CN112097693A - Straightness measuring system and method based on unmanned aerial vehicle - Google Patents
Straightness measuring system and method based on unmanned aerial vehicle Download PDFInfo
- Publication number
- CN112097693A CN112097693A CN202010835051.1A CN202010835051A CN112097693A CN 112097693 A CN112097693 A CN 112097693A CN 202010835051 A CN202010835051 A CN 202010835051A CN 112097693 A CN112097693 A CN 112097693A
- Authority
- CN
- China
- Prior art keywords
- measured object
- aerial vehicle
- unmanned aerial
- straightness
- algorithm
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/26—Measuring arrangements characterised by the use of optical techniques for measuring angles or tapers; for testing the alignment of axes
- G01B11/27—Measuring arrangements characterised by the use of optical techniques for measuring angles or tapers; for testing the alignment of axes for testing the alignment of axes
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B64—AIRCRAFT; AVIATION; COSMONAUTICS
- B64C—AEROPLANES; HELICOPTERS
- B64C39/00—Aircraft not otherwise provided for
- B64C39/02—Aircraft not otherwise provided for characterised by special use
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B64—AIRCRAFT; AVIATION; COSMONAUTICS
- B64U—UNMANNED AERIAL VEHICLES [UAV]; EQUIPMENT THEREFOR
- B64U2101/00—UAVs specially adapted for particular uses or applications
- B64U2101/30—UAVs specially adapted for particular uses or applications for imaging, photography or videography
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- 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
- G06T2207/20024—Filtering details
-
- 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
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Geometry (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a straightness measuring system and method based on an unmanned aerial vehicle, and belongs to the field of non-contact measurement. A straightness measuring system based on an unmanned aerial vehicle comprises an image acquisition system, the unmanned aerial vehicle and a ground station system; a straightness measuring method based on an unmanned aerial vehicle comprises a camera calibration algorithm, an edge detection algorithm, a semantic segmentation algorithm and a straightness calculation algorithm. According to the invention, the unmanned aerial vehicle carries the image acquisition system, and the single picture of the measured object is utilized to measure the straightness of the measured object, so that the measurement precision and the measurement efficiency are improved, the labor intensity is reduced, and the working environment is improved.
Description
Technical Field
The invention relates to a straightness measuring system and method based on an unmanned aerial vehicle, and belongs to the field of non-contact measurement.
Background
In the railway field, a contact network is an important component of a traction power supply system of an electrified railway and is called an 'artery' of the electrified railway. The reliability and stability of the contact network operation state directly relate to the operation order of the high-speed railway. Railway contact net wire is the key and the main part of contact net system, and the pantograph of electric locomotive obtains the electric energy through with wire sliding contact, and the contact quality between the two is the key technical indicator who weighs the contact net quality. The smoothness of the wire, namely the straightness, is an important loop, which not only directly reflects the pantograph-catenary relationship, but also determines the current taking level of the pantograph, and ensures the continuity and the stability of the contact between the wire and the pantograph; once the lead is bent and twisted, off-line arc discharge can be generated in the running process of the train, even the lead is burnt out, and the train can not run normally. Therefore, the construction quality of the railway contact network lead directly determines the operation safety and the transportation capacity of the electrified railway.
The straightness measurement is used as an emergency detection and discrimination means for discriminating and judging the straightness measurement, and the emergency detection and discrimination means mainly comprises two methods: one is direct detection on the web; the other is detection on an indoor horizontal tensile testing machine. The measurement results of both measurement methods can describe the straightness of the wire to some extent. Although the above-mentioned detection method can make a direct judgment on whether the wire is straight, the accuracy of the measurement result is affected by many other factors, and it is difficult to give an accurate value as a criterion, mainly for two reasons: (1) hard bending points caused by artificial acting force are difficult to avoid in the sampling operation and sample straightening process, and if the hard bending points are sampling wires from the net, the hard bending points are also influenced by construction stringing operation; (2) the measurement results from different measurement starting points within a certain length have poor repeatability.
Besides the problems of the measurement mode and the measurement precision, the traditional measurement mode has the following problems:
the measurement efficiency is low. At present, a manual measurement mode is adopted, an engineering vehicle is needed to lift measuring personnel to a high place during measurement, three persons cooperate to complete measurement, measurement data is read and recorded manually, and then the next point is measured. The measuring process is time-consuming and greatly influenced by human factors.
