CN109410237A - A kind of slag picture dividing method based on laser three-D camera - Google Patents
A kind of slag picture dividing method based on laser three-D camera Download PDFInfo
- Publication number
- CN109410237A CN109410237A CN201811075132.5A CN201811075132A CN109410237A CN 109410237 A CN109410237 A CN 109410237A CN 201811075132 A CN201811075132 A CN 201811075132A CN 109410237 A CN109410237 A CN 109410237A
- Authority
- CN
- China
- Prior art keywords
- slag
- camera
- piece
- laser
- connected domain
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20152—Watershed segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30132—Masonry; Concrete
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention proposes a kind of the slag picture dividing method based on laser three-D camera, step are as follows: a feature extraction unit of several scanline groups slagging piece spatial informations that laser three-D camera generates；Using comprehensive altitudinal gradient, the attribute of height mean value and altitudinal gradient mean value comprehensive descision slag piece, slag piece binary image is established according to slag piece attribute；Slag piece binary image is corroded and dilation operation respectively, non-slag panel region is carried out to connected domain respectively based on slag piece morphological feature and is filtered out；The morphological feature between the connected domain after filtering out is obtained, the slag piece that is blocked is filtered out, obtains the surface slag piece segmented image not being blocked.The present invention, which mutually obtains slag piece distribution space information based on laser three-D camera, can avoid influence of the uneven illumination to image segmentation result, judgement is carried out to slag piece occlusion area based on spatial altitude information and realizes the segmentation of slag picture, rejecting is blocked slag piece, improves the segmentation of surface slag piece and feature extraction counts accuracy rate.
Description
Technical field
The present invention relates to the technical field of tunnel piercing equipment construction more particularly to a kind of slags based on laser three-D camera
Picture dividing method, for being split to the slag piece information generated in hard rock TBM digging process and feature extraction.
Background technique
Hard rock tunnel development machine (hereinafter referred to as hard rock TBM) is that one kind is specifically applied to excavate rock tunnel and underpass
The large-scale high-tech construction equipment of engineering.Slag piece is the product that knife cutting rock directly acts on, therefore contains a large amount of rock
Body information.Application No. is the Chinese patent applications of CN201621061653.1 to propose a kind of slag film shooting device, but is not situated between
It continues and how information extraction is carried out to the slag piece of acquisition.Application No. is the Chinese patent applications of CN201710222227.4 to propose base
In the image partition method of watershed algorithm, to obtain the characteristic parameter of slag piece, but not can avoid using traditional CCD camera
When the slag picture of acquisition is split, the characteristic parameter accuracy rate sought after dividing when being blocked due to illumination, slag piece
Instability problem.
Summary of the invention
Obtaining slag picture traditional CCD camera has that uneven illumination, slag piece stack block mutually, to slag piece
The segmentation of image and subsequent characteristics extraction bring very big technical difficulty, and the technical issues of influence accuracy, the present invention is proposed
A kind of slag picture dividing method based on laser three-D camera obtains the spatial information of slag piece using laser three-D camera, uses
In the influence for overcoming traditional CCD camera to be illuminated by the light, spatial altitude information is split slag piece occlusion area and solves traditional CCD
The slag piece occlusion issue that camera can not be handled.
In order to achieve the above object, the technical scheme of the present invention is realized as follows: it is a kind of based on laser three-D camera
Slag picture dividing method, its step are as follows:
Step 1: the slag picture conveyed on belt feeder is obtained using laser three-D camera slag film shooting device；
Step 2: a feature extraction list of several scanline groups slagging piece spatial informations that laser three-D camera generates
The slag piece spatial information of a feature extraction unit is converted three-dimensional space matrix by member；
Step 3: comprehensive altitudinal gradient is sought after carrying out smooth noise reduction to three-dimensional space matrix, seeks feature extraction respectively
The height mean value and altitudinal gradient mean value of unit utilize comprehensive altitudinal gradient, height mean value and altitudinal gradient mean value comprehensive descision
Slag piece establishes slag piece binary image according to slag piece attribute in the attribute of certain point；
Step 4: carrying out corrosion and dilation operation to slag piece binary image respectively, connected domain B1 after indicia etched and
Connected domain B2 after expansion；Non- slag panel region is carried out to connected domain B1 and connected domain B2 respectively based on slag piece morphological feature to filter
It removes, counts as the connected domain B1 ' and connected domain B2 ' after filtering out；
Step 5: obtaining the morphological feature of connected domain B1 ' and connected domain B2 ', is based on geometric center distance and slag piece form
The comprehensive descision index for learning feature filters out the slag piece that is blocked, and obtains the surface slag piece segmented image and slag piece feature not being blocked
Information.
