CN109389612A - A kind of municipal rail train pantograph pan edge detection method - Google Patents
A kind of municipal rail train pantograph pan edge detection method Download PDFInfo
- Publication number
- CN109389612A CN109389612A CN201811009620.6A CN201811009620A CN109389612A CN 109389612 A CN109389612 A CN 109389612A CN 201811009620 A CN201811009620 A CN 201811009620A CN 109389612 A CN109389612 A CN 109389612A
- Authority
- CN
- China
- Prior art keywords
- image
- gradient
- point
- pantograph pan
- follows
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
本发明公开了一种城轨列车受电弓滑板边缘检测方法。该方法为：首先获取受电弓滑板灰度图像信号，使用图像信号与自适应尺寸的高斯卷积核进行卷积运算；然后在3*3的8邻域内用梯度算子计算经过滤波后的图像的梯度幅值及方向；最后对梯度幅值进行非极大值抑制，采用双阈值的自适应提取，进行边缘连接，得到受电弓滑板边缘图像。本发明基于改进的Canny算子，检测结果明显，适用性强。
The invention discloses a method for detecting the edge of a pantograph sliding plate of an urban rail train. The method is as follows: firstly obtain the grayscale image signal of the pantograph slide plate, and use the image signal to perform a convolution operation with a Gaussian convolution kernel of adaptive size; then use the gradient operator to calculate the filtered Gradient amplitude and direction of the image; finally, non-maximum suppression of gradient amplitude, adaptive extraction of double thresholds, edge connection, and edge image of pantograph skateboard are obtained. Based on the improved Canny operator, the invention has obvious detection results and strong applicability.
Description
Technical field
The invention belongs to bow failure detection technique field, especially a kind of municipal rail train pantograph pan edge detection
Method.
Background technique
With the promotion of train speed and the increase of operation mileage, the security performance of train is more and more important.Pantograph one
As be mounted at the top of train, in the process of running with contact line sliding contact, electric energy is obtained from contact net, is that train is being run
The important component of firm energy is obtained in the process.Carbon slipper is the key component that train pantograph system obtains electric energy, in train
During operation, for carbon slipper vulnerable to abrasion, abrasion are once more than that warning value will cause serious traffic accident.
John F.Canny proposes Canny edge detection operator, and this method belongs to multistage edge detection algorithm, by seeking
Optimal detection, oplimal Location and unique response is looked for obtain the edge of image.But Canny edge detection has the drawback that (1)
During first step gaussian filtering, marginal information is also weakened while filtering out picture noise, it is unconspicuous to miss some features
Edge；(2) it is calculated in amplitude and gradient procedure in second step, gradient and amplitude, this method pair is calculated using 2 × 2 fields
Noise is very sensitive, easily detects pseudo-edge；(3) the 4th step be arranged dual threshold during, the selection of dual threshold to edge most
Determination has a significant impact eventually, is difficult to select an optimal value.After threshold value is arranged in artificial experience method, cannot adaptively it be schemed according to every
Piece information selected threshold again.
Summary of the invention
The purpose of the present invention is to provide the good municipal rail train pantograph pan edge detections of a kind of accuracy height, real-time
Method eliminates safe hidden trouble to adopt an effective measure in time.
Realizing the technical solution of the object of the invention is: a kind of municipal rail train pantograph pan edge detection method, base
In improved Canny operator, specifically includes the following steps:
Step 1, pantograph pan gray image signals are obtained, the Gaussian convolution core of picture signal and adaptive size is used
Convolution algorithm is carried out, filtered image is obtained；
Step 2, gradient magnitude and the direction by filtered image are calculated with gradient operator in 8 neighborhoods of 3*3；
Step 3, non-maxima suppression is carried out to gradient magnitude, using the extracted in self-adaptive of dual threshold, carries out edge connection,
Obtain pantograph pan edge image.
Further, acquisition pantograph pan gray image signals described in step 1, use picture signal and adaptive ruler
Very little Gaussian convolution core carries out convolution algorithm, specific as follows:
Gray image signals, and the Gauss with adaptive size are converted by the pantograph pan colour picture signal of acquisition
Convolution kernel carries out convolution algorithm, sets the size of the Gaussian convolution core of adaptive size as m*n, image pixel is ω * h, then certainly
The calculation formula for adapting to the Gaussian convolution core of size is as follows:
Further, the ladder by filtered image is calculated with gradient operator in 8 neighborhoods of 3*3 described in step 2
Amplitude and direction are spent, specific as follows:
The Directional partial derivative and difference formula of 8 neighborhoods of 3*3 are as follows:
Wherein, G (i, j) is the gray value of image of (i, j) point on image, Px(i, j) is the partial derivative in the direction x, Py(i,j)
For the partial derivative in the direction y, P45(i, j) is the partial derivative in 45 ° of directions, P135(i, j) is the partial derivative in 135 ° of directions, fx(i, j) is
Use the image gradient in the direction x that calculus of differences obtains, fy(i, j) is the image gradient in the direction y obtained using calculus of differences；
Then gradient formula are as follows:
Wherein, M (i, j), θ (i, j) are respectively the gradient magnitude and gradient direction of (i, j) point.
Further, non-maxima suppression is carried out to gradient magnitude described in step 3, using adaptively mentioning for dual threshold
It takes, carries out edge connection, obtain pantograph pan edge image, specific as follows:
Step 3.1, the adjacent greatest gradient value for obtaining the point along the gradient direction of point (i, j) using linear interpolation, if
The gradient value is greater than the gradient value of point (i, j), then the gray value of point (i, j) is set to 0；Conversely, the then gray scale of retention point (i, j)
Value；
Step 3.2 defines mean μ in classiAnd variance
Wherein, if the sum that gray value is the pixel of j is nj, pjThe ratio of entire image sum of all pixels is accounted for for it；I is root
According to the subscript for the different gray value intervals that entire image is divided, the gray value interval sum divided is equal to according to whole pictures
The classification number that vegetarian refreshments gradient magnitude divides；
Image after step 3.3, non-maxima suppression, gradient magnitude are divided into 3 classes: C0For non-edge point pixel, gradient width
Being worth range is [0,1 ..., k]；C1For marginal point pixel, gradient magnitude range is [k+1, k+2 ..., m]；C2For doubtful marginal point
Pixel, gradient magnitude range are [m+1, m+2 ..., l-1]；
Define the evaluation function J that high-low threshold value and gradient amplitude histogram are adaptively determined based on minimum interclass variance
(k, m) are as follows:
Formula (7) are substituted into formula (8) and are derived by:
Wherein, gained solution k, m are pixel gradient magnitude threshold value；Image border is carried out with m using k to connect, and can be obtained
To pantograph pan edge image.
Compared with prior art, the present invention its remarkable advantage is: (1) improved Canny operator is based on, using adaptive height
When this accounting method carries out noise reduction filtering, the actual size in target detection is preferably obtained；(2) contiguous range is expanded to 3*
38 neighborhoods, so that detection is more accurate；(3) minimum interclass variance and gradient amplitude histogram are used, edge extracting is improved
Detection accuracy, testing result is obvious, and method applicability is strong.
Detailed description of the invention
Fig. 1 is the flow diagram of municipal rail train pantograph pan edge detection method of the present invention.
Fig. 2 is pantograph left-half image in the embodiment of the present invention.
Fig. 3 is that pantograph left-half is schemed through the processing of traditional Canny operator in the embodiment of the present invention.
Fig. 4 is the improved Canny operator processing figure of pantograph left-half in the embodiment of the present invention.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawing.
In conjunction with Fig. 1, municipal rail train pantograph pan edge detection method of the present invention, first acquisition pantograph pan grayscale image
As signal, convolution algorithm is carried out using the Gaussian convolution core of picture signal and adaptive size；Then it is used in 8 neighborhoods of 3*3
Gradient operator calculates gradient magnitude and direction by filtered image；Non-maxima suppression finally is carried out to gradient magnitude,
Using the extracted in self-adaptive of dual threshold, edge connection is carried out, pantograph pan edge image is obtained, comprising the following steps:
Step 1, pantograph pan gray image signals are obtained, the Gaussian convolution core of picture signal and adaptive size is used
Convolution algorithm is carried out, specific as follows:
Gray image signals, and the Gauss with adaptive size are converted by the pantograph pan colour picture signal of acquisition
Convolution kernel carries out convolution algorithm, sets the size of the Gaussian convolution core of adaptive size as m*n, image pixel is ω * h, then certainly
The calculation formula for adapting to the Gaussian convolution core of size is as follows:
Step 2, gradient magnitude and direction by filtered image, tool are calculated with gradient operator in 8 neighborhoods of 3*3
Body is as follows are as follows:
The Directional partial derivative and difference formula of 8 neighborhoods of 3*3 are as follows:
Wherein, G (i, j) is the gray value of image of (i, j) point on image, Px(i, j) is the partial derivative in the direction x, Py(i,j)
For the partial derivative in the direction y, P45(i, j) is the partial derivative in 45 ° of directions, P135(i, j) is the partial derivative in 135 ° of directions, fx(i, j) is
Use the image gradient in the direction x that calculus of differences obtains, fy(i, j) is the image gradient in the direction y obtained using calculus of differences；
Then gradient formula are as follows:
Wherein, M (i, j), θ (i, j) are respectively the gradient magnitude and gradient direction of (i, j) point.
Step 3, non-maxima suppression is carried out to gradient magnitude, using the extracted in self-adaptive of dual threshold, carries out edge connection,
Pantograph pan edge image is obtained, specific as follows:
Step 3.1, the adjacent greatest gradient value for obtaining the point along the gradient direction of point (i, j) using linear interpolation, if
The gradient value is greater than the gradient value of point (i, j), then the gray value of point (i, j) is set to 0；Conversely, the then gray scale of retention point (i, j)
Value；
Step 3.2 defines mean μ in classiAnd variance
Wherein, if the sum that gray value is the pixel of j is nj, pjThe ratio of entire image sum of all pixels is accounted for for it；I is root
According to the subscript for the different gray value intervals that entire image is divided, the gray value interval sum divided is equal to according to whole pictures
The classification number that vegetarian refreshments gradient magnitude divides；
Image after step 3.3, non-maxima suppression, gradient magnitude are divided into 3 classes: C0For non-edge point pixel, gradient width
Being worth range is [0,1 ..., k]；C1For marginal point pixel, gradient magnitude range is [k+1, k+2 ..., m]；C2For doubtful marginal point
Pixel, gradient magnitude range are [m+1, m+2 ..., l-1]；
Define the evaluation function J that high-low threshold value and gradient amplitude histogram are adaptively determined based on minimum interclass variance
(k, m) are as follows:
Formula (7) are substituted into formula (8) and are derived by:
Wherein, gained solution k, m are pixel gradient magnitude threshold value；Image border is carried out with m using k to connect, and can be obtained
To pantograph pan edge image.
Embodiment 1
Using municipal rail train pantograph pan edge detection method of the present invention, improved edge detection operator is tested
Analysis, Fig. 2 be image capturing system acquisition pantograph left-half image, respectively with tradition Canny operator, adaptively
Canny operator carries out edge detection process, result figure such as Fig. 3, Fig. 4.
It is preferable to the filtration result of noise when using improved Canny operator extraction image border in conjunction with Fig. 3, Fig. 4,
Pseudo-edge in improved edge detection graph is less, and image is apparent, conducive to carrying out image border connection and improving system
Detection accuracy.
Claims (4)
1. a kind of municipal rail train pantograph pan edge detection method, which is characterized in that be based on improved Canny operator, specifically
The following steps are included:
Step 1, pantograph pan gray image signals are obtained, are carried out using the Gaussian convolution core of picture signal and adaptive size
Convolution algorithm obtains filtered image；
Step 2, gradient magnitude and the direction by filtered image are calculated with gradient operator in 8 neighborhoods of 3*3；
Step 3, non-maxima suppression is carried out to gradient magnitude, using the extracted in self-adaptive of dual threshold, carries out edge connection, obtain
Pantograph pan edge image.
2. municipal rail train pantograph pan edge detection method according to claim 1, which is characterized in that described in step 1
Acquisition pantograph pan gray image signals, use the Gaussian convolution core of picture signal and adaptive size to carry out convolution fortune
It calculates, specific as follows:
Gray image signals, and the Gaussian convolution with adaptive size are converted by the pantograph pan colour picture signal of acquisition
Core carries out convolution algorithm, sets the size of the Gaussian convolution core of adaptive size as m*n, image pixel is ω * h, then adaptively
The calculation formula of the Gaussian convolution core of size is as follows:
3. municipal rail train pantograph pan edge detection method according to claim 1, which is characterized in that described in step 2
In 8 neighborhoods of 3*3 with gradient operator calculate by filtered image gradient magnitude and direction, it is specific as follows:
The Directional partial derivative and difference formula of 8 neighborhoods of 3*3 are as follows:
Wherein, G (i, j) is the gray value of image of (i, j) point on image, Px(i, j) is the partial derivative in the direction x, Py(i, j) is the side y
To partial derivative, P45(i, j) is the partial derivative in 45 ° of directions, P135(i, j) is the partial derivative in 135 ° of directions, fx(i, j) is to use
The image gradient in the direction x that calculus of differences obtains, fy(i, j) is the image gradient in the direction y obtained using calculus of differences；
Then gradient formula are as follows:
Wherein, M (i, j), θ (i, j) are respectively the gradient magnitude and gradient direction of (i, j) point.
4. municipal rail train pantograph pan edge detection method according to claim 1, which is characterized in that described in step 3
Non-maxima suppression is carried out to gradient magnitude, using the extracted in self-adaptive of dual threshold, carry out edge connection, it is sliding to obtain pantograph
Plate edge image, specific as follows:
Step 3.1, the adjacent greatest gradient value for obtaining the point along the gradient direction of point (i, j) using linear interpolation, if the ladder
Angle value is greater than the gradient value of point (i, j), then the gray value of point (i, j) is set to 0；Conversely, the then gray value of retention point (i, j)；
Step 3.2 defines mean μ in classiAnd variance
Wherein, if the sum that gray value is the pixel of j is nj, pjThe ratio of entire image sum of all pixels is accounted for for it；I is according to whole
The subscript for the different gray value intervals that width image is divided, the gray value interval sum divided are equal to according to whole pixels
The classification number that gradient magnitude divides；
Image after step 3.3, non-maxima suppression, gradient magnitude are divided into 3 classes: C0For non-edge point pixel, gradient magnitude range
For [0,1 ..., k]；C1For marginal point pixel, gradient magnitude range is [k+1, k+2 ..., m]；C2For doubtful marginal point pixel, ladder
Spending amplitude range is [m+1, m+2 ..., l-1]；
Define the evaluation function J (k, m) that high-low threshold value and gradient amplitude histogram are adaptively determined based on minimum interclass variance
Are as follows:
Formula (7) are substituted into formula (8) and are derived by:
Wherein, gained solution k, m are pixel gradient magnitude threshold value；Image border is carried out using k and m to connect, can be obtained by
Pantograph slider edge image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811009620.6A CN109389612A (en) | 2018-08-31 | 2018-08-31 | A kind of municipal rail train pantograph pan edge detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811009620.6A CN109389612A (en) | 2018-08-31 | 2018-08-31 | A kind of municipal rail train pantograph pan edge detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109389612A true CN109389612A (en) | 2019-02-26 |
Family
ID=65418514
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811009620.6A Pending CN109389612A (en) | 2018-08-31 | 2018-08-31 | A kind of municipal rail train pantograph pan edge detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109389612A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110660071A (en) * | 2019-08-23 | 2020-01-07 | 中山市奥珀金属制品有限公司 | Automatic edge detection double-threshold setting method and system |
| CN111578838A (en) * | 2020-05-25 | 2020-08-25 | 安徽工业大学 | Gear size visual measurement device and measurement method |
| CN111784722A (en) * | 2020-07-15 | 2020-10-16 | 山东派蒙机电技术有限公司 | Improved Canny lane line edge detection algorithm |
| CN111985329A (en) * | 2020-07-16 | 2020-11-24 | 浙江工业大学 | Remote sensing image information extraction method based on FCN-8s and improved Canny edge detection |
| CN114973089A (en) * | 2022-05-30 | 2022-08-30 | 福州大学 | Contact net and pantograph contact point detection method based on image vision algorithm |
| CN120013890A (en) * | 2025-01-20 | 2025-05-16 | 交通运输部公路科学研究所 | A method and system for testing the durability of a pantograph contact grid for highway trucks |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102426649A (en) * | 2011-10-13 | 2012-04-25 | 石家庄开发区冀科双实科技有限公司 | Simple high-accuracy steel seal digital automatic identification method |
| CN102486867A (en) * | 2010-12-04 | 2012-06-06 | 中国科学院沈阳自动化研究所 | An Image Gaussian Filtering Method and Implementation Device with Dynamically Configurable Modes |
| CN104680127A (en) * | 2014-12-18 | 2015-06-03 | 闻泰通讯股份有限公司 | Gesture identification method and gesture identification system |
| CN107067382A (en) * | 2017-05-11 | 2017-08-18 | 南宁市正祥科技有限公司 | A kind of improved method for detecting image edge |
| CN107301661A (en) * | 2017-07-10 | 2017-10-27 | 中国科学院遥感与数字地球研究所 | High-resolution remote sensing image method for registering based on edge point feature |
| CN107452007A (en) * | 2017-07-05 | 2017-12-08 | 国网河南省电力公司 | A kind of visible ray insulator method for detecting image edge |
| CN107478176A (en) * | 2017-08-14 | 2017-12-15 | 广东工业大学 | A kind of pantograph loss detection method and system |
| CN108090498A (en) * | 2017-12-28 | 2018-05-29 | 广东工业大学 | A kind of fiber recognition method and device based on deep learning |
| CN108197619A (en) * | 2017-12-27 | 2018-06-22 | 百度在线网络技术（北京）有限公司 | A kind of localization method based on signboard image, device, equipment and storage medium |
-
2018
- 2018-08-31 CN CN201811009620.6A patent/CN109389612A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102486867A (en) * | 2010-12-04 | 2012-06-06 | 中国科学院沈阳自动化研究所 | An Image Gaussian Filtering Method and Implementation Device with Dynamically Configurable Modes |
| CN102426649A (en) * | 2011-10-13 | 2012-04-25 | 石家庄开发区冀科双实科技有限公司 | Simple high-accuracy steel seal digital automatic identification method |
| CN104680127A (en) * | 2014-12-18 | 2015-06-03 | 闻泰通讯股份有限公司 | Gesture identification method and gesture identification system |
| CN107067382A (en) * | 2017-05-11 | 2017-08-18 | 南宁市正祥科技有限公司 | A kind of improved method for detecting image edge |
| CN107452007A (en) * | 2017-07-05 | 2017-12-08 | 国网河南省电力公司 | A kind of visible ray insulator method for detecting image edge |
| CN107301661A (en) * | 2017-07-10 | 2017-10-27 | 中国科学院遥感与数字地球研究所 | High-resolution remote sensing image method for registering based on edge point feature |
| CN107478176A (en) * | 2017-08-14 | 2017-12-15 | 广东工业大学 | A kind of pantograph loss detection method and system |
| CN108197619A (en) * | 2017-12-27 | 2018-06-22 | 百度在线网络技术（北京）有限公司 | A kind of localization method based on signboard image, device, equipment and storage medium |
| CN108090498A (en) * | 2017-12-28 | 2018-05-29 | 广东工业大学 | A kind of fiber recognition method and device based on deep learning |
Non-Patent Citations (4)
| Title |
|---|
| JING CHEN: "The Effect of Kernel Size of CNNs for Lung Nodule Classification", 《2017 9TH INTERNATIONAL CONFERENCE ON ADVANCED INFOCOMM TECHNOLOGY (ICAIT)》 * |
