CN109443319A - Barrier range-measurement system and its distance measuring method based on monocular vision - Google Patents
Barrier range-measurement system and its distance measuring method based on monocular vision Download PDFInfo
- Publication number
- CN109443319A CN109443319A CN201811570292.7A CN201811570292A CN109443319A CN 109443319 A CN109443319 A CN 109443319A CN 201811570292 A CN201811570292 A CN 201811570292A CN 109443319 A CN109443319 A CN 109443319A
- Authority
- CN
- China
- Prior art keywords
- image
- barrier
- monocular vision
- distance
- module
- 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.)
- Pending
Links
- 230000004888 barrier function Effects 0.000 title claims abstract description 62
- 230000004438 eyesight Effects 0.000 title claims abstract description 56
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000005259 measurement Methods 0.000 title claims abstract description 10
- 238000012545 processing Methods 0.000 claims abstract description 27
- 238000004364 calculation method Methods 0.000 claims abstract description 23
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 33
- 238000013507 mapping Methods 0.000 claims description 10
- 238000004040 coloring Methods 0.000 claims description 7
- 238000011084 recovery Methods 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 description 5
- 238000009434 installation Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000003709 image segmentation Methods 0.000 description 3
- 238000003706 image smoothing Methods 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000010191 image analysis Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000000691 measurement method Methods 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 239000006002 Pepper Substances 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000010668 complexation reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004392 development of vision Effects 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000004310 photopic vision Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/02—Picture taking arrangements specially adapted for photogrammetry or photographic surveying, e.g. controlling overlapping of pictures
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/04—Interpretation of pictures
- G01C11/30—Interpretation of pictures by triangulation
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
- G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Electromagnetism (AREA)
- Aviation & Aerospace Engineering (AREA)
- Automation & Control Theory (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The barrier range-measurement system based on monocular vision that the invention discloses a kind of, comprising: image pickup module shoots barrier single image and sends as image processing module；Image processing module converts image to be analyzed for single image after pre-processing to single image and is sent to distance calculation module；Distance calculation module is analysed to image and carries out digitized description completion obstacle height Information recovering, after determining the location information of the height, size and bottom of barrier in phase plane, according to image pickup module focal length, image pickup module position in phase plane of height and barrier bottom from the ground, pass through the similar triangulation calculation acquired disturbance object distance of optics.The invention also discloses a kind of barrier distance measuring method based on monocular vision.
Description
Technical field
The present invention relates to automotive fields, more particularly to a kind of barrier for automatic Pilot technology based on monocular vision
Range-measurement system.The invention further relates to a kind of barrier distance measuring methods for automatic Pilot technology based on monocular vision.
Background technique
Visual token is exactly the rudimentary knowledge that the mankind obtain range data using computer, and since it is in machine
Extremely important position is occupied and the extensive concern by everybody in device people research, and in machine target following, vision positioning
And it is widely used in vision avoidance etc., while as indispensable heavy in vision guided navigation and SERVO CONTROL
Want technology.With the development of vision technique, more and more location algorithms are emerged, and relatively common is exactly Structure Method, list
Range estimation amount, binocular measurement and more mesh mensurations etc..Wherein structure light is by the influence of light source because limited, it can only be
Fixed occasion is applied；Binocular vision by Feature Points Matching due to being influenced, so precision and efficiency are all not steady enough
It is fixed, therefore how to solve the problems, such as that the Feature Points Matching of binocular vision is the emphasis direction of part research；Compare the two of front
Kind measurement method, monocular vision distance measuring structure is relatively simple, and arithmetic speed is quick, so that it be made above to occupy weight in application in the future
The status wanted.
It is logical for using binocular or more purpose detection methods, the mode of binocular detecting distance in existing visual token instrument
The calculating to two images parallax is crossed, range measurement directly is carried out to front scenery, without judging what class front occurs that
The barrier of type.The principle of binocular camera is similar to human eye.Human eye can perceive the distance of object, be due to two eyes pair
The image that the same object is presented has differences, also referred to as " parallax ".Object distance is remoter, and parallax is smaller；Conversely, parallax is bigger.
