CN106127105A - Unmanned plane vision patrolling method based on the most random Radon conversion - Google Patents
Unmanned plane vision patrolling method based on the most random Radon conversion Download PDFInfo
- Publication number
- CN106127105A CN106127105A CN201610415465.2A CN201610415465A CN106127105A CN 106127105 A CN106127105 A CN 106127105A CN 201610415465 A CN201610415465 A CN 201610415465A CN 106127105 A CN106127105 A CN 106127105A
- Authority
- CN
- China
- Prior art keywords
- radon
- random
- unmanned plane
- image
- line
- 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; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/176—Urban or other man-made structures
-
- 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
- G06T2207/30181—Earth observation
- G06T2207/30184—Infrastructure
Abstract
The invention discloses a kind of unmanned plane vision patrolling method based on the most random Radon conversion, first with unmanned plane image capture device, it is thus achieved that high-tension line graph picture to be processed, and transfer the degraded image of acquisition to gray-scale map;Then the gray-scale map obtained is extracted edge;Then carrying out limiting the Radon conversion of parameter judgement based on angle, the line detected by vertical direction removes;Finally use the random Radon image detection algorithm adding dual threshold span to make decisions, obtain straight line of transmitting electricity.The present invention is based on the line feature that high voltage transmission line is dark target and level of approximation, utilize the multiple dimensioned linear target intensifying method limited based on angle that target is optimized, then to the Radon conversion in the range of result carries out certain angle after screening, and reach to accurately identify the result of high-tension bus-bar by the span of restriction threshold value in Radon converts.
Description
Technical field
The invention belongs to technical field of image processing, be specifically related to unmanned plane vision based on the most random Radon conversion
Patrolling method.
Background technology
In recent years, China's power network development is rapid, but while solution power network development is delayed, intensive high pressure line-group gives height
The damage testing of line ball and maintenance work also bring huge challenge.At present, a lot of countries have carried out high-voltage line Research on Identification
Work.As developed Infrared Detectors, magnetic field detector, ground proximity warning system, millimetre-wave radar, laser radar and being installed on
Survey meter etc. on electric lines of force steel tower.In recent years, domestic correlational study utilizes the RCS electromagnetic scattering of millimere-wave band high-voltage line mostly
Characteristic and as group's property etc. carries out high-voltage line detection, Institutes Of Technology Of Nanjing, South China Science & Engineering University, Harbin Institute of Technology, middle electricity
Ten institutes and middle electric 27 are waited to achieve abundant achievement in this field.But the method is strong to hardware device dependency, use not
Different with detector effect, and detector is expensive, virtually improves the cost of equipment, supply line based on image examines
Survey starts to rise, and becomes study hotspot.
Due to unmanned plane, to have volume little, lightweight, maneuverability, the advantage such as safe and cost-effective, in recent years since, domestic
Development and exploitation to unmanned plane outward all puts into a large amount of man power and material, and many new techniques, new sensor are opened targetedly
Send out and use, such as small industry steam turbine, radar, inertial navigation device etc..But at unmanned plane (particularly SUAV) etc.
Special application field, by volume, the restriction of the factors such as power consumption and weight, the computer with camera as main sensors regards
Feel that technical advantage highlights.
Unmanned plane, using high voltage transmission line as navigation target, carries out the automatic identification of linear target.The most relevant wire mesh
It is that electric wire image zooming-out edge obtains straight line that mark knows method for distinguishing automatically, is then entered high voltage transmission line by Radon conversion
Row identifies.Due to the impact of the factors such as the complexity of high voltage transmission line surrounding, image-forming condition be severe, image mesohigh is transmitted electricity
The edge of line is not it is obvious that additionally plus the interference of noise, the marginal information of extraction is not very reliable, but also comprises a lot
The marginal information of redundancy, brings the biggest difficulty to follow-up judgement and identification.
Summary of the invention
It is an object of the invention to provide unmanned plane vision patrolling method based on the most random Radon conversion, to overcome
The defect that above-mentioned prior art exists, the present invention, based on the line feature that high voltage transmission line is dark target and level of approximation, utilizes
Target is optimized by the multiple dimensioned linear target intensifying method limited based on angle, then result after screening is carried out certain angle
Radon conversion in the range of degree, and reach to accurately identify high-tension electricity by the span of restriction threshold value in Radon converts
The result of line.
For reaching above-mentioned purpose, the present invention adopts the following technical scheme that
Unmanned plane vision patrolling method based on the most random Radon conversion, comprises the following steps:
Step 1: utilize unmanned plane image capture device, it is thus achieved that high-tension line graph picture to be processed, and the figure that degrades that will obtain
As transferring gray-scale map to;
Step 2: the gray-scale map obtained in step 1 is extracted edge;
Step 3: image treated in step 2 carries out limiting based on angle the Radon conversion of parameter judgement, will be perpendicular
Nogata removes to the line detected;
Step 4: use the random Radon image detection algorithm adding dual threshold span that the image through step 3 is entered
Row judgement, obtains straight line of transmitting electricity.
Further, step 2 uses the Canny operator high-voltage line image zooming-out edge to obtaining in step 1.
Further, step 3 limits the Radon conversion of parameter judgement particularly as follows: determine Radon angle of transformation based on angle
Degree is θ ∈ [0,90 °], and the image then processed step 2 carries out Radon conversion, and the accumulation put in obtaining parameter plane is thick
Degree and the coordinate of point.
Further, in step 4, the span of dual threshold is (max_r*threshold1, max_r*
Threshold2), wherein max_r is the maximum of accumulation thickness a little, and threshold1=0.72, threshold2=
0.75。
Further, step 4 adds the random Radon image detection algorithm of dual threshold span particularly as follows: joining
In number plane, when the accumulation thickness of point is in dual threshold span, adjudicate as transmission of electricity straight line;When the accumulation thickness of point does not exists
Time in dual threshold span, adjudicate as interference straight line.
Further, after step 4 obtains transmission of electricity straight line, judged further by identification regulatory factor, obtain the most defeated
Electricity straight line, described identification regulatory factorS is that the high-tension line graph of step 1 collection is as the upper and lower nearby sphere of pixel
The difference of gray value, is mapped to coordinate plane by the coordinate of the point obtained after step 4 is adjudicated, as identification regulatory factor w >
0.02, then prove that the gray value approximation of the upper and lower nearby sphere of straight line, i.e. judgement are final transmission pressure, as identification regulatory factor w
≤ 0.02, then prove that the gray value difference of the upper and lower nearby sphere of straight line is relatively big, i.e. judgement is interference straight line.
Compared with prior art, the present invention has a following useful technique effect:
The present invention, based on the line feature that high voltage transmission line is dark target and level of approximation, utilizes and limits many based on angle
Target is optimized by yardstick linear target intensifying method, then to the Radon in the range of result carries out certain angle after screening
Conversion, the line detected by vertical direction removes, and reaches accurate by the span of restriction threshold value in Radon converts
Identify the result of high-tension bus-bar.
Further, the present invention uses Canny operator to position high-voltage line image zooming-out edge, Canny operator edge
Accuracy and noise resisting ability effect are preferable.
Further, present invention introduces identification regulatory factor, in Radon converts, introduce the identification regulation of grey scale pixel value
The factor can suppress the interference of level of approximation object to a great extent.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of the present invention;
Fig. 2 traditional method and the comparison diagram of the inventive method, wherein (a), (c), (e) are conventional Radon algorithm detection knots
Really;B (), (d), (f) are the present invention quickly random Radon algorithm testing results.
Detailed description of the invention
Below in conjunction with the accompanying drawings the present invention is described in further detail:
Unmanned plane vision patrolling method based on the most random Radon conversion, comprises the following steps:
In step 1 obtain high-tension line graph picture: utilize unmanned plane image capture device, it is thus achieved that high-tension line graph to be processed as rgb,
And transfer acquisition degraded image to gray-scale map i.
Step 2 carries out image edge extraction process with Canny operator:
BW=edge (i, ' canny ', thresh, sigma)
Thresh is susceptibility threshold parameter, and default value is empty matrix [].It is a column vector herein, specifies threshold value for algorithm
Bound.First element is bottom threshold, and second element is upper threshold.If only specifying a threshold element, that
Giving tacit consent to this element is upper threshold, and its value of 0.4 times is as bottom threshold.If threshold parameter is not specified, then algorithm is certainly
Row determines the bound of susceptibility threshold.
In all kinds of edge detection operators, although Roberts operator and Log operator positioning precision are high, but by noise shadow
Ring big;Sobel operator and the relatively large therefore noise removal capability of Prewitt operator template are relatively strong, have smoothing effect, and can filter
Except some noises, remove a part of pseudo-edge, but also smoothed real edge simultaneously, reduce edge precision.Always
From the point of view of body, accuracy and the noise resisting ability effect of location, Canny operator edge are preferable.Therefore the present invention use Canny to calculate
Son carries out image edge extraction.
Step 3 limits parameter judgement based on angle: due to spacing and the impact of angle of camera and high voltage transmission line,
In same scene, during high-voltage line imaging, thickness is inconsistent.Therefore, apply multiple dimensioned linear target to strengthen wave filter,
Can strengthen the high-voltage line target in the range of different scale, but this algorithm high-voltage line target is strengthened out while also electric wire
The vertically target such as bar, trees is strengthened out, and follow-up high-voltage line detection is brought severe jamming by this.In actual application, high pressure is defeated
Electric wire is level of approximation, and the interfering object such as electric pole, trees is vertical.Sentence it is proposed that limit parameter based on angle
Certainly, to suppress the interference to identifying target of the thread-shaped body of vertical direction.Therefore angle is by θ ∈ during this algorithm process each frame data
[0,180 °] narrows down to θ ∈ [0,90 °], theta=randperm (90).
Owing to unmanned plane is that parallel high-voltage line flies and using high-voltage line as navigation target, therefore high-voltage line imaging approximation water
Square to, in order to only strengthen the high-voltage line target of horizontal direction, the high-tension line graph through edge detection process is carried out base as bw
After angle limits parameter judgement, Radon is taked to convert, the accumulation thickness r put in obtaining parameter plane and the coordinate xp of point, with
Realize the detection to the linear pixel in contour images.
Radon formula is as follows:
R (ρ, θ)=∫ ∫ f (x, y) δ (ρ-x cos θ-y sin θ) dxdy (1)
In formula, D is whole image x/y plane;(x y) is gray value;ρ is the zero distance to straight line;θ is distance
With the angle of x-axis, θ ∈ [0,180];δ is Dirac delta function.It makes, and (x, y) linearly ρ=x cos θ+y sin θ amasss
Point.
Radon conversion can be understood as the image projection in ρ-θ space.Each some correspondence image space in ρ-θ space
Straight line, and Radon conversion is image pixel value integration on every straight line, therefore in image, every straight line can be at ρ-θ
Space forms a bright spot. and the detection of straight line is converted in the detection to bright spot of the ρ-θ transform domain, as shown in figure3, in conjunction with 1
Middle direction angle on target limits, the some correspondence image mesohigh line target that brightness value is bigger.
Quick random Radon image detection algorithm in step 4: in order to suppress vertical direction thread-shaped body such as electric pole,
Trees etc., it is proposed that limit parameter judgement based on angle.This algorithm can strengthen the linear target in level of approximation direction, the most again
Can effectively suppress the thread-shaped body of vertical direction, but in some cases, around high-voltage line, there may be the interference of level of approximation
The horizontal edge etc. of object, such as object.When using in this case based on angle restriction parameter judgement, high pressure strengthening
Also can strengthen interfering object or its edge level of approximation while line, thus affect the correct identification of high-voltage line.
In order to overcome this problem, the result after strengthening is being carried out in Radon conversion by limiting the value model of threshold value
Enclose the recognition result reaching only to identify high-tension bus-bar.Accumulation thickness r according to step 3 midpoint, obtains maximum max_r, double
The span of threshold value is (max_r*threshold1, max_r*threshold2).So, by limit in Radon converts
The span determining threshold value can suppress the interference of level of approximation object to a certain extent.Therefore can make
Threshold1=0.72;Threshold2=0.75;
[II, JJ]=find (r >=(max_r*threshold1) &r≤(max_r*threshold2));
Owing to Radon detection is by the direction of each predetermined angular carries out line integral, thus obtain a maximum
Detect straight line, so predetermined angular can be added randomness, theta=randperm (80) by us, increase detection
Randomness, promotes detection speed, reduces the detection time.
Step 5 introduces identification regulatory factor in random Radon image detection algorithm: due to the upper and lower neighbour of high voltage transmission line
The background of nearly scope is much the same, so the gray value approximation of the upper and lower nearby sphere of high voltage transmission line.Therefore definable is adjusted
The joint factorS is the difference of the gray value of the upper and lower nearby sphere of image slices vegetarian refreshments.When for high voltage power transmission line target, by
Similar in the gray value of the upper and lower nearby sphere of high-voltage line, | s | is less, and w is relatively big, so that the gray scale of high voltage transmission line area pixel
The regulatory factor of value becomes big;On the contrary, even if if the interfering object of level of approximation such as rail etc. identify interfering object
Or its edge, but the gray value difference of the upper and lower nearby sphere of this interfering object or its edge is relatively big, and | s | is relatively big, and w is less, from
And make the level of approximation of optimization interfering object (as the horizontal edge of object or riverbank along etc.) grey scale pixel value in region
Regulatory factor diminishes.So, the regulatory factor introducing grey scale pixel value in Radon converts can suppress approximation to a great extent
The interference of horizontal.
In Fig. 2, the scene resolution of 4 width figure tests is respectively 504*373,474*315 and 294*395.Conventional Radon calculates
Method testing result is as shown in figure (a), (c), (e), and the experimental result of inventive algorithm is as shown in figure (b), (d), (f), dotted line
For testing result.The time that calculation procedure runs, the conventional Radon conversionization electric wire detection algorithm time spends and is respectively
11.340241 seconds, 15.140015 seconds, 7.479158 seconds, and erroneous judgement ratio is more serious, field etc. is all mistaken for electric wire, it is judged that just
True part deviation is the biggest, especially in figure (e), and conventional algorithm complete failure, detect substantial amounts of rail and electric pole.And
The new random detection algorithm that the present invention proposes spends respectively 0.755584 second in figure (b), (d), 0.701000 second, the time
Shorten more than 10 times.Especially in result figure (f), the present invention introduces identification regulatory factor in random Radon image detection algorithm,
Gray value difference according to the upper and lower nearby sphere of detected straight line can more accurately detect power transmission line and level of approximation rail, significantly
Promote the accuracy rate of detection, spend the time to be greatly reduced simultaneously, be 0.759467 second.
Claims (6)
1. unmanned plane vision patrolling method based on the most random Radon conversion, it is characterised in that comprise the following steps:
Step 1: utilize unmanned plane image capture device, it is thus achieved that high-tension line graph picture to be processed, and the degraded image of acquisition is turned
For gray-scale map;
Step 2: the gray-scale map obtained in step 1 is extracted edge;
Step 3: image treated in step 2 carries out limiting based on angle the Radon conversion of parameter judgement, will be vertically square
Remove to the line detected;
Step 4: use the random Radon image detection algorithm adding dual threshold span that the image through step 3 is sentenced
Certainly, straight line of transmitting electricity is obtained.
Unmanned plane vision patrolling method based on the most random Radon conversion the most according to claim 1, its feature exists
In, step 2 uses the Canny operator high-voltage line image zooming-out edge to obtaining in step 1.
Unmanned plane vision patrolling method based on the most random Radon conversion the most according to claim 1, its feature exists
In, in step 3 based on angle limit parameter judgement Radon conversion particularly as follows: determine Radon translation-angle as θ ∈ [0,
90 °], the image then processed step 2 carries out Radon conversion, the accumulation thickness put in obtaining parameter plane and the seat of point
Mark.
Unmanned plane vision patrolling method based on the most random Radon conversion the most according to claim 3, its feature exists
In, in step 4, the span of dual threshold is (max_r*threshold1, max_r*threshold2), and wherein max_r is a little
The maximum of accumulation thickness, and threshold1=0.72, threshold2=0.75.
Unmanned plane vision patrolling method based on the most random Radon conversion the most according to claim 3, its feature exists
In, step 4 adds the random Radon image detection algorithm of dual threshold span particularly as follows: in parameter plane, when point
When accumulation thickness is in dual threshold span, adjudicate as transmission of electricity straight line;When the accumulation thickness of point is not in dual threshold span
Time interior, adjudicate as interference straight line.
Unmanned plane vision patrolling method based on the most random Radon conversion the most according to claim 1, its feature exists
In, after step 4 obtains transmission of electricity straight line, judged further by identification regulatory factor, straight line of finally being transmitted electricity, described identification
Regulatory factorS be the high-tension line graph that gathers of step 1 as the difference of the gray value of the upper and lower nearby sphere of pixel, will be through
The coordinate of the point obtained after crossing step 4 judgement is mapped to coordinate plane, as identification regulatory factor w > 0.02, then proves on straight line
The gray value approximation of lower nearby sphere, i.e. judgement are final transmission pressure, when identification regulatory factor w≤0.02, then prove straight line
The gray value difference of nearby sphere is relatively big up and down, i.e. judgement is interference straight line.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610415465.2A CN106127105A (en) | 2016-06-13 | 2016-06-13 | Unmanned plane vision patrolling method based on the most random Radon conversion |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610415465.2A CN106127105A (en) | 2016-06-13 | 2016-06-13 | Unmanned plane vision patrolling method based on the most random Radon conversion |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106127105A true CN106127105A (en) | 2016-11-16 |
Family
ID=57270515
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610415465.2A Pending CN106127105A (en) | 2016-06-13 | 2016-06-13 | Unmanned plane vision patrolling method based on the most random Radon conversion |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106127105A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106777347A (en) * | 2017-01-17 | 2017-05-31 | 广东容祺智能科技有限公司 | A kind of unmanned plane power-line patrolling big data processing system |
CN108509906A (en) * | 2018-03-30 | 2018-09-07 | 长安大学 | Unmanned plane road baseline spot captures dual threshold Radon recognition methods |
CN109426277A (en) * | 2017-08-30 | 2019-03-05 | 广州极飞科技有限公司 | The method and device of motion track planning |
CN109615598A (en) * | 2018-12-10 | 2019-04-12 | 武汉大学 | A kind of power transmission line recognition methods based on the free algorithm of edge mapping parameter |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103247039A (en) * | 2013-05-09 | 2013-08-14 | 河海大学常州校区 | Charged detection method of high-voltage cable based on composite vision |
CN103761524A (en) * | 2014-01-17 | 2014-04-30 | 电子科技大学 | Image-based linear target recognition and extraction method |
CN103810462A (en) * | 2012-11-14 | 2014-05-21 | 中国科学院沈阳自动化研究所 | High voltage transmission line detection method based on linear targets |
CN104573703A (en) * | 2013-10-29 | 2015-04-29 | 中国科学院沈阳自动化研究所 | Method for quickly identifying power transmission line based on partial derivative distribution and boundary strategy |
CN104914434A (en) * | 2015-06-09 | 2015-09-16 | 长安大学 | Cruise helicopter airborne radar high-voltage wire detection and early-warning method |
-
2016
- 2016-06-13 CN CN201610415465.2A patent/CN106127105A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103810462A (en) * | 2012-11-14 | 2014-05-21 | 中国科学院沈阳自动化研究所 | High voltage transmission line detection method based on linear targets |
CN103247039A (en) * | 2013-05-09 | 2013-08-14 | 河海大学常州校区 | Charged detection method of high-voltage cable based on composite vision |
CN104573703A (en) * | 2013-10-29 | 2015-04-29 | 中国科学院沈阳自动化研究所 | Method for quickly identifying power transmission line based on partial derivative distribution and boundary strategy |
CN103761524A (en) * | 2014-01-17 | 2014-04-30 | 电子科技大学 | Image-based linear target recognition and extraction method |
CN104914434A (en) * | 2015-06-09 | 2015-09-16 | 长安大学 | Cruise helicopter airborne radar high-voltage wire detection and early-warning method |
Non-Patent Citations (3)
Title |
---|
曹蔚然 等: "面向旋翼无人机的高压输电线在线检测方法", 《计算机应用研究》 * |
赵利坡 等: "面向巡线无人机高压线实时检测与识别算法", 《小型微型计算机系统》 * |
黄婷婷: "无人机自动巡线方法研究", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106777347A (en) * | 2017-01-17 | 2017-05-31 | 广东容祺智能科技有限公司 | A kind of unmanned plane power-line patrolling big data processing system |
CN109426277A (en) * | 2017-08-30 | 2019-03-05 | 广州极飞科技有限公司 | The method and device of motion track planning |
CN108509906A (en) * | 2018-03-30 | 2018-09-07 | 长安大学 | Unmanned plane road baseline spot captures dual threshold Radon recognition methods |
CN108509906B (en) * | 2018-03-30 | 2022-02-08 | 长安大学 | Double-threshold Radon identification method for line light point capture of unmanned aerial vehicle aerial photography roadbed |
CN109615598A (en) * | 2018-12-10 | 2019-04-12 | 武汉大学 | A kind of power transmission line recognition methods based on the free algorithm of edge mapping parameter |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105528619B (en) | SAR remote sensing image variation detection method based on wavelet transformation and SVM | |
CN103761524B (en) | A kind of linear goal identification based on image and extracting method | |
CN106127105A (en) | Unmanned plane vision patrolling method based on the most random Radon conversion | |
CN108416791A (en) | A kind of monitoring of parallel institution moving platform pose and tracking based on binocular vision | |
CN111255636A (en) | Method and device for determining tower clearance of wind generating set | |
CN108805904A (en) | A kind of moving ship detection and tracking based on satellite sequence image | |
CN106530271B (en) | A kind of infrared image conspicuousness detection method | |
CN105718964B (en) | A kind of visible detection method of power transmission line damper | |
CN111536981B (en) | Embedded binocular non-cooperative target relative pose measurement method | |
CN106156758B (en) | A kind of tidal saltmarsh method in SAR seashore image | |
CN109003277A (en) | A kind of infrared small target in complex background detection method and device | |
CN108537170A (en) | A kind of power equipment firmware unmanned plane inspection pin missing detection method | |
CN109034272A (en) | A kind of transmission line of electricity heat generating components automatic identifying method | |
CN112487912B (en) | Arbitrary direction ship detection method based on improved YOLOv3 | |
CN108229524A (en) | A kind of chimney and condensing tower detection method based on remote sensing images | |
Yang et al. | Detection of morphology defects in pipeline based on 3D active stereo omnidirectional vision sensor | |
CN110675447A (en) | People counting method based on combination of visible light camera and thermal imager | |
CN108052887A (en) | A kind of doubtful illegal land automatic recognition system and method for merging SLAM/GNSS information | |
CN111476762A (en) | Obstacle detection method and device of inspection equipment and inspection equipment | |
CN105787912A (en) | Classification-based step type edge sub pixel localization method | |
CN108109154A (en) | A kind of new positioning of workpiece and data capture method | |
CN109461171A (en) | The small IR targets detection algorithm of DoG filtering is improved based on multichannel | |
CN104408432A (en) | Infrared image target detection method based on histogram modification | |
CN107231553A (en) | Corner location acquisition methods and device | |
Zhao et al. | Infrared small target detection using local component uncertainty measure with consistency assessment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161116 |
|
RJ01 | Rejection of invention patent application after publication |