The working environment is complex. Railway construction is often performed in severe environments such as high temperature, high humidity, high altitude, strong wind and the like, so that the measurement work of the contact net lead is also often performed in a hard working environment. The working environment is severe, the labor intensity is high, and the test is a harsh test for workers.
The digitization and the intellectualization level are low. At present, a measurement mode of manual measurement and manual entry is far away from the digital and intelligent concept of the construction of the current electrified railway, so that the real-time recording and sharing of measurement data cannot be realized, and secondary use such as statistical analysis of detection data cannot be performed.
Besides the straightness measurement of contact net wires in the field of railway lines, the demand scene of straightness measurement is existed in long-distance conveying pipelines, various pipelines in industrial production sites and the like in the field of petrochemical industry, so that a high-precision, high-efficiency and non-contact type straightness measurement system is needed.
Disclosure of Invention
Taking the straightness measurement of the railway contact network wire as an example, the traditional measurement adopts a mode of 'manual measurement + manual recording', and the mode has the defects of low efficiency, poor precision, high labor intensity, difficult working conditions and the like. Aiming at the problems existing in the straightness measurement, the invention aims to design a straightness measurement system and method based on an unmanned aerial vehicle. The system provided by the invention has stable and reliable hardware and high robustness of a software algorithm, can overcome the defects of low current measurement efficiency, poor precision, high labor intensity and the like, and has good practicability.
In order to achieve the purpose, the technical scheme of the invention is as follows:
the straightness measuring system and method based on the unmanned aerial vehicle comprise an image acquisition system, the unmanned aerial vehicle and a ground station system; the method comprises a camera calibration algorithm, an edge detection algorithm, a semantic segmentation algorithm and a straightness calculation algorithm. The specific implementation process comprises the following steps:
the straightness measuring system and method based on the unmanned aerial vehicle calibrate the camera by adopting a Zhang Zhengyou camera calibration algorithm aiming at the camera in the image acquisition system, obtain internal and external parameters and distortion coefficients of the camera, and eliminate imaging distortion. Based on the calibrated camera, the unmanned aerial vehicle is used for carrying the image acquisition system, and under the conditions of different weather, different angles, different illumination and different time, the image acquisition is carried out on the object to be measured to be used as the object to be measured sample atlas. Based on the detected object sample atlas, edge detection is carried out on the pictures in the sample atlas by using a Candy edge detection algorithm, the edge information of the detected object in the pictures is obtained, and the processed pictures are used as a feature atlas.
The Candy edge detection algorithm comprises the following steps:
I. and smoothing the image and filtering noise by using a Gaussian filter.
And II, calculating the gradient strength and the direction of each pixel point in the image.
And III, applying Non-Maximum Suppression (Non-Maximum Suppression) to eliminate spurious response caused by edge detection.
Determining real and potential edges using Double-Threshold (Double-Threshold) detection.
And V, restraining the isolated weak edge to finish edge detection.
Based on the sample set and the feature atlas after the Candy edge detection processing, the data enhancement is carried out by adopting methods such as vertical mirror image, random clipping, rotation, local distortion, principal component analysis enhancement and the like. The data enhancement method adopts the same data enhancement strategy and parameters aiming at the sample graph and the corresponding characteristic graph so as to keep the consistency of the sample graph and the characteristic graph.
Based on the sample map set and the feature map set after data enhancement, a full convolution neural network model is adopted in the semantic segmentation algorithm, the sample set after data enhancement is used as input, the feature map set after data enhancement is used as output, the full convolution neural network model is trained until a preset index is reached, the training is finished, and the training is used as the semantic segmentation model. The semantic segmentation model can effectively improve the identification accuracy of the measured object and the edge precision of the measured object.
When the system and the method are in a working state, the unmanned aerial vehicle carries the image acquisition system to shoot a measured object in the air, and the shooting distance between the camera and the measured object can be controlled by the unmanned aerial vehicle remote controller so as to control the length and the measurement precision of the shot measured object. If the detection precision is p, the shooting length of the measured object meeting the precision is constrained by the following conditions:
wherein l is the length of the object to be measured, PwThe number of pixels in the horizontal direction of the camera. The shooting distance satisfying the precision is constrained by the following conditions:
wherein d is the shooting distance, f is the camera focal length, and w is the transverse dimension of the camera photosensitive element.
Based on the image of the measured object shot by the image acquisition system, the semantic segmentation model is utilized to identify and detect the edge of the measured object in the image, so that the edge information of the measured object can be obtained.
