CN109376623A - Method for detecting transmission tower based on synthetic aperture radar image - Google Patents
Method for detecting transmission tower based on synthetic aperture radar image Download PDFInfo
- Publication number
- CN109376623A CN109376623A CN201811167126.2A CN201811167126A CN109376623A CN 109376623 A CN109376623 A CN 109376623A CN 201811167126 A CN201811167126 A CN 201811167126A CN 109376623 A CN109376623 A CN 109376623A
- Authority
- CN
- China
- Prior art keywords
- pixel
- link area
- image
- transmission tower
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000005540 biological transmission Effects 0.000 title claims abstract description 62
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000001514 detection method Methods 0.000 claims abstract description 34
- 238000013528 artificial neural network Methods 0.000 claims abstract description 24
- 230000005611 electricity Effects 0.000 claims description 15
- 229910002056 binary alloy Inorganic materials 0.000 claims description 5
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 230000010287 polarization Effects 0.000 abstract description 2
- 238000012545 processing Methods 0.000 abstract description 2
- 238000012544 monitoring process Methods 0.000 description 13
- 238000005516 engineering process Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000006641 stabilisation Effects 0.000 description 3
- 238000011105 stabilization Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 239000003595 mist Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000003306 harvesting Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000012634 optical imaging Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
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
- G06V20/182—Network patterns, e.g. roads or rivers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/48—Extraction of image or video features by mapping characteristic values of the pattern into a parameter space, e.g. Hough transformation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The embodiment of the invention provides a method for detecting a transmission tower based on a Synthetic Aperture Radar (SAR) image, which comprises the steps of processing a Synthetic Aperture Radar (SAR) image with high resolution to obtain an interested Region (ROI), calculating the half-variance characteristics of the corresponding Region, and detecting the transmission tower by using a neural network. The invention does not require that the image is a complete polarization SAR image with high price, and for the same SAR image, especially under the conditions that the outline of a high-voltage transmission tower on the image is not clear and a plurality of false targets exist, the invention can detect more transmission tower targets, and the false alarm rate of the detection is lower than that of the prior art.
Description
Technical field
The present invention relates to transmission line of electricity monitoring technical field, more particularly to one kind are defeated based on Synthetic Aperture Radar images detection
The method of electric pole tower.
Background technique
With the fast development of Chinese society economy, the demand to electric energy is continuously increased, so that China's power grid scale also exists
Constantly expand.Important component of the transmission line of electricity as power grid, is widely distributed in various regions, and the ground through area
The environmental abnormalities such as shape, geology, meteorology are complicated.The generation of the events such as extreme climate, geological disaster will all threaten power transmission line road transport
Capable safety and stabilization, or even a wide range of power outages accident of power grid is caused, heavy losses are caused to national economy and people's life.
Since the operating status of transmission line of electricity and shaft tower decides the stabilization and safety of entire power grid, to a wide range of transmission line of electricity and
The safe condition of shaft tower is monitored, and provides foundation to take precautions against and eliminating all kinds of security risks in time, it is ensured that transmission line of electricity operation
Safety and stabilization, have very important significance.
Traditional various on-line monitoring techniques, such as Chinese invention patent 201310025414.5 are disclosed based on distribution
The transmission line online monitoring system of energy harvesting, although accurate prison can be compared to the state of transmission line of electricity and shaft tower
It surveys, but need to be mounted so that all kinds of on-line monitoring sensors and development monitoring data transmission device, therefore these technologies are applied to
Monitoring cost can be increased considerably when in a wide range of transmission line of electricity and shaft tower monitoring, and monitoring data transmission is in communication inconvenience
Remote districts and mountain area are difficult to realize.
Airborne line walking technology based on platforms such as helicopters, such as nothing disclosed in Chinese invention patent 201410789453.7
People's helicopter grid power transmission route inspection method, although may be implemented in larger range transmission line of electricity and shaft tower carry out
Monitoring, but be easy to be influenced by environmental factors such as weather, and monitoring cost is also relatively high.
With the continuous development of satellite remote sensing technology, both at home and abroad to based on high-resolution optical imaging monitor transmission line of electricity
And shaft tower is studied, single width optical image can cover thousands of square kilometres, and price can be controlled in 10,000 yuans
Hereinafter, can ground meet cost-effective monitoring carried out to a wide range of transmission line of electricity and shaft tower.But the technology is vulnerable to cloud and mist etc.
The influence of factor, it is difficult to realize and round-the-clock round-the-clock monitoring is carried out to transmission line of electricity and shaft tower.
Compared with optical sensor, satellite-borne synthetic aperture radar (Synthetic Aperture Radar, SAR) equally may be used
To obtain the high resolution image of decimeter level, it can clearly reflect that the scattering texture of high-voltage power transmission tower and transmission pressure is special
Sign, and be imaged and do not limited by cloud and mist, sleet, solar irradiation condition etc., round-the-clock, round-the-clock can be carried out to interested target
Dynamic monitoring.Single width image can also cover thousands of square kilometres, and price is no more than 20,000 yuans, therefore high resolution SAR
Image can meet well to carry out round-the-clock, round-the-clock to a wide range of transmission line of electricity and shaft tower, cost-effectively monitors.
In conclusion the present invention provides and a kind of detects transmission tower automatically based on high resolution synthetic aperture radar image
Method, can round-the-clock, round-the-clock, automatic accurate detection cost-effectively is carried out to transmission tower.
Summary of the invention
Transmission tower is detected based on high resolution synthetic aperture radar image automatically the embodiment of the invention provides a kind of
Method, comprising:
Step 1: obtaining the High-resolution SAR Images in transmission line of electricity region and being pre-processed；
Step 2: removing the background clutter of the SAR image using global constant false alarm rate method, and generate the first binary system
Image；
Step 3: detecting the linear target in first binary image using Hough transform method, and described two
Linear target is removed in system image, generates the second binary image；
Step 4: obtaining the boundary of link area using eight neighborhood algorithm, the first link area is obtained；
Step 5: merging neighbouring first link area, the second link area is formed, area is removed and is not equal to first
Second link area of area threshold value, obtains third link area；
Step 6: calculating the geometric center of the third link area, as the central point of the third link area, take
It around the central point and include the rectangular area of the third link area as interested region (Region of
Interesting,ROI)；
Step 7: the ROI is divided into L fan-shaped and K annulus, L × K sub-regions are formed, L sector is with described
The central point of third link area is vertex, and K annulus is using the central point of the third link area as the center of circle, wherein first
A annulus includes the minimum rectangle being made of pixel, and n-th annulus includes the rectangle around the N-1 annulus, k-th annulus packet
Containing entire ROI；
Step 8: the ROI is expressed as two-dimensional matrixWherein θkIndicate k-th of fan
The central angle of shape, ρlIndicate the radius of first of annulus, M (ρl,θk) indicate described k-th fan-shaped and first of annulus overlapping region
Remove the described k-th fan-shaped amplitude average value that subregion is formed by with the l-1 annulus overlapping region.
Step 9: along the annular radii construct polar coordinates semi-variance function, and successively with each annulus pair
Semivariance feature is extracted in the fan-shaped region answered:
Wherein h is the distance between two annulus, is fixed value, ζh(θ) and ζh(ρ) respectively indicates different annular and difference
Polar coordinates semi-variance function textural characteristics on direction, wherein from ζhTo find out in the formula of (θ), fan-shaped center of circle angle is constant,
Only annular radii changes；From ζhFind out in the formula of (ρ), annular radii is constant, and fan-shaped direction changes；
Step 10: being examined using polar coordinates semi-variance function textural characteristics as the input quantity of neural network using neural network
Survey transmission tower.
Preferably, above-mentioned that background clutter is removed using global constant false alarm rate method, and the first binary image is generated, packet
It includes:
Using the background clutter of K statistical distribution model f (x) description SAR image；
False alarm rate P is setfa, pass throughCalculate detection threshold value T；
For each pixel, when the intensity value of pixel is greater than detection threshold value T, global constant false alarm rate method detector determines
The pixel is target, the pixel is designated as binary value 1, and retain the range value of the pixel, when the intensity value of pixel
No more than detection threshold value T, then the pixel is determined for clutter, the pixel is designated as binary value 0, and by the width of the pixel
Angle value is set as 0, obtains the first binary image.
Preferably, the above-mentioned boundary that link area is obtained using eight neighborhood algorithm, obtains the first link area, comprising:
From second binary image, the pixel that binary value is 1 is obtained, as search starting point；
In the direction of the clock, respectively from the eight neighborhood of upper and lower and left and right both direction search described search starting point；
When search is less than the pixel that binary value is 1 in the eight neighborhood in described search starting point, described search is risen
Point is rejected；
When searched in the eight neighborhood in described search starting point binary value be 1 pixel, two that described search is arrived
The pixel that hex value is 1 searches for the eight neighborhood of described search starting point, obtains new search starting point as search starting point, until
Return to initial search starting point；
Whole search starting points form the boundary of link area, obtain the first link area.
Preferably, above-mentioned first area threshold value=transmission tower length ×=SAR image resolution.
Preferably, the geometric center of above-mentioned third link area is (Cx, Cy), wherein
A indicates the area of the third link area:xiAnd yiRespectively indicate the third
For each pixel along distance to the coordinate value with orientation, N indicates the sum of pixel in the third link area in link area
Mesh, CxAnd CyThe third link area geometric center is respectively indicated in distance to the coordinate with orientation.
Preferably, above-mentioned L fan-shaped central angle is 10 degree.
Preferably, the above method further include: judge the detection accuracy for the transmission tower that the neural network detects, such as
Fruit detection accuracy is greater than detection threshold value, and the transmission tower that the neural network detects is exported as a result；If detection
Precision is not more than detection threshold value, increases the image quantity for training the neural network and/or increases the neural network
The number of plies detects transmission tower using neural network again, until detection accuracy is greater than detection threshold value.
Compared with prior art, expensive full-polarization SAR image is necessary for present invention does not require image, and for
Identical SAR image, the high-voltage power transmission tower profile especially on image is less clear and there are many false targets
In the case of, the present invention can detecte out more transmission tower targets, and it is lower than the prior art to detect false alarm rate.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, will make below to required in the embodiment of the present invention
Attached drawing is briefly described, for those of ordinary skill in the art, without creative efforts, also
Other drawings may be obtained according to these drawings without any creative labor.
Fig. 1 shows in one embodiment of the invention and detects power transmission rod automatically based on high resolution synthetic aperture radar image
The flow diagram of the method for tower；
Fig. 2 shows the schematic diagrames that ROI is divided into L × K sub-regions in one embodiment of the invention.
Specific embodiment
The feature and exemplary embodiment of various aspects of the invention is described more fully below, in order to make mesh of the invention
, technical solution and advantage be more clearly understood, with reference to the accompanying drawings and embodiments, the present invention is further retouched in detail
It states.It should be understood that specific embodiment described herein is only configured to explain the present invention, it is not configured as limiting the present invention.
To those skilled in the art, the present invention can be real in the case where not needing some details in these details
It applies.Below the description of embodiment is used for the purpose of better understanding the present invention to provide by showing example of the invention.
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to
Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence " including ... ", it is not excluded that including
There is also other identical elements in the process, method, article or equipment of the element.
Power transmission rod is detected based on high resolution synthetic aperture radar image automatically as shown in Figure 1, the present embodiment provides one kind
The method of tower, method includes the following steps:
Step 1: obtaining the High-resolution SAR Images in transmission line of electricity region and being pre-processed；The present embodiment is using Germany
The high resolution image that business TerraSAR-X satellite is provided along rail lift direction, specific performance parameter are as follows: X-band；It hangs down
Directly-vertical polarization；3 meters of resolution ratio；37 degree of incidence angles；Single width image coverage area is 50km (orientation) × 30km (distance
To).It include that geocoding, registration and speckle noise such as filter out at the routine operations to the pretreatment that SAR image carries out.
Step 2: removing the background clutter of the SAR image using global constant false alarm rate method, and generate the first binary system
Image；The step specifically includes:
Step 2.1, the background clutter that SAR image is described using K statistical distribution model f (x)；
Step 2.2, setting false alarm rate Pfa, pass throughCalculate detection threshold value T；
Step 2.3, for each pixel, when the intensity value of pixel is greater than detection threshold value T, global CFAR detector determines
The pixel is target, the pixel is designated as binary value 1, and retain the range value of the pixel, when the intensity value of pixel
No more than detection threshold value T, then the pixel is determined for clutter, the pixel is designated as binary value 0, and by the width of the pixel
Angle value is set as 0, obtains the first binary image.
Step 3: detecting the linear target in first binary image using Hough transform method, and described two
Linear target is removed in system image, generates the second binary image；Due to line caused by SAR image such as road, bridge
Shape target can detect to transmission tower and bring adverse effect, it is therefore desirable to first remove these linear targets.
Step 4: obtaining the boundary of link area using eight neighborhood algorithm, the first link area is obtained；It specifically includes:
Step 4.1, from second binary image, obtain binary value be 1 pixel, as search starting point；
Step 4.2, in the direction of the clock, respectively from up and down and eight neighbours of left and right both direction search described search starting point
Domain；
Step 4.3, when in the eight neighborhood in described search starting point search for less than binary value be 1 pixel when, by institute
State search starting point rejecting；
Step 4.4, when searched in the eight neighborhood in described search starting point binary value be 1 pixel, searched described
The pixel that the binary value that rope arrives is 1 searches for the eight neighborhood of described search starting point as search starting point, obtains new search and rises
Point, until returning to initial search starting point；
Step 4.5, whole search starting points form the boundary of link area, obtain the first link area.
Step 5: merging neighbouring first link area, the second link area is formed, area is removed and is not equal to first
Second link area of area threshold value, obtains third link area；
The first area threshold value of the present embodiment can determine by transmission tower height and SAR image resolution, example
Such as, the first area threshold value is equal to transmission tower length × SAR image resolution.Optionally, the height of transmission tower
About 50 meters of degree, image resolution is 3 meters, and the first threshold value may be configured as 150 square metres.
Step 6: calculating the geometric center of the third link area, as the central point of the third link area, take
It around the central point and include the rectangular area of the third link area as interested region (Region of
Interesting,ROI)；
The geometric center of the present embodiment third link area is (Cx, Cy), wherein
A indicates the area of the third link area:xiAnd yiRespectively indicate the third
For each pixel along distance to the coordinate value with orientation, N indicates the sum of pixel in the third link area in link area
Mesh, CxAnd CyThe third link area geometric center is respectively indicated in distance to the coordinate with orientation.
Step 7: as shown in Fig. 2, the small rectangle of grey be spatial sampling point, be referred to as pixel.The ROI is drawn
It is divided into L fan-shaped and K annulus, forms L × K sub-regions, the L fan-shaped central point with the third link area is top
Point, K annulus is using the central point of the third link area as the center of circle, wherein first annulus includes to be made of most pixel
Small rectangle, n-th annulus include the rectangle around the N-1 annulus, and k-th annulus includes entire ROI；In the present embodiment, L
Fan-shaped central angle can be set to 10 degree.
Step 8: the ROI is expressed as two-dimensional matrixWherein θkIndicate k-th of fan
The central angle of shape, ρlIndicate the radius of first of annulus, M (ρl,θk) indicate described k-th fan-shaped and first of annulus overlapping region
Remove the described k-th fan-shaped amplitude average value that subregion is formed by with the l-1 annulus overlapping region.
Step 9: along the annular radii construct polar coordinates semi-variance function, and successively with each annulus pair
Semivariance feature is extracted in the fan-shaped region answered:
Wherein h is the distance between two annulus, is fixed value, ζh(θ) and ζh(ρ) respectively indicates different annular and difference
Polar coordinates semi-variance function textural characteristics on direction, wherein from ζhTo find out in the formula of (θ), fan-shaped center of circle angle is constant,
Only annular radii changes；From ζhFind out in the formula of (ρ), annular radii is constant, and fan-shaped direction changes；
Step 10: being examined using polar coordinates semi-variance function textural characteristics as the input quantity of neural network using neural network
Survey transmission tower.
Step 110, using polar coordinates semi-variance function textural characteristics as the input quantity of neural network, examined using neural network
Survey transmission tower.
The image that the present embodiment extracts a part from SAR image obtained is used to train neural network, these images
In not only include true transmission tower target, but also be also required to comprising the false target as caused by the woods, house etc..
The present embodiment method further include: Step 11: judging the detection for the transmission tower that the neural network detects
Precision exports the transmission tower that the neural network detects if detection accuracy is greater than detection threshold value as a result；Such as
Fruit detection accuracy is not more than detection threshold value, increases the image quantity for training the neural network and/or increases the nerve
The number of plies of network detects transmission tower using neural network again, until detection accuracy is greater than detection threshold value.
Compared with prior art, expensive full-polarization SAR image is necessary for present invention does not require image, and for
Identical SAR image, the high-voltage power transmission tower profile especially on image is less clear and there are many false targets
In the case of, the present invention can detecte out more transmission tower targets, and it is lower than the prior art to detect false alarm rate.
It should be clear that the invention is not limited to specific configuration described above and shown in figure and processing.
For brevity, it is omitted here the detailed description to known method.In the above-described embodiments, several tools have been described and illustrated
The step of body, is as example.But method process of the invention is not limited to described and illustrated specific steps, this field
Technical staff can be variously modified, modification and addition after understanding spirit of the invention, or suitable between changing the step
Sequence.
It should be noted that the exemplary embodiment referred in the present invention, is described based on a series of step or device
Certain methods or system.But the present invention is not limited to the sequence of above-mentioned steps, that is to say, that can be according to mentioning in embodiment
And sequence execute step, may also be distinct from that the sequence in embodiment or several steps are performed simultaneously.
The above description is merely a specific embodiment, it should be appreciated that protection scope of the present invention is not limited to
This, anyone skilled in the art in the technical scope disclosed by the present invention, can readily occur in various equivalent
Modifications or substitutions, these modifications or substitutions should be covered by the protection scope of the present invention.
Claims (7)
1. a kind of method for detecting transmission tower automatically based on high resolution synthetic aperture radar image, it is characterised in that the side
Method includes:
Step 1: obtaining the High-resolution SAR Images in transmission line of electricity region and being pre-processed；
Step 2: removing the background clutter of the SAR image using global constant false alarm rate method, and generate the first binary system shadow
Picture；
Step 3: detecting the linear target in first binary image using Hough transform method, and in the binary system
Linear target is removed in image, generates the second binary image；
Step 4: obtaining the boundary of link area using eight neighborhood algorithm, the first link area is obtained；
Step 5: merging neighbouring first link area, the second link area is formed, area is removed and is not equal to the first area
Second link area of threshold value, obtains third link area；
Step 6: calculating the geometric center of the third link area, as the central point of the third link area, takes and surround
The central point and include the rectangular area of the third link area as interested region (Region of
Interesting,ROI)；
Step 7: the ROI is divided into L fan-shaped and K annulus, L × K sub-regions are formed, L sector is with the third
The central point of link area is vertex, and K annulus is using the central point of the third link area as the center of circle, wherein first circle
Ring includes the minimum rectangle being made of pixel, and n-th annulus includes the rectangle around the N-1 annulus, and k-th annulus includes whole
A ROI；
Step 8: the ROI is expressed as two-dimensional matrixWherein θkIndicate k-th of sector
Central angle, ρlIndicate the radius of first of annulus, M (ρl,θk) indicate that k-th of sector is removed with first of annulus overlapping region
The described k-th fan-shaped amplitude average value that subregion is formed by with the l-1 annulus overlapping region.
Step 9: constructing polar coordinates semi-variance function along the annular radii, and successively corresponding with each annulus
Semivariance feature is extracted in the fan-shaped region:
Wherein h is the distance between two annulus, is fixed value, ζh(θ) and ζh(ρ) respectively indicates different annular and different directions
On polar coordinates semi-variance function textural characteristics, wherein from ζhSee that 5 go out in the formula of (θ), fan-shaped center of circle angle is constant, only
Annular radii changes；From ζhFind out in the formula of (ρ), annular radii is constant, and fan-shaped direction changes；
Step 10: being detected using neural network defeated using polar coordinates semi-variance function textural characteristics as the input quantity of neural network
Electric pole tower.
2. the side according to claim 1 for detecting defeated 10 electric pole tower automatically based on high resolution synthetic aperture radar image
Method, which is characterized in that it is described that background clutter is removed using global constant false alarm rate method,
And generate the first binary image, comprising:
Using the background clutter of K statistical distribution model f (x) description SAR image；
False alarm rate P is setfa, pass throughCalculate detection threshold value T；
For each pixel, when the intensity value of pixel is greater than detection threshold value T, described in global constant false alarm rate method detector determines
Pixel is target, the pixel is designated as binary value 1, and retain the range value of the pixel, when the intensity value of pixel is little
In detection threshold value T, then the pixel is determined for clutter, the pixel is designated as binary value 0, and by the range value of the pixel
It is set as 0, obtains the first binary image.
3. the method according to claim 1 that transmission tower is detected based on high resolution synthetic aperture radar image automatically,
It is characterized in that, the boundary for obtaining link area using eight neighborhood algorithm, obtains the first link area, comprising:
From second binary image, the pixel that binary value is 1 is obtained, as search starting point；
In the direction of the clock, respectively from the eight neighborhood of upper and lower and left and right both direction search described search starting point；
When search is less than the pixel that binary value is 1 in the eight neighborhood in described search starting point, described search starting point is picked
It removes；
When searched in the eight neighborhood in described search starting point binary value be 1 pixel, the binary system that described search is arrived
The pixel that value is 1 searches for the eight neighborhood of described search starting point, new search starting point is obtained, until returning to as search starting point
Initial search starting point；
Whole search starting points form the boundary of link area, obtain the first link area.
4. the method according to claim 1 that transmission tower is detected based on high resolution synthetic aperture radar image automatically,
It is characterized in that, the first area threshold value=transmission tower length ×=SAR image resolution.
5. the method according to claim 1 that transmission tower is detected based on high resolution synthetic aperture radar image automatically,
It is characterized in that, the geometric center of the third link area is (Cx, Cy), wherein
A indicates the area of the third link area:xiAnd yiRespectively indicate the third link
For each pixel along distance to the coordinate value with orientation, N indicates the total number of pixel in the third link area, C in regionx
And CyThe third link area geometric center is respectively indicated in distance to the coordinate with orientation.
6. the method according to claim 1 that transmission tower is detected based on high resolution synthetic aperture radar image automatically,
It is characterized in that, described L fan-shaped central angle is 10 degree.
7. the method according to claim 1 that transmission tower is detected based on high resolution synthetic aperture radar image automatically,
Characterized by further comprising:
Judge the detection accuracy for the transmission tower that the neural network detects, it, will if detection accuracy is greater than detection threshold value
The transmission tower that the neural network detects exports as a result；If detection accuracy is not more than detection threshold value, increases and use
In the image quantity of the training neural network and/or the number of plies of the increase neural network, detected again using neural network
Transmission tower, until detection accuracy is greater than detection threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811167126.2A CN109376623B (en) | 2018-10-08 | 2018-10-08 | Method for detecting transmission tower based on synthetic aperture radar image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811167126.2A CN109376623B (en) | 2018-10-08 | 2018-10-08 | Method for detecting transmission tower based on synthetic aperture radar image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109376623A true CN109376623A (en) | 2019-02-22 |
| CN109376623B CN109376623B (en) | 2021-12-21 |
Family
ID=65403700
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811167126.2A Active CN109376623B (en) | 2018-10-08 | 2018-10-08 | Method for detecting transmission tower based on synthetic aperture radar image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109376623B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112990071A (en) * | 2021-03-31 | 2021-06-18 | 香港中文大学（深圳） | Power transmission line anomaly detection method and device and computer readable storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105976376A (en) * | 2016-05-09 | 2016-09-28 | 电子科技大学 | High resolution SAR image target detection method based on part model |
| CN108256429A (en) * | 2017-12-19 | 2018-07-06 | 国网山西省电力公司阳泉供电公司 | A kind of transmission tower object detection method using high spatial resolution single polarization SAR image |
| CN108549080A (en) * | 2018-02-28 | 2018-09-18 | 中国电力科学研究院有限公司 | A kind of transmission tower position extracting method and system |
-
2018
- 2018-10-08 CN CN201811167126.2A patent/CN109376623B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105976376A (en) * | 2016-05-09 | 2016-09-28 | 电子科技大学 | High resolution SAR image target detection method based on part model |
| CN108256429A (en) * | 2017-12-19 | 2018-07-06 | 国网山西省电力公司阳泉供电公司 | A kind of transmission tower object detection method using high spatial resolution single polarization SAR image |
| CN108549080A (en) * | 2018-02-28 | 2018-09-18 | 中国电力科学研究院有限公司 | A kind of transmission tower position extracting method and system |
Non-Patent Citations (3)
| Title |
|---|
| 何宇婷: ""半方差函数纹理提取在遥感图像分类中的应用"", 《遥感技术与应用》 * |
| 刘艳: ""高分辨率SAR卫星监测特高压输电铁塔形变"", 《高电压技术》 * |
| 周筑博: ""基于深度卷积神经网络的输电线路可见光图像目标检测"", 《液晶与显示》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112990071A (en) * | 2021-03-31 | 2021-06-18 | 香港中文大学（深圳） | Power transmission line anomaly detection method and device and computer readable storage medium |
| CN112990071B (en) * | 2021-03-31 | 2024-01-05 | 香港中文大学（深圳） | Power transmission line abnormality detection method and device and computer readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109376623B (en) | 2021-12-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Guan et al. | Extraction of power-transmission lines from vehicle-borne lidar data | |
| Wang et al. | The role of spatial heterogeneity in detecting urban slums | |
| CN114419825B (en) | High-speed rail perimeter intrusion monitoring device and method based on millimeter wave radar and camera | |
| CN106356757A (en) | Method for inspecting electric power lines by aid of unmanned aerial vehicle on basis of human vision characteristics | |
| Kim et al. | Detection of the Hebei Spirit oil spill on SAR imagery and its temporal evolution in a coastal region of the Yellow Sea | |
| Lehtomäki et al. | Power line mapping technique using all-terrain mobile laser scanning | |
| CN111191673B (en) | Ground surface temperature downscaling method and system | |
| CN106156758B (en) | A kind of tidal saltmarsh method in SAR seashore image | |
| CN109254287A (en) | Method for detecting power transmission conductor based on synthetic aperture radar image | |
| CN106950574B (en) | The remote sensing measuring method and device of gray haze total amount in a kind of atmosphere | |
| Yan et al. | Inland water mapping based on GA-LinkNet from CyGNSS data | |
| CN103400137A (en) | Method for extracting geometrical building parameters of synthetic aperture radar (SAR) image | |
| CN103996027A (en) | Space-based space target recognizing method | |
| Liu et al. | Using SPOT 5 fusion-ready imagery to detect Chinese tamarisk (saltcedar) with mathematical morphological method | |
| Hu et al. | Dynamic monitoring of land subsidence in mining area from multi-source remote-sensing data–a case study at Yanzhou, China | |
| CN118053087B (en) | Lin Caoguang V complementary evaluation and monitoring method based on unmanned aerial vehicle and model analysis | |
| Hon et al. | Application of short‐range LIDAR in early alerting for low‐level windshear and turbulence at Hong Kong International Airport | |
| Haroun et al. | A review of vegetation encroachment detection in power transmission lines using optical sensing satellite imagery | |
| CN109376623A (en) | Method for detecting transmission tower based on synthetic aperture radar image | |
| Li et al. | Probabilistic Convective Initiation Nowcasting Using Himawari-8 AHI with Explainable Deep Learning Models | |
| Mandroux et al. | Wind turbine detection on sentinel-2 images | |
| CN114239379A (en) | Transmission line geological disaster analysis method and system based on deformation detection | |
| Colin et al. | Segmentation of rainfall regimes by machine learning on a colocalized Nexrad/Sentinel-1 dataset | |
| Ahmed et al. | Partice swarm optimization based fusion of MODIS and PALSAR images for hotspot detection | |
| Wang et al. | Modification of CFAR algorithm for oil spill detection from SAR data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |