A kind of unmanned plane boat is taken the photograph the three-dimensional imaging treatment facilityTechnical field
The utility model relates to imaging device, specifically relates to a kind of unmanned plane boat and takes the photograph the three-dimensional imaging treatment facility.
Background technology
Existing unmanned plane then carries out interpretation by the two-dimensional signal in photoelectricity and digital platform acquisition space, and this equipment adopts technique of binocular stereoscopic vision, realizes the 3-d reproduction of object scene, and finishes the space orientation to scenery.The image that photoelectricity and digital platform obtain can extract the two-dimensional signal of target, but can't obtain the depth information of target.The space multistory information of obtaining the target in the image is the difficult point problem that unmanned plane boat photogra is used.
From current unmanned plane boat photogra treatment technology and treatment facility present situation thereof, mainly there is the shortcoming of the following aspects:
(1) lacks systematic three-dimensional localization process technology.
Most unmanned planes do not adopt three-dimensional localization process technology, only use the photo that obtains to carry out simple image interpretation and analysis use.
(2) lack the equipment of realizing that three-dimensional imaging is handled.
Existing unmanned plane mainly adopts single-sensor to be navigated in ground and takes the photograph, and each takes the photograph the station only can obtain a width of cloth image, does not make full use of each boat of taking the photograph the station and takes the photograph information, the equipment that does not also have three-dimensional imaging to handle simultaneously.
Summary of the invention
Technical problem to be solved in the utility model is to provide a kind of unmanned plane boat to take the photograph the three-dimensional imaging treatment facility, problem at unmanned plane aerophoto prior art and disposal route existence, base oneself upon twin camera unmanned plane boat and take the photograph processing, primary study on unmanned plane, set up the problem of binocular tri-dimensional vision system, realized solid location to ground scenery.
The utility model solves the problems of the technologies described above by the following technical programs: a kind of unmanned plane boat is taken the photograph the three-dimensional imaging treatment facility, comprise image capture module, camera calibration module, the image pretreatment module, video camera interior orientation demarcating module, video camera outer orientation demarcating module, three-dimensional matching module and three-dimensional reconstruction module, described image capture module connects camera calibration module and image pretreatment module simultaneously, image pretreatment module output terminal is connected to the input end of three-dimensional matching module, the output terminal of three-dimensional matching module, the output terminal of video camera interior orientation demarcating module, and the output terminal of video camera outer orientation demarcating module is connected to three rebuilding module simultaneously.
This utility model further is specially:
Described image capture module comprises preceding ccd video camera, back ccd video camera, image pick-up card, preceding ccd video camera, back ccd video camera input end are connected to video camera interior orientation demarcating module, output terminal is connected to image pick-up card, and the output terminal of image pick-up card is connected to the image pretreatment module.
Described image pick-up card comprises low-pass filter, sampling hold circuit, A/D converter, image storage unit, CPU and the display that connects in regular turn.。
Described image pretreatment module comprises image enhancing unit, image restoration unit, noise smoothing filter unit and geometry correction unit, the output terminal of image pick-up card is connected to the input end of image enhancing unit, image restoration unit, noise smoothing filter unit and geometry correction unit simultaneously, and the output terminal of image enhancing unit, image restoration unit, noise smoothing filter unit and geometry correction unit is connected to three-dimensional matching module.
Described three-dimensional matching module comprises feature extraction unit and the characteristic matching unit that connects in regular turn, described video camera interior orientation demarcating module comprises the controlling filed unit, feature extraction unit, and the video camera elements of interior orientation resolves the unit, ccd video camera before described, back ccd video camera input end is connected to the controlling filed unit of video camera interior orientation demarcating module, the image enhancing unit of described image pretreatment module, image restoration unit, the output terminal of noise smoothing filter unit and geometry correction unit is connected to the feature extraction unit of three-dimensional matching module and the feature extraction unit of video camera interior orientation demarcating module simultaneously, described video camera interior orientation demarcating module comprises GPS and inertial navigation system, the video camera elements of exterior orientation resolves the unit, the output terminal of the characteristic matching unit of described three-dimensional matching module, the video camera elements of interior orientation of video camera interior orientation demarcating module resolves the output terminal of unit, and the video camera elements of exterior orientation of the video camera outer orientation demarcating module output terminal that resolves the unit is connected to the spatial coordinates calculation unit of three-dimensional reconstruction module simultaneously, and described three-dimensional reconstruction unit also comprises the depth map generation unit that is connected to spatial coordinates calculation unit output terminal.
The utility model has the advantages that: solved following technical matters:
1. the establishment of binocular tri-dimensional vision model.
According to the ultimate principle of binocular stereo vision, work out a cover and be fit to the model that the unmanned plane aerophoto obtains stereopsis, realize the extraction of depth information.
2. Camera calibration technology.
Need s internal and external orientation at three-dimensional reconstruction, the method that proposes to use controlling filed to demarcate is carried out the demarcation of the elements of interior orientation of video camera, has improved stated accuracy.
3. the three-dimensional reconstruction technology of scenery.
Employing realizes three-dimensional reconstruction based on the solid matching method of Edge Gradient Feature.
Compared with prior art have following beneficial effect:
1. simple to operate, system only needs manual intervention seldom can realize the Three-dimension Target location is overcome the shortcoming that traditional treatment method needs a large amount of artificial complex operations.
2. the depth information precision is effective, based on the stereotactic system of binocular stereo vision can be fast, accurately, flexibly target is carried out space orientation, and traditional way needs a large amount of ground control points to calculate.
3. binocular stereo vision localization process speed is fast, with traditional manually the object space location being compared, this system can realize real-time positioning, beam back ground acceptance by the photo that binocular vision system obtains by radio, directly use the spatial information that target is resolved by this system, speed is fast, and real-time is good, can satisfy the demand of obtaining information fast.
[description of drawings]
Fig. 1 takes the photograph the structured flowchart of three-dimensional imaging treatment facility for a kind of unmanned plane boat of the utility model.
Fig. 2 is the theory diagram of image pick-up card.
[embodiment]
The utility model will be further described in conjunction with the embodiments with reference to the accompanying drawings, so that those skilled in the art can better understand the utility model and being implemented, but illustrated embodiment is not as to qualification of the present utility model.
As shown in Figure 1, a kind of unmanned plane of the utility model boat is taken the photograph the three-dimensional imaging treatment facility and is comprised image capture module, camera calibration module, image pretreatment module, video camera interior orientation demarcating module, video camera outer orientation demarcating module, three-dimensional matching module and three-dimensional reconstruction module.
Described image capture module connects camera calibration module and image pretreatment module simultaneously, image pretreatment module output terminal is connected to the input end of three-dimensional matching module, the output terminal of three-dimensional matching module, the output terminal of video camera interior orientation demarcating module, and the output terminal of video camera outer orientation demarcating module is connected to three rebuilding module simultaneously.
Described image capture module comprises preceding CCD(charge-coupled image sensor) video camera, back ccd video camera, image pick-up card, it is right to the airphoto of same target that this image capture module module can generate two ccd video cameras.Preceding ccd video camera, back ccd video camera input end are connected to video camera interior orientation demarcating module, and preceding ccd video camera, back ccd video camera output terminal are connected to image pick-up card, and the output terminal of image pick-up card is connected to the image pretreatment module.
Image pick-up card is mainly finished the digitized process to analog video signal.Please consult shown in Figure 2ly simultaneously, described image pick-up card comprises in regular turn low-pass filter, sampling hold circuit, A/D converter, image storage unit, the CPU that connects, and display.
Preceding ccd video camera, back ccd video camera is gathered in real time to the vision signal of the scenery of input, vision signal is at first through low-pass filter filtering, be converted to continuous analog signal in time, according to the requirement of application system to image resolution ratio, with sampling hold circuit continuous vision signal is carried out interval sampling in time, vision signal is converted to discrete simulating signal, after the A/D conversion, be stored in one or more channels of image storage unit, send instruction by computing machine, one two field picture is still in the image memory channel, promptly gather or caught a two field picture, computing machine is handled the image of gathering, and image is shown on the display in real time.In addition, image pick-up card also has video image analysis, processing capacity, and can effectively control video camera.
Described image pretreatment module comprises image enhancing unit, image restoration unit, noise smoothing filter unit and geometry correction unit.The output terminal of image pick-up card is connected to the input end of image enhancing unit, image restoration unit, noise smoothing filter unit and geometry correction unit simultaneously, and the output terminal of image enhancing unit, image restoration unit, noise smoothing filter unit and geometry correction unit is connected to three-dimensional matching module.
The pretreated purpose of image has two: the one, improve the visual effect of image, and improve the sharpness of image; The 2nd, image is become more help the processing of computing machine, be convenient to various signature analysises.Image enhancing unit mainly is to realize the adjustment of the gray scale (or brightness) of image, the detailed information of outstanding image; Image restoration unit mainly solves the IMC problem of image, improves the sharpness of image; The noise smoothing filter unit mainly solves the squelch problem of Image Acquisition and transmission course; Geometry correction unit realizes the function of the geometric distortion correction of image photo coordinate conversion and image inside according to elements of interior orientation.
Described three-dimensional matching module comprises feature extraction unit and the characteristic matching unit that connects in regular turn.Feature extraction is exactly to extract the feature primitive of being convenient to mate from two width of cloth images, as point-like character, line feature or provincial characteristics etc.Characteristic matching is according to the calculating to selected feature primitive, sets up the one-to-one relationship between two width of cloth characteristics of image primitives, and obtains corresponding anaglyph thus.
Described video camera interior orientation demarcating module comprises controlling filed unit, feature extraction unit, and the video camera elements of interior orientation resolves the unit.Ccd video camera, back ccd video camera input end are connected to the controlling filed unit of video camera interior orientation demarcating module before described.The output terminal of the image enhancing unit of described image pretreatment module, image restoration unit, noise smoothing filter unit and geometry correction unit is connected to the feature extraction unit of three-dimensional matching module and the feature extraction unit of video camera interior orientation demarcating module simultaneously.
Described video camera interior orientation demarcating module comprises that GPS and inertial navigation system, video camera elements of exterior orientation resolve the unit.The video camera elements of interior orientation of the output terminal of the characteristic matching unit of described three-dimensional matching module, video camera interior orientation demarcating module resolves the output terminal of unit, and, the video camera elements of exterior orientation of the video camera outer orientation demarcating module output terminal that resolves the unit is connected to the spatial coordinates calculation unit of three-dimensional reconstruction module simultaneously.Described three-dimensional reconstruction unit also comprises the depth map generation unit that is connected to spatial coordinates calculation unit output terminal.
Camera calibration is for position, the property parameters of determining video camera and sets up imaging model, so as to determine in the space coordinates object point with it at the corresponding relation between the picture point on the plane of delineation.The geological information of object from the image information calculating three dimensions that video camera obtains, and rebuild thus and recognition object, and the three-dimensional geometry position of certain point of space object surface, mutual relationship in itself and the image between the corresponding point is that these geometric model parameters are exactly camera parameters by the geometric model decision of video camera imaging.Camera calibration need be determined the three-dimensional position and the direction (elements of exterior orientation) of video camera inner geometry and the optical characteristics (elements of interior orientation) and the camera coordinate system of a relative world coordinate system.
The brief operation process that a kind of unmanned plane boat of the utility model is taken the photograph the three-dimensional imaging treatment facility is:
1. camera calibration.
At first, utilize two video cameras to take the control field pattern, obtain image after the shooting by image pick-up card; Secondly, by the image pretreatment module, image is carried out processing such as figure image intensifying, image recovery, noise smoothing filtering and geometry correction, extract minutiae, the elements of interior orientation of calculating video camera.Two video cameras are during to the ground scene shot, and position and attitude information that record unmanned plane GPS and inertial navigation system provide resolve the elements of exterior orientation of two video cameras.
2. stereoscopic vision is rebuild
The first step, two video cameras are taken ground scenery, and the image that is obtained scenery by image pick-up card is right; Second step, to Flame Image Process, utilize the method for characteristic matching to extract the right unique point of image through the image pretreatment module, carry out the solid coupling; In the 3rd step,, resolve the volume coordinate of scenery by the s internal and external orientation of two video cameras.