The laser three-D camera slag film shooting device includes laser three-D camera and shooting bracket, and shooting bracket is fixed on
The surface of the belt feeder of slag piece is transported, laser three-D camera is fixed on shooting bracket, and phase is equipped with above laser three-D camera
Machine protective cover.
The laser three-D camera is connected with rotary encoder, and the shooting frame rate of laser three-D camera is by rotary encoder
Pulse control, the pulse that laser three-D camera is generated according to rotary encoder generates corresponding scan line.
The three-dimensional space matrix is A1=[x, y, Z (x, y)], wherein x is the row coordinate of certain point in slag picture, y
For column coordinate, Z (x, y) is the height value of this coordinate points；The comprehensive altitudinal gradient G includes carrying out smooth noise reduction by the direction x
The altitudinal gradient G sought afterwardsxWith the altitudinal gradient G for seek after smooth noise reduction by the direction yy, andSmoothly
The method of noise reduction is including but not limited to sliding average or median filtering.
The judgment method of the slag piece attribute are as follows: when three-dimensional space matrix is the certain point (x, y) in A1 while meeting Z
(x, y)>k1*Zmean&G (x, y)<k2*G (mean), this point are slag piece, attribute 1；Otherwise attribute is 0；Wherein, k1 and k2 are
Adjustable parameter, Zmean is height mean value and Gmean is altitudinal gradient mean value, and G (x, y) represents the gradient at coordinate points (x, y)
Value.
The slag piece morphological feature includes containing but being not limited to region area S, long axis a, short axle b, perimeter L, circularity C；And
Non- slag panel region criterion are as follows: S > SmaxOr S < SminOr a > amaxOr b < bminOr C < Cmin；Wherein, maximum
Region area Smax, Minimum Area area Smin, maximum long axis amax, minimum short axle bmin, minimum roundness CminIt is that slag piece form is special
Levy threshold value.
The method for building up of comprehensive descision index in the step 5 are as follows: based on the connected domain B1 ' after corroding, individually count
Calculate the comprehensive evaluation index vector M in each region in each region connected domain B1 ' and connected domain B2 ', and comprehensive evaluation index value:
Ci=4 π Si/Li 2
Wherein, Z1Ix, Z1Iy is the center-of-mass coordinate of the ith zone of connected domain B1 ', i=1,2 ... p；Z2Jx, Z2Jy is to connect
The center-of-mass coordinate in j-th of region of logical domain B2 ', j=1,2 ... q；P and q is respectively B1 ' and B2 ' independent communication domain number；Ci,Si,
LiThe respectively circularity of the ith zone of connected domain B1 ', region area, perimeter；
By p comprehensive evaluation index value MiForm comprehensive evaluation index vector M={ M1, M2..., Mi..., Mp}。
The dividing method of the surface slag piece are as follows: the number according to independent communication domain in connected domain B1 ' and connected domain B2 ' is poor
N, rejects the corresponding connected region of preceding n maximum value of comprehensive evaluation index vector M, and remaining connected domain is to reject to be blocked
The segmentation of surface slag piece is completed in independent communication domain after slag piece.