And measuring the straightness of the measured object in the image by adopting a straightness calculation algorithm based on the edge information of the measured object of the semantic segmentation model. As shown in fig. 4, in the three-dimensional coordinate system O-XYZ, a plane ABCD (expression: Z ═ f, where f is the camera focal length) is an imaging plane; plane EFGH (expression: Z ═ d, where d>0, parameter to be solved) is parallel to the plane ABCD, and the plane EIJH is measuredThe object is actually in a plane, and the plane EFGH intersects with the plane EIJH at EH. According to the edge information of the measured object of the semantic segmentation model, the intersecting lengths of the measured object in the imaging plane ABCD and the AD and BC are p respectively1、p2。
The coordinate of the point A isWherein w is the image imaging surface width and h is the image imaging surface height; coordinate of point E isSimilarly, point coordinates B, C, D, F, G, H, E 'and F' can be obtained. According to the image imaging principle, the following relationship is satisfied:
can be seen from the figureI' point coordinates can be obtained. From the point E, I ', the H coordinate, an expression for the plane EI' H can be determined:
Arx+Bry+Crz+Dr=0
selecting q on the edge of the measured object in the imaging plane1、q2、q3Three points, whereinLet plane EHI' and straight lineHas a cross point of r1The point r can be obtained according to the intersection relationship between the plane and the straight line1I.e. the mid-point q of the image plane1Point r corresponding to the actual object1In the same way, q can be obtained2、q3Corresponding r2、r3And (4) point. Passing point r2To a straight line r1r3As a perpendicular line, the foot is r4Then, thenThe diameter D of the measured object is obtained. In the actual railway construction process, the diameter D of the measured object of the contact net is uniform and known, and the diameter D is determined according to theThe specific numerical value of the parameter d to be solved can be obtained. And obtaining the plane EFGH of the actual position of the measured object. According to the imaging principle, the original size reduction of the measured object can be realized.
And calculating the central point of the measured object based on the original size of the measured object, and fitting a straight line corresponding to the central line according to the central line of the measured object. And obtaining the farthest distance point from the central line to the fitting straight line, namely the measured object straightness measurement result, as shown in fig. 5.
In the embodiment of the invention, the straightness measuring system and method based on the unmanned aerial vehicle acquire the sample image of the measured object by carrying the image acquisition equipment by the unmanned aerial vehicle. And processing the pictures in the sample picture set by using an edge detection algorithm to obtain a feature picture set, and respectively using the sample picture set and the feature picture set after data enhancement as input and output of a training full convolution neural network to obtain a full convolution neural network model. When the system is in a working state, the unmanned aerial vehicle carries image acquisition equipment and acquires images of a measured object according to a set measurement route and an image acquisition task; and transmitting the acquired image to a ground station system in real time, wherein the ground station system adopts a semantic segmentation algorithm to identify and detect the edge of the object to be measured, and finally obtains a straightness measurement result by utilizing a straightness calculation algorithm.
The application of the invention can realize the measuring processes of automatic detection, digital transmission, intelligent identification and measurement, improve the detection efficiency, improve the detection precision and reduce the labor intensity of workers; and is suitable for severe working environments such as high temperature, cold, humidity, high altitude and the like.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic diagram of the system;
FIG. 2 is a flow chart of semantic segmentation model training;
FIG. 3 is a system work flow diagram;
FIG. 4 is a schematic diagram of the solution of the parameter d to be solved;
FIG. 5 is a diagram illustrating the solution of straightness results.
The labels in the figure are: the method comprises the following steps of 1-a measured object, 2-a camera, 3-a stable holder, 4-an unmanned aerial vehicle, 5-a ground station, 6-the edge of the measured object, 7-the axis of the measured object, 8-a straight line fitted by the axis of the measured object, 9-a point with the axis farthest from the straight line, and 10-a point with the axis farthest from the straight line (a straightness measurement result).
Detailed Description
The technical solution in the embodiments of the present invention will be clearly and perfectly described below with reference to the drawings in the embodiments of the present invention, taking railway contact wire straightness measurement as an example. It is to be understood that the described embodiments are merely exemplary of some, and not necessarily all, embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without any inventive step, are within the scope of the present invention.