The size of parallax corresponds to the distance of distance between object and eyes.It is poor with photopic vision for a simple experiment, in tester
Eyes before lift his finger vertically, then alternately close his eyes, tester it will be noted that he finger relative to
The background of scene is beated to the left and to the right, and the movement or parallax in level are inversely proportional at a distance from tester.
Compare binocular or more mesh measurement methods, and monocular vision ranging is simple with its structure, and arithmetic speed is fast, at low cost etc.
Advantage has obtained widest use.Single camera vision system method needs the accurate positioning and complicated calibration of camera at present, and
Matching due to finding the i.e. image of picture of same object point between image needs a large amount of extremely complex calculating, limits it and answers
Use range.In addition current single camera vision system method has used a half-reflecting mirror that the light of same viewpoint is divided into two parts,
Two camera lenses of different apertures are reused two images are obtained on light-ray condensing to two target surfaces, using same point in two width figures
Defocusing degree difference as in calculates image distance.This method camera structure is complex, and installation requirement is higher, calculation amount also compared with
Greatly.Intelligent driving needs to measure the distance of vehicle front barrier, binocular used at present or brilliant distance measuring method, so that being
Cost of uniting is relatively high, and is also required to bigger installation space, this just needs a detection accuracy to meet automotive safety spacing
Early warning requirement, the detection method that cost is relatively low, installation space is small.
Summary of the invention
The barrier range-measurement system based on monocular vision that the technical problem to be solved in the present invention is to provide a kind of.The present invention is also
Provide a kind of barrier distance measuring method based on monocular vision.
Image smoothing refer to enlarged regions for protruding image, low-frequency component, trunk portion or inhibit picture noise and
Interfere the image processing method of radio-frequency component, it is therefore an objective to make the gentle gradual change of brightness of image, reduce mutation gradient, improve image matter
Amount.The method of image smoothing includes: interpolation method, linear smoothing method, convolution method etc..Such processing method is according to image
The different of noise carry out smoothly, such as salt-pepper noise, just use linear smoothing method！
Mathematical morphology (mathematical morphology, MM): being tight to have according to morphology concept development
The science on lattice mathematical theory basis, and be applied successfully in image procossing and area of pattern recognition.It is with certain shape
The structural element of state goes to measure and extract the correspondingly-shaped in image to achieve the purpose that image analysis and identification.Morphology figure
As processing shows as a kind of neighborhood operation form；
Image segmentation is exactly to divide the image into several regions specific, with unique properties and propose interesting target
Technology and process.It is by the committed step of image procossing to image analysis.Existing image partition method mainly divides following
Several classes: the dividing method based on threshold value, the dividing method based on region, the dividing method based on edge and be based on specific theory
Dividing method etc..From the point of view of mathematical angle, image segmentation is that digital picture is divided into the process in mutually disjoint region.Figure
As segmentation process be also a labeling process, i.e., belong to the same area as rope imparting be identically numbered.
Shape from shading (shape from shading, abbreviation SFS) is that 3D shape restores (3D in computer vision
Shape recovery) problem one of key technology, task be using in single width image body surface light and shade variation come
Restore the parameter values such as relative altitude or the surface method direction of its surface each point, establishes base further to carry out three-dimensionalreconstruction to object
Plinth.
In order to solve the above technical problems, the barrier range-measurement system provided by the invention based on monocular vision, comprising: image
Acquisition module, image processing module and distance calculation module；
Described image acquisition module shoots barrier single image and sends as image processing module；
Described image processing module converts the single image to after pre-processing to the single image to be analyzed
Image is sent to distance calculation module；
The distance calculation module is analysed to image and carries out digitized description completion obstacle height Information recovering, determines
After location information of height, size and the bottom of barrier in phase plane, according to image pickup module focal length, image capture mould
Block position in phase plane of height and barrier bottom from the ground, passes through the similar triangulation calculation acquired disturbance object distance of optics
From.
It is further improved the obstacle distance computing system based on monocular vision, described image acquisition module is monocular
Camera or video camera, described image processing module and distance calculation module are MCU or PC.
It is further improved the obstacle distance computing system based on monocular vision, the pretreatment includes at least image
Smoothing processing, image dividing processing and/or morphology processing.
It is further improved the obstacle distance computing system based on monocular vision, described image processing module can be differentiated
Gray level image and color image.
It is further improved the obstacle distance computing system based on monocular vision, the single image is converted into wait divide
Analysing image is that the color image parts of single image are converted to gray level image.
It is further improved the obstacle distance computing system based on monocular vision, the distance calculation module will be wait divide
It is to be analysed to image to intend mapping from two-dimensional space to three-dimensional space that analysis image, which carries out digitized description,.
Be further improved the obstacle distance computing system based on monocular vision, image to be analyzed from two-dimensional space to
The quasi- mapping of three-dimensional space is in the following ways；
It is analysed to image pixel to describe using two-dimensional matrix, the two-dimensional matrix ranks indicate the space bit of pixel
It sets, the colouring information of pixel is indicated with numerical value or vector.
It is further improved the obstacle distance computing system based on monocular vision, image to be analyzed is converted into grayscale image
As its two-dimensional matrix is as follows；
F (i, j) is the gray value of the i-th row in matrix, jth column in F (x, y) two-dimensional matrix, wherein 1≤i≤m, 1≤j≤
n。
It is further improved the obstacle distance computing system based on monocular vision, the obstacle height Information recovering
Using shape from shading (shape from shading, SFS).
It is further improved the obstacle distance computing system based on monocular vision, the shape from shading is basis
The gray scale light and shade variation of image to be analyzed digital picture restores the relative altitude or surface normal of each point on the three-dimensional surface of object
Measure parameter value.That is, the functional relation between object surface shape and brightness of image is established by determining illumination reflection model,
And according to the prior model building to surface shape to the constraint condition of form parameter, then to the simultaneous under these constraint conditions
Equation solution, to obtain the 3D shape of body surface.This process is equivalent to completion one from two-dimensional space to three-dimensional space
Quasi- mapping, be imaging inverse process, it is therefore an objective to complete restoring again for object height information.
The intensity of Ideal Diffuse Reflection light is directly proportional with the folder cosine of an angle between incident light and body surface normal vector, it may be assumed that
E (x, y)=I (x, y) ρ cos θ
E (x, y) is the intensity that diffuses, and I (x, y) is the intensity of light source, and ρ is surface reflection coefficient, and θ is incident light and surface
Angle between normal vector.Assuming thatFor light source incidence vector,For body surface each point
Law vector, the functional form of the law vector of body surface is (zx, zy, -1), (p, q) is the surface graded of object.Then have:
When to have cos θ value be 1, E (x, y) has a maximum value, combines above formula, find out first brightness and the light source direction of image with
Determine surface graded (p, q) of object, then according to the relationship of (p, q) and z can further find out object apparent height z (x,
y)。
The present invention provides a kind of barrier distance measuring method based on monocular vision, comprising the following steps:
1) barrier single image is shot；
2) single image is pre-processed；
3) image to be analyzed is converted by the single image；
4) it is analysed to image and carries out digitized description completion obstacle height Information recovering；
5) location information of the height, size and bottom of barrier in phase plane is determined；
6) position according to shooting focal length, shooting distance ground level and barrier bottom in phase plane, passes through light
It learns similar triangulation calculation and obtains distance between shooting point and barrier.
It is further improved the barrier distance measuring method based on monocular vision, implementation steps 2) when, the pretreatment is extremely
It less include picture smooth treatment, image dividing processing and/or morphology processing.
It is further improved the barrier distance measuring method based on monocular vision, implementation steps 3) when, the single image
Being converted into image to be analyzed is that the color image parts of single image are converted to gray level image.
It is further improved the barrier distance measuring method based on monocular vision, implementation steps 4) when, it is analysed to image
Carrying out digitized description is to be analysed to image to intend mapping from two-dimensional space to three-dimensional space.
It is further improved the barrier distance measuring method based on monocular vision, image to be analyzed is from two-dimensional space to three-dimensional
The quasi- mapping in space is in the following ways；
It is analysed to image pixel to describe using two-dimensional matrix, the two-dimensional matrix ranks indicate the space bit of pixel
It sets, the colouring information of pixel is indicated with numerical value or vector.
Be further improved the barrier distance measuring method based on monocular vision, image to be analyzed be converted into gray level image its
Two-dimensional matrix is as follows；
F (i, j) is the gray value of the i-th row in matrix, jth column in F (x, y) two-dimensional matrix, wherein 1≤i≤m, 1≤j≤
n。
It is further improved the barrier distance measuring method based on monocular vision, the obstacle height Information recovering uses
Shape from shading (shape from shading, SFS).
It is further improved the barrier distance measuring method based on monocular vision, the shape from shading is according to wait divide
The gray scale light and shade variation for analysing image digital image restores the relative altitude of each point or surface normal ginseng on the three-dimensional surface of object
Numerical value.That is, the functional relation between object surface shape and brightness of image is established by determining illumination reflection model, and according to
According to the prior model building to surface shape to the constraint condition of form parameter, then to the simultaneous equations under these constraint conditions
It solves, to obtain the 3D shape of body surface.It is quasi- from two-dimensional space to three-dimensional space that this process is equivalent to completion one
Mapping is the inverse process of imaging, it is therefore an objective to complete restoring again for object height information.
The intensity of Ideal Diffuse Reflection light is directly proportional with the folder cosine of an angle between incident light and body surface normal vector, it may be assumed that
E (x, y)=I (x, y) ρ cos θ
E (x, y) is the intensity that diffuses, and I (x, y) is the intensity of light source, and ρ is surface reflection coefficient, and θ is incident light and surface
Angle between normal vector.Assuming thatFor light source incidence vector,For body surface each point
Law vector, the functional form of the law vector of body surface is (zx, zy, -1), (p, q) is the surface graded of object.Then have:
When to have cos θ value be 1, E (x, y) has a maximum value, combines above formula, find out first brightness and the light source direction of image with
Determine surface graded (p, q) of object, then according to the relationship of (p, q) and z can further find out object apparent height z (x,
y)。
It is ranging working principle of the present invention is as follows restoring the 3D structural information of barrier (target) from single image
When in the reflection model of radiation direction, body surface be known conditions, the light and shade variation institute using gray scale in single image is hidden
The shape information contained, the surface three dimension shape of reconstruction of objects.The reconstruct of target object 3D shape is built upon the basis of image
On, first image is pre-processed when carrying out the reconstruct of 3D shape, such as picture smooth treatment, image dividing processing and mathematics
Morphological scale-space.The figure to be analyzed of gray scale is converted by digitized description, a width digital picture is by limited pixel group
At there are two types of feature, that is, space characteristics and color characteristics for each pixel tool.Usual pixel is retouched in the form of two-dimensional matrix
It states.The spatial position that pixel is indicated with the ranks of matrix element indicates the colouring information of pixel with numerical value or vector.For
For gray level image, it can be indicated using two-dimensional matrix.Next do shape from shading (shape from shading,
It SFS) is the relative altitude z (x, y) for changing each point on the three-dimensional surface for restoring object according to the gray scale light and shade of single digital image
Or surface normal (zx, zy, -1) etc. parameter values.Purpose is restoring again for object height information to be completed.3D structure after recovery
It can determine the location information of the height, size and bottom of barrier in phase plane.Finally according to known focal length of camera,
Barrier (target) is extrapolated according to the similar triangle of optics in the position of height and barrier bottom in phase plane from the ground
Distance.
The present invention drives suitable for automobile intelligent, estimates road obstacle distance using single image, installs on vehicle
Then monocular camera system is believed with vehicle captured in real-time front obstacle figure by restoring the 3D structure of target from single image
Breath, the surface three dimension shape of reconstruction of objects can determine the location information of the height, size and bottom of barrier in phase plane,
The finally position according to known focal length of camera, height from the ground and barrier bottom in phase plane, according to optics
Similar triangle extrapolates the distance of target.Cost of the present invention is relatively low, installation space is small, can be widely used in after forming product
Intelligent network joins automobile.
Detailed description of the invention
Present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments:
Fig. 1 is shape from shading (shape from shading, SFS) algorithm flow schematic diagram.
Fig. 2 is one implementation diagram of distance measuring method of the present invention.
Specific embodiment
Barrier range-measurement system one embodiment provided by the invention based on monocular vision, comprising: image pickup module, figure
As processing module and distance calculation module；Image pickup module uses monocular camera, image processing module and distance calculation module
It is integrated in car-mounted computer.
Described image acquisition module shoots barrier single image and sends as image processing module；
Described image processing module can differentiate gray level image and color image, after capable of pre-processing to the single image
(the i.e. single width figure that shows for gray scale chart of single image conversion after the color image parts of single image are converted to gray level image
Picture) it is sent to distance calculation module；
The distance calculation module is analysed to image and carries out digitized description completion obstacle height Information recovering, determines
After location information of height, size and the bottom of barrier in phase plane, according to image pickup module focal length, image capture mould
Block position in phase plane of height and barrier bottom from the ground, passes through the similar triangulation calculation acquired disturbance object distance of optics
From.
It is analysed to image in the following ways and intends mapping from two-dimensional space to three-dimensional space.
It is analysed to image pixel to describe using two-dimensional matrix, the two-dimensional matrix ranks indicate the space bit of pixel
It sets, indicates the colouring information of pixel with numerical value or vector, it is as follows that image to be analyzed is converted into gray level image its two-dimensional matrix；
F (i, j) is the gray value of the i-th row in matrix, jth column in F (x, y) two-dimensional matrix, wherein 1≤i≤m, 1≤j≤
n。
The obstacle height Information recovering uses shape from shading, and the shape from shading is according to figure to be analyzed
As the gray scale light and shade variation of digital picture restores the relative altitude or surface normal parameter value of each point on the three-dimensional surface of object.
The present invention provides a kind of one embodiment of barrier distance measuring method based on monocular vision, comprising the following steps:
1) barrier single image is shot；
2) single image is pre-processed, described pre-process includes at least picture smooth treatment, at image segmentation
Reason and/or morphology processing；
3) image to be analyzed is converted by the single image, it is by single width that the single image, which is converted into image to be analyzed,
The color image parts of image are converted to gray level image；
4) it is analysed to image and carries out digitized description completion obstacle height Information recovering, the obstacle height information
Restore to use shape from shading, the shape from shading is extensive according to the variation of the gray scale light and shade of image to be analyzed digital picture
The relative altitude or surface normal parameter value of each point on the three-dimensional surface of multiple object；
It is analysed to image and carries out digitized description be to be analysed to image in the following ways from two-dimensional space to three-dimensional
The quasi- mapping in space.
It is analysed to image pixel to describe using two-dimensional matrix, the two-dimensional matrix ranks indicate the space bit of pixel
It sets, indicates the colouring information of pixel with numerical value or vector, it is as follows that image to be analyzed is converted into gray level image its two-dimensional matrix；
F (i, j) is the gray value of the i-th row in matrix, jth column in F (x, y) two-dimensional matrix, wherein 1≤i≤m, 1≤j≤
n。
5) location information of the height, size and bottom of barrier in phase plane is determined；
6) position according to shooting focal length, shooting distance ground level and barrier bottom in phase plane, passes through light
It learns similar triangulation calculation and obtains distance between shooting point and barrier.
Known camera height H, camera focal length f, coordinate points y1 of the barrier point in image coordinate, based on practical
Pixel Dimensions yPixel.It can then solve:
The distance between all shooting point P and barrier B areSimilarly
Above by specific embodiment and embodiment, invention is explained in detail, but these are not composition pair
Limitation of the invention.Without departing from the principles of the present invention, those skilled in the art can also make many deformations and change
Into these also should be regarded as protection scope of the present invention.
Claims (18)
1. a kind of barrier range-measurement system based on monocular vision characterized by comprising image pickup module, image procossing
Module and distance calculation module；
Described image acquisition module shoots barrier single image and sends as image processing module；
Described image processing module by the single image converts image to be analyzed after pre-processing to the single image
It is sent to distance calculation module；
The distance calculation module is analysed to image and carries out digitized description completion obstacle height Information recovering, determines obstacle
After location information of height, size and the bottom of object in phase plane, according to image pickup module focal length, image pickup module from
The position of ground level and barrier bottom in phase plane passes through the similar triangulation calculation acquired disturbance object distance of optics.
2. the obstacle distance computing system based on monocular vision as described in claim 1, it is characterised in that: described image intake
Module is monocular camera or video camera, and described image processing module and distance calculation module are MCU or PC.
3. the obstacle distance computing system based on monocular vision as described in claim 1, it is characterised in that: the pretreatment is extremely
It less include picture smooth treatment, image dividing processing and/or morphology processing.
4. the obstacle distance computing system based on monocular vision as claimed in claim 3, it is characterised in that: described image processing
Module can differentiate gray level image and color image.
5. the obstacle distance computing system based on monocular vision as claimed in claim 4, it is characterised in that: the single image
Being converted into image to be analyzed is that the color image parts of single image are converted to gray level image.
6. the obstacle distance computing system based on monocular vision as claimed in claim 5, it is characterised in that: the distance calculates
Module is analysed to image and carries out digitized description be to be analysed to image to intend mapping from two-dimensional space to three-dimensional space.
7. the obstacle distance computing system based on monocular vision as claimed in claim 6, it is characterised in that: image to be analyzed from
Two-dimensional space maps in the following ways to three-dimensional space is quasi-；
It is analysed to image pixel to describe using two-dimensional matrix, the two-dimensional matrix ranks indicate the spatial position of pixel, use
Numerical value or vector indicate the colouring information of pixel.
8. the obstacle distance computing system based on monocular vision as claimed in claim 7, it is characterised in that: image to be analyzed turns
It is as follows to turn to gray level image its two-dimensional matrix；
F (i, j) is the gray value of the i-th row in matrix, jth column in F (x, y) two-dimensional matrix, wherein 1≤i≤m, 1≤j≤n.
9. the obstacle distance computing system based on monocular vision as claimed in claim 8, it is characterised in that: the barrier is high
It spends Information recovering and uses shape from shading (shape from shading, SFS).
10. the obstacle distance computing system based on monocular vision as claimed in claim 9, it is characterised in that: the light and shade is extensive
Complex shape is the relatively high of each point on the three-dimensional surface according to the gray scale light and shade of image to be analyzed digital picture variation recovery object
Degree or surface normal parameter value.
11. a kind of barrier distance measuring method based on monocular vision, which comprises the following steps:
1) barrier single image is shot；
2) single image is pre-processed；
3) image to be analyzed is converted by the single image；
4) it is analysed to image and carries out digitized description completion obstacle height Information recovering；
5) location information of the height, size and bottom of barrier in phase plane is determined；
6) position according to shooting focal length, shooting distance ground level and barrier bottom in phase plane passes through optics phase
Distance between shooting point and barrier is obtained like triangulation calculation.
12. the barrier distance measuring method based on monocular vision as claimed in claim 11, it is characterised in that: implementation steps 2) when,
The pretreatment includes at least picture smooth treatment, image dividing processing and/or morphology processing.
13. the barrier distance measuring method based on monocular vision as claimed in claim 11, it is characterised in that: implementation steps 3) when,
It is that the color image parts of single image are converted to gray level image that the single image, which is converted into image to be analyzed,.
14. the barrier distance measuring method based on monocular vision as claimed in claim 11, it is characterised in that: implementation steps 4) when,
It is analysed to image and carries out digitized description be to be analysed to image to intend mapping from two-dimensional space to three-dimensional space.
15. the barrier distance measuring method based on monocular vision as claimed in claim 14, it is characterised in that: image to be analyzed is from two
Dimension space maps in the following ways to three-dimensional space is quasi-；
It is analysed to image pixel to describe using two-dimensional matrix, the two-dimensional matrix ranks indicate the spatial position of pixel, use
Numerical value or vector indicate the colouring information of pixel.
16. the barrier distance measuring method based on monocular vision as claimed in claim 15, it is characterised in that: image to be analyzed conversion
For gray level image, its two-dimensional matrix is as follows；
F (i, j) is the gray value of the i-th row in matrix, jth column in F (x, y) two-dimensional matrix, wherein 1≤i≤m, 1≤j≤n.
17. the barrier distance measuring method based on monocular vision as claimed in claim 16, it is characterised in that: the obstacle height
Information recovering uses shape from shading (shape from shading, SFS).
18. the barrier distance measuring method based on monocular vision as claimed in claim 17, it is characterised in that: the light and shade restores shape
Shape be according to the gray scale light and shade of image to be analyzed digital picture change restore object three-dimensional surface on each point relative altitude or
Surface normal parameter value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811570292.7A CN109443319A (en) | 2018-12-21 | 2018-12-21 | Barrier range-measurement system and its distance measuring method based on monocular vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811570292.7A CN109443319A (en) | 2018-12-21 | 2018-12-21 | Barrier range-measurement system and its distance measuring method based on monocular vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109443319A true CN109443319A (en) | 2019-03-08 |
Family
ID=65534839
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811570292.7A Pending CN109443319A (en) | 2018-12-21 | 2018-12-21 | Barrier range-measurement system and its distance measuring method based on monocular vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109443319A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111521117A (en) * | 2019-08-21 | 2020-08-11 | 长城汽车股份有限公司 | Monocular vision distance measuring method, storage medium and monocular camera |
| CN112639400A (en) * | 2020-01-21 | 2021-04-09 | 深圳市大疆创新科技有限公司 | Distance measurement method, device, equipment and system based on semantic segmentation |
| WO2022233932A1 (en) | 2021-05-05 | 2022-11-10 | F. Hoffmann-La Roche Ag | Monitoring device for monitoring a sample handling system |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH05209730A (en) * | 1992-01-31 | 1993-08-20 | Toshiba Corp | Image processing apparatus |
| CN101750049A (en) * | 2008-12-05 | 2010-06-23 | 南京理工大学 | Monocular vision vehicle distance measuring method based on road and vehicle information |
| CN103852060A (en) * | 2014-03-19 | 2014-06-11 | 哈尔滨工业大学 | Visible light image distance measuring method based on monocular vision |
| CN104899554A (en) * | 2015-05-07 | 2015-09-09 | 东北大学 | Vehicle ranging method based on monocular vision |
| CN106289099A (en) * | 2016-07-28 | 2017-01-04 | 汕头大学 | A kind of single camera vision system and three-dimensional dimension method for fast measuring based on this system |
| CN205919783U (en) * | 2016-07-28 | 2017-02-01 | 汕头大学 | Monocular vision system |
| CN106443650A (en) * | 2016-09-12 | 2017-02-22 | 电子科技大学成都研究院 | Monocular vision range finding method based on geometric relation |
| CN107218922A (en) * | 2016-12-29 | 2017-09-29 | 恩泊泰（天津）科技有限公司 | A kind of distance-finding method based on monocular camera |
| CN107255813A (en) * | 2017-06-30 | 2017-10-17 | 努比亚技术有限公司 | Distance-finding method, mobile terminal and storage medium based on 3D technology |
| CN108088414A (en) * | 2017-12-05 | 2018-05-29 | 苏州天瞳威视电子科技有限公司 | A kind of monocular distance measuring method |
| CN108375368A (en) * | 2018-01-09 | 2018-08-07 | 上海未来伙伴机器人有限公司 | A kind of embedded onboard distance measuring method and its system based on monocular vision |
-
2018
- 2018-12-21 CN CN201811570292.7A patent/CN109443319A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH05209730A (en) * | 1992-01-31 | 1993-08-20 | Toshiba Corp | Image processing apparatus |
| CN101750049A (en) * | 2008-12-05 | 2010-06-23 | 南京理工大学 | Monocular vision vehicle distance measuring method based on road and vehicle information |
| CN103852060A (en) * | 2014-03-19 | 2014-06-11 | 哈尔滨工业大学 | Visible light image distance measuring method based on monocular vision |
| CN104899554A (en) * | 2015-05-07 | 2015-09-09 | 东北大学 | Vehicle ranging method based on monocular vision |
| CN106289099A (en) * | 2016-07-28 | 2017-01-04 | 汕头大学 | A kind of single camera vision system and three-dimensional dimension method for fast measuring based on this system |
| CN205919783U (en) * | 2016-07-28 | 2017-02-01 | 汕头大学 | Monocular vision system |
| CN106443650A (en) * | 2016-09-12 | 2017-02-22 | 电子科技大学成都研究院 | Monocular vision range finding method based on geometric relation |
| CN107218922A (en) * | 2016-12-29 | 2017-09-29 | 恩泊泰（天津）科技有限公司 | A kind of distance-finding method based on monocular camera |