Beneficial effects of the present invention: relative to traditional image partition method based on Visible Light CCD Camera, it is based on laser
Three-dimensional camera, which carries out Image Acquisition, can avoid influence of the uneven illumination to image segmentation result；Meanwhile laser three-D camera obtains
Slag piece distribution space information is taken, judgement is carried out to slag piece occlusion area based on spatial altitude information and realizes the segmentation of slag picture, is picked
Except the slag piece that is blocked, improves the segmentation of surface slag piece and feature extraction counts accuracy rate.The present invention can be used as separate functional blocks
It is embedded into host computer or remote control terminal, for obtaining driving relevant information, auxiliary master according to laser three-D slag piece segmented image
Driver or intelligent driving system carry out digging control.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is the installation diagram of laser three-D camera slag film shooting device of the invention.
Fig. 2 is the slag picture after binaryzation of the present invention.
Fig. 3 is the image of connected domain after present invention corrosion.
Fig. 4 is the image of connected domain after present invention expansion.
Fig. 5 is the surface slag piece segmentation result image rejected after capped slag piece.
In figure, 1 is belt feeder, and 2 be shooting bracket, and 3 be phase machine protective cover, and 4 be laser three-D camera.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under that premise of not paying creative labor
Embodiment shall fall within the protection scope of the present invention.
A kind of slag picture dividing method based on laser three-D camera, its step are as follows:
Step 1: the slag picture conveyed on belt feeder is obtained using laser three-D camera slag film shooting device.
As shown in Figure 1, the laser three-D camera slag film shooting device includes laser three-D camera 4 and shooting bracket 2, clap
It takes the photograph bracket 2 and is fixed on the surface for transporting the belt feeder 1 of slag piece, laser three-D camera 4 is fixed on shooting bracket 2, laser three
It ties up and is equipped with phase machine protective cover above camera 4, for protective cover for protecting laser three-D camera 4, reduction flies up slag piece to laser three-D
The influence of the acquisition image of camera 4.
The laser three-D camera 4 is connected with rotary encoder, and the shooting frame rate of laser three-D camera is by rotary coding
The pulse control of device, laser three-D camera 4 generate corresponding scan line according to the pulse that rotary encoder generates.Rotary encoder
The triggering that is shot as laser three-D camera of pulse kept away for realizing the slag piece acquisition of information of the fixed conveying distance of belt feeder
Exempt from the slag piece three-dimensional information obtained when belt feeder velocity jump deformation.
Step 2: a feature extraction list of several scanline groups slagging piece spatial informations that laser three-D camera generates
The slag piece spatial information of a feature extraction unit is converted three-dimensional space matrix by member.
The every traveling 1m of belt feeder, rotary encoder generate 1000 pulses, and rotary encoder starts laser three-D camera and produces
Raw 1000 scan lines, every 1000 scanline groups at slag piece spatial information as a feature extraction unit.The three-dimensional
Space matrix is A1=[x, y, Z (x, y)], wherein x is the row coordinate of certain point in slag picture, and y is column coordinate, Z (x, y)
For the height value of this coordinate points.
Step 3: comprehensive altitudinal gradient is sought after carrying out smooth noise reduction to three-dimensional space matrix, seeks feature extraction respectively
The height mean value and altitudinal gradient mean value of unit utilize comprehensive altitudinal gradient, height mean value and altitudinal gradient mean value comprehensive descision
The attribute in certain point of slag piece establishes slag piece binary image according to slag piece attribute.
The comprehensive altitudinal gradient G includes the altitudinal gradient G for carrying out seeking after smooth noise reduction by the direction xxWith by the direction y into
The altitudinal gradient G sought after the smooth noise reduction of rowy, andThe method of smooth noise reduction is flat including but not limited to sliding
Equal or median filtering.Establish space-time matrix A 2=[x, y, Z (x, y), G (x, y)].
The judgment method of the slag piece attribute are as follows: when three-dimensional space matrix is the certain point (x, y) in A1 while meeting Z
(x, y)>k1*Zmean&G (x, y)<k2*G (mean), this point are slag piece, attribute 1；Otherwise attribute is 0；Wherein, k1 and k2 are
Adjustable parameter is defaulted as 1.Wherein, Zmean is height mean value and Gmean is altitudinal gradient mean value, and G (x, y) represents coordinate points
Gradient value at (x, y).
Attribute of the slag piece information unit at certain point is judged simultaneously according to height and gradient information, and judging result is slag
Piece, four seed type of slag piece edge, slag piece gap and background, wherein attribute sets 1 when being slag piece, sets 0 when being other, establishes slag piece
Binary image B, as shown in Figure 2.
Step 4: carrying out corrosion and dilation operation to slag piece binary image respectively, connected domain B1 after indicia etched and
Connected domain B2 after expansion；Non- slag panel region is carried out to connected domain B1 and connected domain B2 respectively based on slag piece morphological feature to filter
It removes, counts as the connected domain B1 ' and connected domain B2 ' after filtering out.
The slag piece morphological feature includes containing but being not limited to region area S, long axis a, short axle b, perimeter L, circularity C；And
Non- slag panel region criterion are as follows: S > SmaxOr S < SminOr a > amaxOr b < bminOr C < Cmin；Wherein, maximum
Region area Smax, Minimum Area area Smin, maximum long axis amax, minimum short axle bmin, minimum roundness CminIt is that slag piece form is special
Levy threshold value.
It carries out non-slag panel region to connected domain B1 and connected domain B2 respectively by above-mentioned criterion to filter out, i.e. connected domain
B1 ' is to carry out non-slag panel region criterion again after carrying out etching operation to binary image B, and non-slag panel region is filtered out removal
Connection area image afterwards；Connected domain B2 ' is to carry out non-slag panel region again after carrying out binary image B expansive working to determine mark
Non- slag panel region is filtered out the connection area image after removal, as shown in Figures 3 and 4 by standard, wherein rectangle mark is connected region
Minimum circumscribed rectangle.
Step 5: obtaining the morphological feature of connected domain B1 ' and connected domain B2 ', is based on geometric center distance and slag piece form
The comprehensive descision index for learning feature filters out the slag piece that is blocked, and obtains the surface slag piece segmented image and slag piece feature not being blocked
Information.
Connected domain B1 ' after corrosion separates the slag piece for occurring blocking, and the connected domain B2 ' after expansion, which has slag piece, to be hidden
The region of gear is combined together, and therefore, the present invention carries out comprehensive descision by the morphological feature of connected domain B1 ' and connected domain B2 '.
The method for building up of comprehensive descision index in the step 5 are as follows: based on the connected domain B1 ' after corroding, individually count
Calculate the comprehensive evaluation index vector M in each region in each region connected domain B1 ' and connected domain B2 ', and comprehensive evaluation index value:
Ci=4 π Si/Li 2
Wherein, Z1Ix, Z1Iy is the center-of-mass coordinate of the ith zone of connected domain B1 ', i=1,2 ... p；Z2Jx, Z2Jy is to connect
The center-of-mass coordinate in j-th of region of logical domain B2 ', j=1,2 ... q；P and q is respectively B1 ' and B2 ' independent communication domain number；Ci,Si,
LiThe respectively circularity of the ith zone of connected domain B1 ', region area, perimeter.In field of image processing, center-of-mass coordinate, circle
Degree, region area, perimeter are all common features, can be acquired by ripe algorithm.
By p comprehensive evaluation index value MiForm comprehensive evaluation index vector M={ M1, M2..., Mi..., Mp}。
The dividing method of the surface slag piece are as follows: the number according to independent communication domain in connected domain B1 ' and connected domain B2 ' is poor
N, rejects the corresponding connected region of preceding n maximum value of comprehensive evaluation index vector M, and remaining connected domain is to reject to be blocked
The segmentation of surface slag piece is completed in independent communication domain after slag piece.The number in independent communication domain in connected domain B1 ' and connected domain B2 '
It finds out after coming, the two, which is subtracted each other, can find out number difference n.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Within mind and principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (8)
1. a kind of slag picture dividing method based on laser three-D camera, which is characterized in that its step are as follows:
Step 1: the slag picture conveyed on belt feeder is obtained using laser three-D camera slag film shooting device；
Step 2: a feature extraction unit of several scanline groups slagging piece spatial informations that laser three-D camera generates,
Three-dimensional space matrix is converted by the slag piece spatial information of a feature extraction unit；
Step 3: comprehensive altitudinal gradient is sought after carrying out smooth noise reduction to three-dimensional space matrix, seeks feature extraction unit respectively
Height mean value and altitudinal gradient mean value, utilize comprehensive altitudinal gradient, height mean value and altitudinal gradient mean value comprehensive descision slag piece
Attribute, slag piece binary image is established according to slag piece attribute；
Step 4: corrosion and dilation operation, connected domain B1 and expansion after indicia etched are carried out respectively to slag piece binary image
Connected domain B2 afterwards；Non- slag panel region is carried out to connected domain B1 and connected domain B2 respectively based on slag piece morphological feature to filter out, and is united
It is calculated as the connected domain B1 ' after filtering out and connected domain B2 '；
Step 5: obtaining the morphological feature of connected domain B1 ' and connected domain B2 ', special based on geometric center distance and slag piece morphology
The comprehensive descision index of sign filters out the slag piece that is blocked, and obtains the surface slag piece segmented image not being blocked and slag piece characteristic information.
2. the slag picture dividing method according to claim 1 based on laser three-D camera, which is characterized in that described to swash
Light three-dimensional camera slag film shooting device includes laser three-D camera (4) and shooting bracket (2), and shooting bracket (2) is fixed on transport
The surface of the belt feeder (1) of slag piece, laser three-D camera (4) is fixed in shooting bracket (2), on laser three-D camera (4)
Side is equipped with phase machine protective cover.
3. the slag picture dividing method according to claim 2 based on laser three-D camera, which is characterized in that described to swash
Light three-dimensional camera (4) is connected with rotary encoder, the shooting frame rate of laser three-D camera by rotary encoder pulse control,
Laser three-D camera (4) generates corresponding scan line according to the pulse that rotary encoder generates.
4. the slag picture dividing method according to claim 1 based on laser three-D camera, which is characterized in that described three
Dimension space matrix be A1=[x, y, Z (x, y)], wherein x be slag picture in certain point row coordinate, y be column coordinate, Z (x,
It y) is the height value of this coordinate points；The comprehensive altitudinal gradient G includes that the height sought after smooth noise reduction ladder is carried out by the direction x
Spend GxWith the altitudinal gradient G for seek after smooth noise reduction by the direction yy, andSmoothly the method for noise reduction includes
But it is not limited to sliding average or median filtering.
5. the slag picture dividing method according to claim 4 based on laser three-D camera, which is characterized in that the slag
The judgment method of piece attribute are as follows: when three-dimensional space matrix is the certain point (x, y) in A1 while meeting Z (x, y) > k1*Zmean&G
(x, y) < k2*G (mean), this point are slag piece, attribute 1；Otherwise attribute is 0；Wherein, k1 and k2 is adjustable parameter, and Zmean is
Height mean value and Gmean are altitudinal gradient mean value, and G (x, y) represents the gradient value at coordinate points (x, y).
6. the slag picture dividing method according to claim 1 based on laser three-D camera, which is characterized in that the slag
Piece morphological feature includes containing but being not limited to region area S, long axis a, short axle b, perimeter L, circularity C；And non-slag panel region determines
Standard are as follows: S > SmaxOr S < SminOr a > amaxOr b < bminOr C < Cmin；Wherein, maximum region area Smax, most
Zonule area Smin, maximum long axis amax, minimum short axle bmin, minimum roundness CminIt is slag piece morphological feature threshold value.
7. the slag picture dividing method according to claim 1 based on laser three-D camera, which is characterized in that the step
The method for building up of comprehensive descision index in rapid five are as follows: based on the connected domain B1 ' after corroding, it is every individually to calculate connected domain B1 '
The comprehensive evaluation index vector M in a region and each region in connected domain B2 ', and comprehensive evaluation index value:
Wherein, Z1Ix, Z1Iy is the center-of-mass coordinate of the ith zone of connected domain B1 ', i=1,2 ... p；Z2Jx, Z2Jy is connected domain
The center-of-mass coordinate in j-th of region of B2 ', j=1,2 ... q；P and q is respectively B1 ' and B2 ' independent communication domain number；Ci,Si,LiPoint
It Wei not the circularity of ith zone of connected domain B1 ', region area, perimeter；
By p comprehensive evaluation index value MiForm comprehensive evaluation index vector M={ M1, M2..., Mi..., Mp}。
8. the slag picture dividing method according to claim 7 based on laser three-D camera, which is characterized in that the table
The dividing method of face slag piece are as follows: according to the number difference n in independent communication domain in connected domain B1 ' and connected domain B2 ', reject overall merit
The corresponding connected region of preceding n maximum value of indicator vector M, remaining connected domain are independently connecting after rejecting the slag piece that is blocked
Logical domain, completes the segmentation of surface slag piece.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811075132.5A CN109410237B (en) | 2018-09-14 | 2018-09-14 | Slag slice image segmentation method based on laser three-dimensional camera |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811075132.5A CN109410237B (en) | 2018-09-14 | 2018-09-14 | Slag slice image segmentation method based on laser three-dimensional camera |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109410237A true CN109410237A (en) | 2019-03-01 |
| CN109410237B CN109410237B (en) | 2020-06-26 |
Family
ID=65464882
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811075132.5A Active CN109410237B (en) | 2018-09-14 | 2018-09-14 | Slag slice image segmentation method based on laser three-dimensional camera |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109410237B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112127896A (en) * | 2020-09-18 | 2020-12-25 | 武汉大学 | Automatic acquisition and analysis system and method for TBM excavation rock slag information |
| CN113096059A (en) * | 2019-12-19 | 2021-07-09 | 合肥君正科技有限公司 | Method for eliminating night light source interference shielding detection of in-vehicle monitoring camera |
| CN113569943A (en) * | 2021-07-26 | 2021-10-29 | 中铁工程装备集团有限公司 | Deep neural network-based slag piece bulk early warning method, system and device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN206016814U (en) * | 2016-09-18 | 2017-03-15 | 中铁工程装备集团有限公司 | A kind of record and the device for analyzing TBM slag pieces |
| CN107133963A (en) * | 2017-04-07 | 2017-09-05 | 中国铁建重工集团有限公司 | Image processing method and device, the method and device of slag piece distributional analysis |
| CN107194943A (en) * | 2017-04-07 | 2017-09-22 | 北京工业大学 | Image partition method and device, image partition method and device for slag piece |
| CN107219227A (en) * | 2017-04-07 | 2017-09-29 | 中国铁建重工集团有限公司 | Slag piece on-line analysis and system |
| CN107862677A (en) * | 2017-10-16 | 2018-03-30 | 中铁第四勘察设计院集团有限公司 | The Tunnel Lining Cracks recognition methods of thresholding algorithm and system between a kind of class based on gradient |
| CN108182440A (en) * | 2018-01-09 | 2018-06-19 | 中铁工程装备集团有限公司 | It is a kind of that the method for obtaining surrounding rock category is identified based on slag picture |
-
2018
- 2018-09-14 CN CN201811075132.5A patent/CN109410237B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN206016814U (en) * | 2016-09-18 | 2017-03-15 | 中铁工程装备集团有限公司 | A kind of record and the device for analyzing TBM slag pieces |
| CN107133963A (en) * | 2017-04-07 | 2017-09-05 | 中国铁建重工集团有限公司 | Image processing method and device, the method and device of slag piece distributional analysis |
| CN107194943A (en) * | 2017-04-07 | 2017-09-22 | 北京工业大学 | Image partition method and device, image partition method and device for slag piece |
| CN107219227A (en) * | 2017-04-07 | 2017-09-29 | 中国铁建重工集团有限公司 | Slag piece on-line analysis and system |
| CN107862677A (en) * | 2017-10-16 | 2018-03-30 | 中铁第四勘察设计院集团有限公司 | The Tunnel Lining Cracks recognition methods of thresholding algorithm and system between a kind of class based on gradient |
| CN108182440A (en) * | 2018-01-09 | 2018-06-19 | 中铁工程装备集团有限公司 | It is a kind of that the method for obtaining surrounding rock category is identified based on slag picture |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113096059A (en) * | 2019-12-19 | 2021-07-09 | 合肥君正科技有限公司 | Method for eliminating night light source interference shielding detection of in-vehicle monitoring camera |
| CN113096059B (en) * | 2019-12-19 | 2023-10-31 | 合肥君正科技有限公司 | Method for eliminating interference shielding detection of night light source by in-vehicle monitoring camera |
| CN112127896A (en) * | 2020-09-18 | 2020-12-25 | 武汉大学 | Automatic acquisition and analysis system and method for TBM excavation rock slag information |
| CN112127896B (en) * | 2020-09-18 | 2021-12-21 | 武汉大学 | Automatic acquisition and analysis system and method for TBM excavation rock slag information |
| CN113569943A (en) * | 2021-07-26 | 2021-10-29 | 中铁工程装备集团有限公司 | Deep neural network-based slag piece bulk early warning method, system and device |
| CN113569943B (en) * | 2021-07-26 | 2023-10-24 | 中铁工程装备集团有限公司 | Slag piece large-block early warning method, system and device based on deep neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109410237B (en) | 2020-06-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210264176A1 (en) | Hazard detection from a camera in a scene with moving shadows | |
| CN109410237A (en) | A kind of slag picture dividing method based on laser three-D camera | |
| US9773317B2 (en) | Pedestrian tracking and counting method and device for near-front top-view monitoring video | |
| CN111079611B (en) | Automatic extraction method for road surface and marking line thereof | |
| CN105046198B (en) | A kind of lane detection method | |
| US9111444B2 (en) | Video and lidar target detection and tracking system and method for segmenting moving targets | |
| JP6274557B2 (en) | Moving surface information detection apparatus, moving body device control system using the same, and moving surface information detection program | |
| CN105426864A (en) | Multiple lane line detecting method based on isometric peripheral point matching | |
| US9189691B2 (en) | Three-dimensional object detection device and three-dimensional object detection method | |
| CN106525000A (en) | A road marking line automatic extracting method based on laser scanning discrete point strength gradients | |
| CN107273816B (en) | Traffic speed limit label detection recognition methods based on vehicle-mounted forward sight monocular camera | |
| JPWO2013157301A1 (en) | Three-dimensional object detection apparatus and three-dimensional object detection method | |
| CN106254827B (en) | Intelligent group mist identification early warning method and device | |
| CN102867311A (en) | Target tracking method and target tracking device | |
| EP2879383B1 (en) | Three-dimensional object detection device and three-dimensional object detection method | |
| CN107563373B (en) | Unmanned aerial vehicle landing area active safety detection method based on stereoscopic vision and application | |
| CN108280401A (en) | A kind of pavement detection method, apparatus, cloud server and computer program product | |
| CN104966062A (en) | Video monitoring method and device | |
| Malinovskiy et al. | Video-based vehicle detection and tracking using spatiotemporal maps | |
| CN110334692A (en) | A kind of blind way recognition methods based on image procossing | |
| JPWO2014017521A1 (en) | Three-dimensional object detection device | |
| US20150009292A1 (en) | Three-dimensional object detection device | |
| CN102043957B (en) | Method of Vehicle Segmentation based on concave spots of image | |
| CN104361317A (en) | Bayonet type video analysis based safety belt unsecured behavior detection system and method | |
| CN115273138B (en) | Human body detection system and passenger flow camera |
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 |