Referring to fig. 1, fig. 1 shows a working schematic diagram of a system and a method for measuring linearity based on an unmanned aerial vehicle for measuring linearity of a railway contact network conductor according to an embodiment of the present invention, where the working schematic diagram includes a measured object 1, an unmanned aerial vehicle 4, an image acquisition system (a camera 2, a stabilizing pan-tilt 3), and a ground station system 5. When the wire is measured, the unmanned aerial vehicle carries the image acquisition system to shoot the wire, the image acquisition system sends the shot wire picture to the ground station system in a wireless mode, and the ground station system uses corresponding software and algorithm to complete the straightness accuracy measurement on the wire.
In the embodiment, a Xinjiang longitude and latitude M300RTK unmanned aerial vehicle is adopted, a VC-151MC-5H camera of visiworks is carried, the resolution of the camera is 14192 multiplied by 10640, wherein P isw14192. The pixel size is 3.76um × 3.76um, and the focal length f is 66.7 mm. The size of the camera imaging plane is 53361.92um × 40006.4um, where w ═ 53361.92um and h ═ 40006.4 um. In this embodiment, if the measurement accuracy p is 1mm, the maximum measurement length of the wire is lmax4.73m, the maximum shooting distance is dmax=5.913m。
After the camera shoots the wire picture, the wire picture is transmitted to the ground station system in a wireless communication mode. The ground station system realizes the identification of the conducting wire and the conducting wire edge in the picture by utilizing a semantic segmentation model, wherein the lengths of the conducting wire at the left side and the right side in the picture are respectively p1、p2. The coordinates of the point E can be obtained according to the above conditionsCoordinates of point HThe point A, B, C, D, F, G, H, E 'and F' can be obtained by the same method. According to the imaging principle, the following steps are known:
and as can be seen from the figures,the coordinates of point I' can be obtained. From the coordinates of point E, H, I ', a plane expression A of the plane EHI' can be determinedrx+Bry+Crz+Dr=0。
Selecting q on the edge of a conductor in the imaging plane1、q2、q3Three points, wherein Solving the plane EHI' and the lineAt the intersection point of (a), the imaging plane q can be obtained1Point r in the physical coordinate system1In the same way, q can be obtained2、q3Corresponding r2、r3And (4) point. Passing point r2To a straight line r1r3As a perpendicular line, the foot is r4。I.e. the diameter of the wire. The wire diameter is uniform and known, and is therefore according to the equationThe specific value of the shooting distance d can be obtained, and the expression of the plane where the conducting wire is located is obtained.
And restoring the coordinates of the edge of the conducting wire in the picture in the actual imaging plane according to the relation of the intersection point of the straight line and the plane. In the plane EHI', the two-dimensional point cloud coordinates of the central line of the wire are solved, and a linear regression method is used for carrying out linear fitting on the central line. And traversing and solving the distance from the center line point cloud to the fitting straight line, wherein the maximum distance is the straightness measuring result.
In order to improve the system measurement precision and reduce the system error, the straightness result can be calculated for many times in a mode of cutting the edge of the picture; at the same time, q at different positions can be selected for multiple times1、q2、q3Calculating a plurality of straightness results; calculating the average value of the tail removal according to the multiple measurement results; the measurement accuracy can be improved.
The method applies the unmanned aerial vehicle technology, the artificial intelligence technology and the image processing technology to the field of railway contact network wire straightness detection, has higher detection efficiency and precision compared with the traditional manual measurement, and avoids accidental errors caused by the manual measurement; the labor intensity of workers is reduced, and the working comfort of the workers is improved; the railway construction efficiency and the railway construction quality are improved.
The above description is only a preferred embodiment of the present invention, and the present invention is not limited to the above embodiments, and those skilled in the art can make modifications, equivalents, improvements and the like without departing from the spirit and principle of the present invention.
Claims (5)
1. A straightness measuring system based on an unmanned aerial vehicle is characterized by comprising the unmanned aerial vehicle, an image acquisition system and a ground station system; wherein the content of the first and second substances,
the unmanned aerial vehicle is used for carrying an image acquisition system to realize intelligent non-contact measurement;
the image acquisition system comprises a camera and a stabilizing pan-tilt and is used for acquiring an image of a measured object;
the ground station system is a high-performance workstation and is responsible for performing software and algorithm processing on an image of a measured object to realize measurement of the straightness of the measured object.
2. A straightness measuring method based on an unmanned aerial vehicle is characterized by comprising a camera calibration algorithm, an edge detection algorithm, a semantic segmentation algorithm and a straightness calculation algorithm;
the camera calibration algorithm adopts Zhangyingyou camera calibration algorithm to obtain internal and external parameters and distortion coefficients of the camera and eliminate imaging distortion;
based on the calibrated camera, the unmanned aerial vehicle carries an image acquisition system to acquire an image of the measured object as a sample atlas, and the edge detection algorithm identifies the edge information of the measured object in the sample atlas as a feature atlas;
based on the sample atlas and the feature atlas, the semantic segmentation algorithm trains a model by using the sample atlas and the feature atlas after data enhancement to realize the identification of a measured object and the edge of the measured object in an unknown image;
and based on the edge information of the measured object in the image, the straightness calculation algorithm is used for calculating the straightness of the measured object.
3. The unmanned aerial vehicle-based straightness measuring method according to claim 2, wherein the measured object is an object with a circular cross section in the length direction, and straightness of the measured object is measured by using a single picture taken by a monocular camera.
4. The straightness measuring method based on the unmanned aerial vehicle as claimed in claim 2, wherein the edge detection algorithm and the semantic segmentation algorithm are implemented by the following processes: firstly, carrying out sample collection on a measured object by an unmanned aerial vehicle carrying image collection system under different backgrounds to serve as a sample atlas; then, performing edge detection on the sample image by adopting an edge detection algorithm through a Candy algorithm to obtain edge information, and taking a processed result set as a feature image set; then, aiming at the sample graph and the corresponding characteristic graph, the same data enhancement operation is adopted; and finally, training the model by adopting a full convolution neural network model and taking the sample atlas and the feature atlas after data enhancement as the input and the output of the model respectively to obtain a final semantic segmentation model.
5. The straightness measuring method based on the unmanned aerial vehicle as claimed in claim 2, wherein based on a measured object with a uniform diameter and a known size, the straightness calculating algorithm calculates a plane where the measured object is located in a world coordinate system, converts the measured object in an image coordinate system into the world coordinate system, and obtains the actual size of the measured object; fitting the central line of the measured object into a straight line, and calculating the maximum distance between the central line and the fitted straight line to obtain the linearity result of the measured object.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010835051.1A CN112097693A (en) | 2020-08-19 | 2020-08-19 | Straightness measuring system and method based on unmanned aerial vehicle |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010835051.1A CN112097693A (en) | 2020-08-19 | 2020-08-19 | Straightness measuring system and method based on unmanned aerial vehicle |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112097693A true CN112097693A (en) | 2020-12-18 |
Family
ID=73754001
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010835051.1A Pending CN112097693A (en) | 2020-08-19 | 2020-08-19 | Straightness measuring system and method based on unmanned aerial vehicle |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112097693A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113126628A (en) * | 2021-04-26 | 2021-07-16 | 上海联适导航技术股份有限公司 | Method, system and equipment for automatic driving of agricultural machinery and readable storage medium |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100082287A1 (en) * | 2008-09-30 | 2010-04-01 | Xerox Corporation | Methods and systems for detecting spectrophotometer misalignment |
CN102519400A (en) * | 2011-12-15 | 2012-06-27 | 东南大学 | Large slenderness ratio shaft part straightness error detection method based on machine vision |
CN106529555A (en) * | 2016-11-04 | 2017-03-22 | 四川大学 | DR (Digital Radiography) slice lung contour extraction method based on fully convolutional network |
CN107044821A (en) * | 2016-12-30 | 2017-08-15 | 苏州西博三维科技有限公司 | A kind of measuring method and system of contactless tubing object |
CN109284779A (en) * | 2018-09-04 | 2019-01-29 | 中国人民解放军陆军工程大学 | Object detection method based on deep full convolution network |
CN109931889A (en) * | 2019-03-27 | 2019-06-25 | 中铁大桥科学研究院有限公司 | Offset detection system and method based on image recognition technology |
CN109974631A (en) * | 2019-04-26 | 2019-07-05 | 太原科技大学 | A kind of bar Linearity surveying system and detection method |
CN110473175A (en) * | 2019-07-26 | 2019-11-19 | 长春工程学院 | A kind of train axle space linearity measurement method |