| CN107255813A (en) * | 2017-06-30 | 2017-10-17 | 努比亚技术有限公司 | Distance-finding method, mobile terminal and storage medium based on 3D technology |
| CN108088414A (en) * | 2017-12-05 | 2018-05-29 | 苏州天瞳威视电子科技有限公司 | A kind of monocular distance measuring method |
| CN108375368A (en) * | 2018-01-09 | 2018-08-07 | 上海未来伙伴机器人有限公司 | A kind of embedded onboard distance measuring method and its system based on monocular vision |
Non-Patent Citations (1)
| Title |
|---|
| 张利: "《基于单目视觉和双目视觉的图像三维重建技术研究》", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111521117A (en) * | 2019-08-21 | 2020-08-11 | 长城汽车股份有限公司 | Monocular vision distance measuring method, storage medium and monocular camera |
| CN112639400A (en) * | 2020-01-21 | 2021-04-09 | 深圳市大疆创新科技有限公司 | Distance measurement method, device, equipment and system based on semantic segmentation |
| WO2022233932A1 (en) | 2021-05-05 | 2022-11-10 | F. Hoffmann-La Roche Ag | Monitoring device for monitoring a sample handling system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110569704B (en) | Multi-strategy self-adaptive lane line detection method based on stereoscopic vision | |
| JP6305171B2 (en) | How to detect objects in a scene | |
| CN102774325A (en) | Rearview reversing auxiliary system and method for forming rearview obstacle images | |
| CN103458261B (en) | Video scene variation detection method based on stereoscopic vision | |
| CN109544599A (en) | A kind of three-dimensional point cloud method for registering based on the estimation of camera pose | |
| CN105913013A (en) | Binocular vision face recognition algorithm | |
| CN110288659A (en) | A kind of Depth Imaging and information acquisition method based on binocular vision | |
| CN110189375A (en) | A kind of images steganalysis method based on monocular vision measurement | |
| CN109443319A (en) | Barrier range-measurement system and its distance measuring method based on monocular vision | |
| Lin et al. | Research on 3D reconstruction in binocular stereo vision based on feature point matching method | |
| CN110378341A (en) | A kind of binocular vision pedestrian distance detection method | |
| CN105352482B (en) | 332 dimension object detection methods and system based on bionic compound eyes micro lens technology | |
| CN111951339A (en) | Image processing method for performing parallax calculation by using heterogeneous binocular cameras | |
| CN106709432B (en) | Human head detection counting method based on binocular stereo vision | |
| CN118244281A (en) | Vision and radar fusion target positioning method and device | |
| CN115690190B (en) | Moving target detection and positioning method based on optical flow image and pinhole imaging | |
| CN113538545B (en) | Monocular depth estimation method based on electro-hydraulic adjustable-focus lens and corresponding camera and storage medium | |
| CN113221739B (en) | Monocular vision-based vehicle distance measuring method | |
| CN110514140B (en) | Three-dimensional imaging method, device, equipment and storage medium | |
| CN110599407B (en) | Human body noise reduction method and system based on multiple TOF cameras in downward inclination angle direction | |
| CN116952154A (en) | Method and system for measuring depth of mouth of detonating tube assembly based on machine vision | |
| CN113129348B (en) | Monocular vision-based three-dimensional reconstruction method for vehicle target in road scene | |
| Li et al. | The method of detecting nearest distance between obstacles and vehicle tail based on binocular vision system | |
| Ying et al. | Technique of measuring leading vehicle distance based on digital image processing theory | |
| CN109690555A (en) | Face detector based on curvature |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200605 Address after: 200031 No. 390, Wukang Road, Xuhui District, Shanghai Applicant after: SAIC MOTOR Corp.,Ltd. Applicant after: DIAS AUTOMOTIVE ELECTRONIC SYSTEM Co.,Ltd. Address before: 201206 Shanghai city Pudong New Area Jinji Road 33 Lane 4 Building 4 floor Applicant before: DIAS AUTOMOTIVE ELECTRONIC SYSTEM Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190308 |
|
| WD01 | Invention patent application deemed withdrawn after publication |