| 姚小文: "基于图像处理的城轨列车受电弓检测技术研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
| 李牧: "自适应Canny算子边缘检测技术", 《万方数据知识服务平台》 * |
| 陈叶健: "城轨列车受电弓滑板磨耗检测技术研究", 《万方数据知识服务平台》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110660071A (en) * | 2019-08-23 | 2020-01-07 | 中山市奥珀金属制品有限公司 | Automatic edge detection double-threshold setting method and system |
| CN111578838A (en) * | 2020-05-25 | 2020-08-25 | 安徽工业大学 | Gear size visual measurement device and measurement method |
| CN111784722A (en) * | 2020-07-15 | 2020-10-16 | 山东派蒙机电技术有限公司 | Improved Canny lane line edge detection algorithm |
| CN111985329A (en) * | 2020-07-16 | 2020-11-24 | 浙江工业大学 | Remote sensing image information extraction method based on FCN-8s and improved Canny edge detection |
| CN111985329B (en) * | 2020-07-16 | 2024-03-29 | 浙江工业大学 | Remote sensing image information extraction method based on FCN-8s and improved Canny edge detection |
| CN114973089A (en) * | 2022-05-30 | 2022-08-30 | 福州大学 | Contact net and pantograph contact point detection method based on image vision algorithm |
| CN120013890A (en) * | 2025-01-20 | 2025-05-16 | 交通运输部公路科学研究所 | A method and system for testing the durability of a pantograph contact grid for highway trucks |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109389612A (en) | A kind of municipal rail train pantograph pan edge detection method | |
| CN109410230B (en) | Improved Canny image edge detection method capable of resisting noise | |
| CN105354541B (en) | The SAR image object detection method of view-based access control model attention model and constant false alarm rate | |
| CN109633633B (en) | Life signal enhancement method based on segmented classification enhancement processing | |
| CN107808383B (en) | Rapid detection method for SAR image target under strong sea clutter | |
| Lawgaly et al. | Weighted averaging-based sensor pattern noise estimation for source camera identification | |
| CN104700421A (en) | Edge detection algorithm based on canny self-adaptive threshold value | |
| CN101739686A (en) | Moving object tracking method and system thereof | |
| CN104268872A (en) | Consistency-based edge detection method | |
| CN102968625A (en) | Ship distinguishing and tracking method based on trail | |
| CN109376740A (en) | A kind of water gauge reading detection method based on video | |
| CN105719283A (en) | Road surface crack image detection method based on Hessian matrix multi-scale filtering | |
| CN108872997A (en) | A kind of seabed line detecting method based on side scan sonar data fusion and accurate processing | |
| CN108921813A (en) | Unmanned aerial vehicle detection bridge structure crack identification method based on machine vision | |
| CN104424641A (en) | Detection method for image fuzzy tampering | |
| CN110321855A (en) | A kind of greasy weather detection prior-warning device | |
| CN108830899A (en) | A kind of center of circle detection method that the quick self-adapted angle point based on geometrical characteristic combines | |
| CN103488992A (en) | Oil spilling detection method for complex SAR image scenes | |
| CN105225244A (en) | Based on the noise detection method that minimum local mean square deviation calculates | |
| CN109118453A (en) | A kind of image processing method that background inhibits | |
| CN107798688A (en) | Motion estimate method, method for early warning and automobile anti-rear end collision prior-warning device | |
| CN110717900B (en) | Pantograph abrasion detection method based on improved Canny edge detection algorithm | |
| Wang et al. | License plate location algorithm based on edge detection and morphology | |
| Ma et al. | Artificial object edge detection based on enhanced Canny algorithm for high-speed railway apparatus identification | |
| CN110473255A (en) | A kind of ship bollard localization method divided based on multi grid |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190226 |
|
| RJ01 | Rejection of invention patent application after publication |