US20190370965A1 (en) * | 2017-02-22 | 2019-12-05 | The United States Of America, As Represented By The Secretary, Department Of Health And Human Servic | Detection of prostate cancer in multi-parametric mri using random forest with instance weighting & mr prostate segmentation by deep learning with holistically-nested networks |
CN110781924A (en) * | 2019-09-29 | 2020-02-11 | 哈尔滨工程大学 | Side-scan sonar image feature extraction method based on full convolution neural network |
-
2020
- 2020-08-19 CN CN202010835051.1A patent/CN112097693A/en active Pending
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100082287A1 (en) * | 2008-09-30 | 2010-04-01 | Xerox Corporation | Methods and systems for detecting spectrophotometer misalignment |
CN102519400A (en) * | 2011-12-15 | 2012-06-27 | 东南大学 | Large slenderness ratio shaft part straightness error detection method based on machine vision |
CN106529555A (en) * | 2016-11-04 | 2017-03-22 | 四川大学 | DR (Digital Radiography) slice lung contour extraction method based on fully convolutional network |
CN107044821A (en) * | 2016-12-30 | 2017-08-15 | 苏州西博三维科技有限公司 | A kind of measuring method and system of contactless tubing object |
US20190370965A1 (en) * | 2017-02-22 | 2019-12-05 | The United States Of America, As Represented By The Secretary, Department Of Health And Human Servic | Detection of prostate cancer in multi-parametric mri using random forest with instance weighting & mr prostate segmentation by deep learning with holistically-nested networks |
CN109284779A (en) * | 2018-09-04 | 2019-01-29 | 中国人民解放军陆军工程大学 | Object detection method based on deep full convolution network |
CN109931889A (en) * | 2019-03-27 | 2019-06-25 | 中铁大桥科学研究院有限公司 | Offset detection system and method based on image recognition technology |
CN109974631A (en) * | 2019-04-26 | 2019-07-05 | 太原科技大学 | A kind of bar Linearity surveying system and detection method |
CN110473175A (en) * | 2019-07-26 | 2019-11-19 | 长春工程学院 | A kind of train axle space linearity measurement method |
CN110781924A (en) * | 2019-09-29 | 2020-02-11 | 哈尔滨工程大学 | Side-scan sonar image feature extraction method based on full convolution neural network |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113126628A (en) * | 2021-04-26 | 2021-07-16 | 上海联适导航技术股份有限公司 | Method, system and equipment for automatic driving of agricultural machinery and readable storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11551341B2 (en) | Method and device for automatically drawing structural cracks and precisely measuring widths thereof | |
CN109544679B (en) | Three-dimensional reconstruction method for inner wall of pipeline | |
CN110580717B (en) | Unmanned aerial vehicle autonomous inspection route generation method for electric power tower | |
CN110230998B (en) | Rapid and precise three-dimensional measurement method and device based on line laser and binocular camera | |
CN112686938B (en) | Power transmission line clear distance calculation and safety alarm method based on binocular image ranging | |
JPS62503121A (en) | Method and apparatus for obtaining information regarding the cross-sectional shape or contour of at least a portion of the surface of an object | |
CN106996748A (en) | Wheel diameter measuring method based on binocular vision | |
CN110910443B (en) | Contact net geometric parameter real-time measuring method and device based on single monitoring camera | |
CN110599538B (en) | Method and device for identifying icing thickness of transmission line conductor | |
CN115482195B (en) | Train part deformation detection method based on three-dimensional point cloud | |
CN113313047B (en) | Lane line detection method and system based on lane structure prior | |
CN115810041B (en) | Cable size detection method and system based on multi-camera vision | |
CN103196429B (en) | Method for quickly obtaining and measuring orthophotoquad of city skyline contour line facade | |
CN112802004B (en) | Portable intelligent video detection device for health of power transmission line and pole tower | |
CN115063472A (en) | Deep learning-based luggage identification and measurement method and device | |
CN116152697A (en) | Three-dimensional model measuring method and related device for concrete structure cracks | |
CN112097693A (en) | Straightness measuring system and method based on unmanned aerial vehicle | |
CN114596301A (en) | Coating roller surface defect detection system adopting 3D line laser profile technology | |
CN112396651B (en) | Method for realizing equipment positioning through two-angle image | |
CN113673011A (en) | Method for intelligently identifying tunnel invasion boundary in operation period based on point cloud data | |
CN116958092A (en) | Power transmission line sag rapid measurement method based on improved stereo matching algorithm | |
CN112200778A (en) | Non-contact multi-cable type parameter measurement method and system | |
CN116147477A (en) | Joint calibration method, hole site detection method, electronic device and storage medium | |
RU2650857C1 (en) | System for determining geometric parameters of three-dimensional objects | |
CN115560695A (en) | Global calibration method for linear structured light vision sensor of plane mirror without overlapping view field |
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 |