information Acquistion devices and methods therefor of combination TOF technologies and binocular vision
CN106815867A (en) * 2015-11-30 2017-06-09 宁波舜宇光电信息有限公司 TOF camera is demarcated and correction system and its apparatus and method for

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106815867A (en) * 2015-11-30 2017-06-09 宁波舜宇光电信息有限公司 TOF camera is demarcated and correction system and its apparatus and method for
CN106772431A (en) * 2017-01-23 2017-05-31 杭州蓝芯科技有限公司 A kind of Depth Information Acquistion devices and methods therefor of combination TOF technologies and binocular vision

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
JIA WU ET AL.: "Improved 3D Depth Image Estimation Algorithm for Visual Camera", 《2009 2ND INTERNATIONAL CONGRESS ON IMAGE AND SIGNAL PROCESSING》 *
周杰等: "飞行时间深度相机和彩色相机的联合标定", 《信号处理》 *
本书编委会: "《2017测绘地理信息仪器装备发展研究》", 31 October 2017, 测绘出版社 *
舒娜: "摄像机标定方法的研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112415492A (en) * 2019-08-22 2021-02-26 Enscape有限公司 Inspection apparatus of camera module with TOF function
CN112581542A (en) * 2020-12-24 2021-03-30 北京百度网讯科技有限公司 Method, device and equipment for evaluating automatic driving monocular calibration algorithm
CN113689391A (en) * 2021-08-16 2021-11-23 炬佑智能科技(苏州)有限公司 ToF device installation parameter acquisition method and system and ToF device

Also Published As

Publication number Publication date
CN110136203B (en) 2022-02-18

Similar Documents

Publication Publication Date Title
CN106412433B (en) Atomatic focusing method and system based on RGB-IR depth camera
CN106454090B (en) Atomatic focusing method and system based on depth camera
US9443308B2 (en) Position and orientation determination in 6-DOF
CN104335005B (en) 3D is scanned and alignment system
US11354810B2 (en) Methods and systems for imaging a scene, such as a medical scene, and tracking objects within the scene
US20140307100A1 (en) Orthographic image capture system
CN109831660A (en) Depth image acquisition method, depth image obtaining module and electronic equipment
CN110136203A (en) The scaling method and its calibration system of TOF device
CN108234984A (en) Binocular depth camera system and depth image generation method
CN108718373A (en) Device for image
CN107449374A (en) The vision auxiliary laser galvanometer scanning system and its field calibration method of flexible layout
CN116017164A (en) System and method for capturing and generating panoramic three-dimensional images
CN109819173A (en) Depth integration method and TOF camera based on TOF imaging system
KR20170133119A (en) Depth extracting camera system using multi focus image and operation method thereof
CN113034567A (en) Depth truth value acquisition method, device and system and depth camera
CN110880161B (en) Depth image stitching and fusion method and system for multiple hosts and multiple depth cameras
CN114359406A (en) Calibration of auto-focusing binocular camera, 3D vision and depth point cloud calculation method
CN210719062U (en) Tracking device and scanning system based on wireless synchronization
CN107564051A (en) A kind of depth information acquisition method and system
CN111654626B (en) High-resolution camera containing depth information
CN113711229A (en) Control method of electronic device, and computer-readable storage medium
CN107478172B (en) Laser three-D curved profile positioning projection method based on binocular vision
JP7300895B2 (en) Image processing device, image processing method, program, and storage medium
CN110378964A (en) Join scaling method and device, storage medium outside a kind of video camera
CN107301672A (en) A kind of indoor scene becomes more meticulous model building device and modeling method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant