CN114359143A - Remote sensing image road extraction method - Google Patents
Remote sensing image road extraction method Download PDFInfo
- Publication number
- CN114359143A CN114359143A CN202111455442.1A CN202111455442A CN114359143A CN 114359143 A CN114359143 A CN 114359143A CN 202111455442 A CN202111455442 A CN 202111455442A CN 114359143 A CN114359143 A CN 114359143A
- Authority
- CN
- China
- Prior art keywords
- road
- shadow
- image
- area
- region
- 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
- 238000000605 extraction Methods 0.000 title claims abstract description 39
- 238000000034 method Methods 0.000 claims description 22
- 238000012545 processing Methods 0.000 claims description 20
- 238000001914 filtration Methods 0.000 claims description 19
- 230000000877 morphologic effect Effects 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims description 9
- 230000011218 segmentation Effects 0.000 claims description 7
- 230000000694 effects Effects 0.000 claims description 6
- 238000009499 grossing Methods 0.000 claims description 5
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 230000035772 mutation Effects 0.000 claims description 4
- 238000005429 filling process Methods 0.000 claims description 3
- 238000003709 image segmentation Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000010521 absorption reaction Methods 0.000 description 2
- 238000013075 data extraction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000029553 photosynthesis Effects 0.000 description 2
- 238000010672 photosynthesis Methods 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4053—Scaling of whole images or parts thereof, e.g. expanding or contracting based on super-resolution, i.e. the output image resolution being higher than the sensor resolution
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a remote sensing image road extraction method with shadow interference, aiming at the condition that the conventional extraction method is interrupted due to the fact that a road section in a high-resolution remote sensing image is shielded by the shadows of ground objects on two sides of the road section, firstly, a road shadow area in the image is detected, the shadow area is compensated by utilizing the gray information of a non-shielded road section area, and then, the compensated road is extracted.
Description
Technical Field
The invention belongs to the technical field of optical image processing, and particularly relates to a remote sensing image road extraction method which is mainly applied to an airborne photoelectric pod.
Background
In the high-resolution remote sensing image, trees and buildings are often arranged at two sides of a road, due to the sun angle, shadows generated by such ground objects are projected on the road to cause uneven road surface gray scale, shadow areas in the image are darker than surrounding ground objects due to lack of illumination, meanwhile, the spectral characteristics of original ground objects existing in the shadow coverage area are not completely covered by the shadow effect, and therefore the original ground object characteristics are still kept to a certain extent.
By combining the gray features of the shadow areas, the shadow can be detected and extracted in the image, the gray features of the original ground object covered by the shadow areas are recovered to a certain extent, and the detection of the ground object is realized.
The airborne photoelectric pod integrates optical, mechanical, automatic control and communication technologies, is important search and reconnaissance equipment in the field of aerospace, and is often loaded with optical sensors such as visible light and near infrared, so that the research on the target extraction technology with shadow interference applied to the airborne photoelectric pod is of great significance.
Disclosure of Invention
The invention provides a remote sensing image road extraction method with shadow interference aiming at the background.
The technical scheme adopted by the invention for solving the technical problems is as follows: a remote sensing image road extraction method comprises the following steps:
(1) color space conversion: firstly, converting RGB to HSV color space by combining the characteristics of low brightness, high tone and high saturation embodied by a road shadow region in a high-resolution remote sensing image;
(2) primarily detecting a shadow area;
(3) shadow area filtering: after the shadow is preliminarily detected, generating a series of indexes by combining with road characteristics, further filtering the preliminarily extracted shadow area according to the indexes, removing the shadow of the non-road area, and keeping the shadow of the road section;
(4) road shadow area compensation: after the road shadow detection is finished, compensating the shadow region by using a Wallis filter, then smoothing the primary compensation result by using median filtering, solving the problem of edge region gray value mutation caused by light, road material and the like, extracting the primary extracted road by using a region growing method, and then performing morphological processing to improve the extraction effect;
(5) road extraction: and finally, extracting the mesh road route by using a morphological skeleton algorithm, and optimizing the problems of burrs, interruption and the like in the extraction result to obtain a complete road skeleton line.
The remote sensing image road extraction method comprises the steps of (1) converting the color space of an original image into HSV (hue, saturation, value) and then splitting the converted image into H, S, V three separate wave bands.
The remote sensing image road extraction method comprises the step (2) of respectively solving respective segmentation threshold values delta of the three components H, S, V by an otsu methodh、δs、δvAnd respectively carrying out image segmentation by using segmentation threshold values, and solving intersection of the results by the following formula:
f=fh∩fs∩fv
the result is then superimposed on the color image and the extracted result is displayed in blue.
The remote sensing image road extraction method comprises the steps of (3) processing adjacent planar shadow areas of the shadow areas by taking blocks as units, counting the area S of each shadow block, and setting an area threshold value deltasRemoving the block shadow below the threshold value; counting the standard deviation of each shadow area, and setting a standard deviation threshold deltavarRemoving the areas which do not meet the conditions;
r, G, B three wave band images are mutually differed to generate a difference image, and different points embodied by the blue-purple tree and the real road shadow are analyzed and compared in the difference image; for the shadow regions which are reserved after being filtered, the integral gray value mean value mu of all the regions in the blue-red band difference image is calculatedaveAnd the average value mu of the gray scale corresponding to each shadow areaiIn μaveAs a threshold value according to which the judgment is made, all shadow regions S satisfying that the average value of the pixel gray levels of the single block regions is larger than the average value of the pixel gray levels of all the regions as a whole are reservediRemoving the area which does not meet the condition, and obtaining the final road shadow result S by the following formula:
S={Si|μi≥μave}
further improved extraction results are obtained by using a hole filling process.
The remote sensing image road extraction method comprises the step (4) of performing compensation processing on a finally detected shadow region by using a Wallis filter, wherein the Wallis filter is represented by the following form:
wherein g (x, y) represents the original image, f represents the reference region information, sfAs reference area standard deviation, mfIs the mean value of the gray levels of the reference region, sgAnd mgThe standard deviation and the mean value of the gray levels, g, corresponding to the region to be enhancedc(x, y) is the gray value of the image after transformation, c is the expansion coefficient of variance, b is the expansion coefficient of brightness, and the value ranges of the two are [0,1];
Selecting partial ideal road seed points in the image, counting 3 multiplied by 3 or 5 multiplied by 5 neighborhood information of the seed points as a reference area, and assigning m with neighborhood mean and standard deviationfAnd sfMeanwhile, counting the finally obtained shadow detection result, calculating the gray level mean value and standard deviation corresponding to each single area, and assigning a value to mgAnd sgAnd performing Wallis filtering processing by taking each region as a unit to obtain a preliminary compensation result, and performing smoothing processing on the edge of the compensation region by using median filtering of a 3 multiplied by 3 size template.
The method for extracting the remote sensing image road comprises the steps of (5) extracting the road by adopting a region growing method, selecting seed points in a road section, comparing the seed points with the gray values of the seed points to see whether threshold conditions are met, further judging whether pixels are the road points, and realizing the road searching and distinguishing of the whole image.
The method for extracting the remote sensing image road comprises the steps of processing an extraction result by using a morphological skeleton algorithm, removing burrs in an extracted skeleton line, simultaneously carrying out connection processing on a broken part in the skeleton line, finally obtaining a complete road skeleton line, and superposing the complete road skeleton line on an original image to extract a mesh road route.
The invention has the beneficial effects that: aiming at the problem that part of trees and shadows are not easy to distinguish, the method utilizes the characteristic that the correlation of blue-red light and blue-green light is not large to obtain the mutual difference image of three wave bands, and eliminates the trees extracted by mistake according to the gray threshold; according to the method, Wallis filtering is combined with median filtering to process the shadow area, so that the problem of edge area gray value mutation caused by light, road material and the like is solved, and extraction and compensation of road shadow are realized; the invention adopts morphological processing to optimize the extraction result, better inhibits noise interference and has more stable extraction result.
Drawings
FIG. 1 shows the image results of each channel in HSV space according to the present invention;
FIG. 2 is a diagram of the shadow region extracted primarily by the present invention;
FIG. 3 is a diagram illustrating the preliminary filtering result of the shadow area according to the present invention;
FIG. 4 shows the difference between RGB bands according to the present invention;
FIG. 5 shows the optimization results of hole filling according to the present invention;
FIG. 6 shows the road compensation result of the present invention;
FIG. 7 shows the compensated road extraction result of the present invention;
FIG. 8 shows the road skeleton line extracted by the present invention;
FIG. 9 shows the second set of road data extraction results according to the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
The invention discloses a remote sensing image road extraction method, which combines the low brightness, high tone and high saturation characteristics embodied by a shadow region, firstly carries out RGB-HSV color space conversion, preliminarily detects the shadow, then combines road characteristics to generate a series of indexes, further filters the preliminarily extracted shadow region according to the indexes, removes the shadow of a non-road region, and reserves the shadow of a road section. And after the road shadow detection is finished, performing compensation processing on the shadow area by using a Wallis filter. And then, smoothing the primary compensation result by adopting median filtering, and solving the problem of edge region gray value mutation caused by light, road material and the like. And extracting the primary extraction road by adopting a region growing method, and then performing morphological treatment to improve the extraction effect. And finally, extracting the mesh road route by using a morphological skeleton algorithm, and optimizing the problems of burrs, interruption and the like in the extraction result to obtain a complete road skeleton line.
(1) And (4) color space conversion.
Firstly, the color space of the original image is converted into HSV, and then the converted image is divided into H, S, V three separate bands, and the display result is shown in fig. 1, wherein (a) is the original color image, (b) is the H component image, (c) is the S component image, and (d) is the V component image.
(2) And (5) primarily detecting a shadow area.
Because the image histogram with shadow area generally presents the characteristic of double peak, and the shadow area is the brighter area on H component and S component, and appears darker on V component correspondingly, so the threshold segmentation is carried out on three components, and the respective segmentation threshold value delta is obtained by otsu method for H, S, V three components respectivelyh、δs、δvThe three components are respectively subjected to image segmentation by using segmentation threshold values, the intersection of the results is obtained as shown in the following formula, then the results are superposed on the color image, and the extracted results are displayed in blue as shown in fig. 2:
f=fh∩fs∩fv
and 3, filtering out shadow areas.
After the initial extraction, the shadow areas mostly exist in the unit of pixels, the adjacent planar shadow areas are processed in the unit of blocks, the area S of each shadow block is counted, the extraction result is observed, the fact that the extraction area with small partial area exists in the image and comprises correct shadows and wrongly extracted ground objects is found, and the area is manually setThreshold deltasBlock shadows below the threshold are removed. Counting the standard deviation of each shadow area, and setting a standard deviation threshold deltavarThe areas not meeting the condition are removed, and the shadow result obtained after the preliminary filtering is shown in fig. 3.
Comparing the preliminary removal result with the original image can find that the trees still cannot be completely removed well after the trees are filtered by taking the area and the variance as indexes due to the existence of partial blue-purple trees. As can be seen from one of the above-mentioned shadow region characteristics, the shadow region has a small correlation between blue-red light and blue-green light, and due to the effect of plant photosynthesis, there are different degrees of absorption and reflection for three lights, so that an attempt is made to generate a differential image by mutually differencing R, G, B three band images, and the differences between the blue-purple tree and the real road shadow are analyzed and compared in the differential image, as shown in fig. 4, where (a) is an original image, (b) is a green-red band differential image, (c) is a green-blue band differential image, and (d) is a blue-red band differential image.
Comparing the three differential band images, it can be seen that in fig. 4(b) and (c), the real shadow and the tree in the corresponding reserved shadow region still cannot be well distinguished, while in fig. 4(d), due to the absorption of red and blue light by tree photosynthesis and the advantage of blue light in scattering, more blue light is reserved than red light, so that the spectral value of the tree region in the real shadow is brighter than that in the shadow, so that the tree region can be distinguished. For the shadow regions which are reserved after being filtered, the integral gray value mean value mu of all the regions in the blue-red band difference image is calculatedaveAnd the average value mu of the gray scale corresponding to each shadow areaiIn μaveAs a threshold value according to which the judgment is made, all shadow regions S satisfying that the average value of the pixel gray levels of the single block regions is larger than the average value of the pixel gray levels of all the regions as a whole are reservediRemoving the area which does not meet the condition to obtain the final road shadow result S: s ═ Si|μi≥μave}。
Due to the influence of the local individual noise, some holes still exist in the final detection result and are not processed, and in order to reduce the interference to the subsequent processing, a hole filling process is adopted, and the further improved extraction result is shown in fig. 5.
(4) And compensating the road shadow area.
After the road shadow detection is completed, in order to reduce the influence of the shadow and realize the final complete road segment extraction, the shadow compensation is needed.
And (3) performing compensation processing on the finally detected shadow region by using a Wallis filter, wherein the representation form of the commonly used Wallis filter is as follows:
wherein g (x, y) represents the original image, f represents the reference region information, sfAs reference area standard deviation, mfIs the mean value of the gray levels of the reference region, sgAnd mgThe standard deviation and the mean value of the gray levels, g, corresponding to the region to be enhancedcAnd (x, y) is the gray value of the image after transformation. c is the expansion coefficient of the variance, b is the brightness expansion coefficient, and the value ranges of the two are [0, 1%]。
In order to simplify the operation, assuming that the gray values of all road sections of the road image are relatively close, manually selecting partial ideal road seed points in the image, counting 3 × 3 or 5 × 5 neighborhood information of the seed points as a reference region, and assigning m with a neighborhood mean value and a standard deviationfAnd sf. Meanwhile, counting the finally obtained shadow detection result, calculating the gray level mean value and standard deviation corresponding to each single area, and assigning a value to mgAnd sgThe Wallis filtering process is performed for each region as a unit, and the obtained preliminary compensation result is smoothed for the edge of the compensation region by using the median filtering of a 3 × 3 template, and the result is shown in FIG. 6.
(5) And (6) extracting the road.
After the shadow compensation is completed, the extraction of the road is performed, the road is extracted by adopting a region growing method, the seed points are manually selected from the road section, whether the threshold condition is met is judged according to the comparison with the gray value of the seed points, whether the pixels are the road points is further judged, the road searching and distinguishing of the whole image are realized, the result obtained by the primary processing is shown in fig. 7(a), and due to the images of noise and the like, the road section obtained by the primary region growing has the phenomena of default, holes and the like, so that the further morphological processing is performed to improve the extraction result, as shown in fig. 7 (b).
In order to further extract the mesh road route, the extraction result is processed by using a morphological skeleton algorithm, then burrs in the extracted skeleton line are removed, meanwhile, the connection processing is required for the interruption position in the skeleton line, finally, a complete road skeleton line is obtained, and the complete road skeleton line is superimposed on the original image to display the result as shown in fig. 8.
The second set of road data extraction results of the present invention are shown in fig. 9, where (a) is the original road image, (b) is the preliminarily extracted shadow region, (c) is the optimized shadow region, (d) is the Wallis filtering result, (e) is the corresponding gray image, and (f) is the extracted skeleton line.
The above-described embodiments are merely illustrative of the principles and effects of the present invention, and some embodiments may be applied, and it will be apparent to those skilled in the art that various changes and modifications may be made without departing from the inventive concept of the present invention, and these embodiments are within the scope of the present invention.
Claims (7)
1. A remote sensing image road extraction method is characterized in that: comprises the following steps
(1) Color space conversion: firstly, converting RGB to HSV color space by combining the characteristics of low brightness, high tone and high saturation embodied by a road shadow region in a high-resolution remote sensing image;
(2) primarily detecting a shadow area;
(3) shadow area filtering: generating a series of indexes by combining road characteristics, further filtering the preliminarily extracted shadow area, removing the shadow of the non-road area, and keeping the shadow of the road section;
(4) road shadow area compensation: after the road shadow detection is finished, compensating the shadow region by using a Wallis filter, then smoothing the primary compensation result by using median filtering, solving the problem of edge region gray value mutation caused by light and road material, extracting the primary extracted road by using a region growing method, and then performing morphological treatment to improve the extraction effect;
(5) road extraction: and finally, extracting the mesh road route by using a morphological skeleton algorithm, and optimizing the problems of burrs and interruption in the extraction result to obtain a complete road skeleton line.
2. The method for extracting a remote sensing image road according to claim 1, wherein the step (1) is that the color space of the original image is converted into HSV, and then the converted image is divided into H, S, V three separate wave bands.
3. The method for extracting a remote sensing image road as claimed in claim 2, wherein in the step (2), H, S, V three components are respectively subjected to otsu method to obtain respective segmentation threshold values δh、δs、δvAnd respectively carrying out image segmentation by using segmentation threshold values, and solving intersection of the results by the following formula:
f=fh∩fs∩fv
the result is then superimposed on the color image and the extracted result is displayed in blue.
4. The method for extracting a remote sensing image road as claimed in claim 3, wherein the step (3) is to process the adjacent planar shadow areas of the shadow areas in units of blocks, count the area S of each shadow block, and set an area threshold δsRemoving the block shadow below the threshold value; counting the standard deviation of each shadow area, and setting a standard deviation threshold deltavarRemoving the areas which do not meet the conditions;
by R, G, B IIIThe seed wave band images are mutually differenced to generate a difference image, and different points embodied by the blue-purple trees and the real road shadow are analyzed and compared in the difference image; calculating the integral gray value mean value mu of all the areas in the blue-red band difference imageaveAnd the average value mu of the gray scale corresponding to each shadow areaiIn μaveAs a threshold value according to which the judgment is made, all shadow regions S satisfying that the average value of the pixel gray levels of the single block regions is larger than the average value of the pixel gray levels of all the regions as a whole are reservediRemoving the area which does not meet the condition, and obtaining the final road shadow result S through a formula: s ═ Si|μi≥μave}; further improved extraction results are obtained by using a hole filling process.
5. The method for extracting a remote sensing image road as claimed in claim 4, wherein the step (4) is to perform compensation processing on the finally detected shadow region by using a Wallis filter, and the Wallis filter is expressed as:wherein g (x, y) represents the original image, f represents the reference region information, sfAs reference area standard deviation, mfIs the mean value of the gray levels of the reference region, sgAnd mgThe standard deviation and the mean value of the gray levels, g, corresponding to the region to be enhancedc(x, y) is the gray value of the image after transformation, c is the expansion coefficient of variance, b is the expansion coefficient of brightness, and the value ranges of the two are [0,1];
Selecting partial ideal road seed points in the image, counting 3 multiplied by 3 or 5 multiplied by 5 neighborhood information of the seed points as a reference area, and assigning m with neighborhood mean and standard deviationfAnd sfMeanwhile, counting the finally obtained shadow detection result, calculating the gray level mean value and standard deviation corresponding to each single area, and assigning a value to mgAnd sgAnd performing Wallis filtering processing by taking each region as a unit to obtain a preliminary compensation result, and performing smoothing processing on the edge of the compensation region by using median filtering of a 3 multiplied by 3 size template.
6. The method for extracting the remote sensing image road according to claim 5, wherein in the step (5), the road is extracted by adopting a region growing method, the seed points are selected from the road section, and whether the threshold condition is met or not is judged according to the comparison with the gray value of the seed points, so that the road searching and distinguishing of the whole image are realized.
7. The method for extracting the remote sensing image road according to claim 6, characterized in that a morphological skeleton algorithm is used for processing an extraction result, then burrs in the extracted skeleton line are removed, meanwhile, connection processing is needed for the break in the skeleton line, finally, a complete road skeleton line is obtained and is superposed on an original image to extract a mesh road route.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111455442.1A CN114359143A (en) | 2021-12-01 | 2021-12-01 | Remote sensing image road extraction method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111455442.1A CN114359143A (en) | 2021-12-01 | 2021-12-01 | Remote sensing image road extraction method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114359143A true CN114359143A (en) | 2022-04-15 |
Family
ID=81097041
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111455442.1A Pending CN114359143A (en) | 2021-12-01 | 2021-12-01 | Remote sensing image road extraction method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114359143A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114998412A (en) * | 2022-08-01 | 2022-09-02 | 北京中科慧眼科技有限公司 | Shadow region parallax calculation method and system based on depth network and binocular vision |
CN115641512A (en) * | 2022-12-26 | 2023-01-24 | 成都国星宇航科技股份有限公司 | Satellite remote sensing image road identification method, device, equipment and medium |
CN115995046A (en) * | 2022-11-18 | 2023-04-21 | 北京市农林科学院信息技术研究中心 | Rural road remote sensing extraction method and device under shadow shielding state |
-
2021
- 2021-12-01 CN CN202111455442.1A patent/CN114359143A/en active Pending
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114998412A (en) * | 2022-08-01 | 2022-09-02 | 北京中科慧眼科技有限公司 | Shadow region parallax calculation method and system based on depth network and binocular vision |
CN115995046A (en) * | 2022-11-18 | 2023-04-21 | 北京市农林科学院信息技术研究中心 | Rural road remote sensing extraction method and device under shadow shielding state |
CN115995046B (en) * | 2022-11-18 | 2023-08-04 | 北京市农林科学院信息技术研究中心 | Rural road remote sensing extraction method and device under shadow shielding state |
CN115641512A (en) * | 2022-12-26 | 2023-01-24 | 成都国星宇航科技股份有限公司 | Satellite remote sensing image road identification method, device, equipment and medium |
CN115641512B (en) * | 2022-12-26 | 2023-04-07 | 成都国星宇航科技股份有限公司 | Satellite remote sensing image road identification method, device, equipment and medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114359143A (en) | Remote sensing image road extraction method | |
CN107103591B (en) | Single image defogging method based on image haze concentration estimation | |
CN109191432B (en) | Remote sensing image cloud detection method based on domain transformation filtering multi-scale decomposition | |
CN1623171A (en) | Method for producing cloud free and cloud-shadow free images | |
CN105913421B (en) | Remote sensing images cloud detection method of optic based on adaptive shape dark | |
CN102509262B (en) | Method for removing thin cloud of remote sensing image | |
CN112307901B (en) | SAR and optical image fusion method and system for landslide detection | |
CN106096491B (en) | Automatic identification method for microaneurysms in fundus color photographic image | |
CN110687121B (en) | Intelligent online detection and automatic grading method and system for ceramic tiles | |
CN104217440B (en) | A kind of method extracting built-up areas from remote sensing images | |
WO2019076326A1 (en) | Shadow detection method and system for surveillance video image, and shadow removing method | |
CN108711160B (en) | Target segmentation method based on HSI (high speed input/output) enhanced model | |
CN110175556B (en) | Remote sensing image cloud detection method based on Sobel operator | |
CN110930321A (en) | Blue/green screen digital image matting method capable of automatically selecting target area | |
CN109003238B (en) | Image haze removal method based on model, histogram and gray level enhancement | |
CN112465708A (en) | Improved image defogging method based on dark channel | |
CN108280810A (en) | Automatic processing method for repairing cloud coverage area of single-time phase optical remote sensing image | |
CN114881869A (en) | Inspection video image preprocessing method | |
CN102855627A (en) | City remote sensing image shadow detection method based on spectral characteristic and topological relation | |
CN113744326A (en) | Fire detection method based on seed region growth rule in YCRCB color space | |
CN111192332B (en) | Smoke detection-based smoke machine control method and smoke machine | |
CN106339995A (en) | Space-time multiple feature based vehicle shadow eliminating method | |
CN111931824B (en) | Coal rock identification method based on drilling slag return image | |
CN102982512B (en) | Image shadow detection method for Baidu satellite map | |
CN114119634A (en) | Automatic building extraction method and system combining vegetation elimination and image feature consistency constraint |
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 |