CN104063870A - Automatic land and sea template segmentation method based on scanning line detection and application thereof - Google Patents
Automatic land and sea template segmentation method based on scanning line detection and application thereof Download PDFInfo
- Publication number
- CN104063870A CN104063870A CN201410317380.1A CN201410317380A CN104063870A CN 104063870 A CN104063870 A CN 104063870A CN 201410317380 A CN201410317380 A CN 201410317380A CN 104063870 A CN104063870 A CN 104063870A
- Authority
- CN
- China
- Prior art keywords
- sea
- land
- region
- template
- gray
- 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
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to an automatic land and sea template segmentation method based on scanning line detection and the application of the automatic land and sea template segmentation method. The automatic land and sea template segmentation method comprises the steps that (1) a high-resolution visible light port remote-sensing image to be detected is input and is re-sampled; (2) sea surface seed points are detected for the re-sampled port remote-sensing image; (3) among the obtained sea surface area seed points, all points with the gray values meeting following conditions in the adjacent area are searched for, wherein the conditions include that the difference between the pixel gray of the point and the gray of the adjacent seed point is not larger than 2, and the difference between the pixel gray of the point and the gray of the initial seed point is not larger than 8; the pixel points meeting the conditions are the sear surface area, and the other area is set as a land area; (4) the sea area and the land area are binarized; (5) morphological processing is carried out to obtain a land and sea segmentation template after the port remote-sensing image is re-sampled; (6) interpolation is carried out on the re-sampled land and sea segmentation template, and a land and sea template as big as the original remote-sensing image is obtained. The automatic land and sea template segmentation method based on scanning line detection and the application of the automatic land and sea template segmentation method can be widely applied to a sea surface target screening process of various civil and military fields.
Description
Technical field
The present invention relates to a kind of image detecting method and application thereof, particularly about a kind of extra large land template automatic division method and application thereof of detecting based on sweep trace that is applicable to high definition remote sensing images.
Background technology
The existing sea-surface target detection method based on extra large land template auto Segmentation, in the time realizing the separation of extra large land, generally all adopts based on Threshold segmentation or the method based on Texture Segmentation.In the time processing the abundant remote sensing images of high-resolution, background complexity, details, the extra large land template automatic division method based on Threshold segmentation exists that extra large land boundary alignment precision is low, the problems such as hole easily appear in sea and land area; And extra large land template auto Segmentation speed in the time of texture feature extraction based on Texture Segmentation is very slow, the problem that also have that positioning precision is poor simultaneously, hole easily appears in sea and land area.Therefore for high-resolution, extra large land background complexity, remote sensing images that details is abundant, existingly often can not obtain desirable result based on Threshold segmentation and the extra large land template dividing method based on Texture Segmentation, existing detection method can not realize the detection of the sea-surface target of high definition remote sensing images under the background of Complex Sea land quickly and accurately simultaneously.
Summary of the invention
For the problems referred to above, the object of this invention is to provide a kind of extraction rate fast, be applicable to the extra large land template automatic division method and the application thereof that detect based on sweep trace of high definition remote sensing images.
For achieving the above object, the present invention takes following technical scheme: a kind of extra large land template automatic division method detecting based on sweep trace, it comprises the following steps: 1) input High Resolution Visible Light harbour remote sensing images to be detected, according to based on the relevant resampling disposal route of elemental area, carry out resampling processing; 2) for resampling harbour after treatment remote sensing images, adopt region, the sea Seed Points detection method based on sweep trace, detect region, sea Seed Points; 3) according to step 2) region, the sea initial seed point that obtains, utilize region growing algorithm to search the point that all gray-scale values meet the following conditions in adjacent domain: this pixel grey scale with adjoin Seed Points gray scale difference value in 2 and with initial seed point gray scale difference value in 8, the pixel that mark meets this condition is region, sea, and other region is set to land area; 4) to step 3) region, extra large land that obtains carries out binaryzation, is set to 255 by land area gray-scale value, and sea area grayscale codomain is set to 0; 5) binary conversion treatment result is carried out to morphology processing, i.e. once corrosion operation and an expansive working, obtains harbour remote sensing images and adopts resampling Hai Lu after treatment to cut apart template; 6) to by step 5) the resampling Hai Lu after treatment that obtains cuts apart template, and according to based on the relevant resampling disposal route interpolation of elemental area, obtain the Hai Lu onesize with former remote sensing images and cut apart template.
Described step 2) middle region, the sea Seed Points that detects, comprise the following steps: the resampling harbour after treatment remote sensing images of 1. lining by line scan from top to bottom, first scan the first row; 2. the number of inactive pixels point in this row of judgement scanning: if the number of inactive pixels point is more than or equal to N/10, this row is decided to be to inactive line, 2. scanning next line, get back to step, and wherein N is this row pixel count; If the number of inactive pixels point is less than N/10, this row gray scale is done to difference after by forward direction, the difference result of each pixel is designated as difference value, and the difference value of every last pixel of row is set to 0; 3. investigate difference result, in the result of difference, if difference value is less than 2, be set to 0; 4. judge difference value is whether 0 contiguous pixels number exceedes N/5: if exceeded, think and occur continuous flat site, be region, sea, get pixel now as region, sea initial seed point, enter described step 3); Otherwise 2. scanning next line, get back to step, until complete the scanning of last column.
The application of the above-mentioned extra large land template automatic division method detecting based on sweep trace in the sea-surface target screening of military field.
The application of the above-mentioned extra large land template automatic division method detecting based on sweep trace in the sea-surface target screening of civil area.
The present invention is owing to taking above technical scheme, it has the following advantages: 1, the present invention proposes a kind of extra large land template automatic division method detecting based on sweep trace, it is not only applicable to the rapid extraction of the high definition remote sensing images under the background of Complex Sea land, and has evaded satellite image and proofreaied and correct the problem of bringing inactive pixels.2, the present invention has applied the detection method of region, the sea detection Seed Points of high definition remote sensing images under the background of Complex Sea land, the method in region, calmodulin binding domain CaM growing method mark sea, realize the method for extra large land template auto Segmentation by binaryzation and morphology processing, process the method for interpolation etc. by resampling, realized and quick and precisely obtained the goal of the invention that the Hai Lu onesize with former remote sensing images cut apart template.The present invention can be widely used in the template separation of extra large land and object filtering process in the remote sensing images of high definition harbour, various civilian and militaries field.
Brief description of the drawings
Fig. 1 is that the inventive method is applied to the schematic flow sheet that sea-surface target detects
Fig. 2 is the present invention's High Resolution Visible Light to be detected harbour remote sensing images schematic diagram
Fig. 3 is the schematic flow sheet that the present invention is based on region, the sea Seed Points detection method of sweep trace
Fig. 4 be the present invention obtain cut apart template schematic diagram with the onesize Hai Lu of former remote sensing images
Embodiment
Below in conjunction with drawings and Examples, the present invention is described in detail.
The present invention is based on the extra large land template automatic division method that sweep trace detects, comprise the following steps (as shown in first dotted line frame of Fig. 1):
1) input High Resolution Visible Light harbour remote sensing images (as shown in Figure 2) to be detected, based on the relevant resampling disposal route of elemental area, carry out resampling processing according to prior art;
2) for resampling harbour after treatment remote sensing images, adopt region, the sea Seed Points detection method based on sweep trace, detect region, sea Seed Points, its process following (as shown in Figure 3):
1. the resampling harbour after treatment remote sensing images of lining by line scan from top to bottom, first scan the first row;
2. the number of inactive pixels point (inactive pixels point refers to by satellite photo and proofreaies and correct some complete black area pixel point in the remote sensing images that cause) in this row of judgement scanning:
If the number of inactive pixels point is more than or equal to N/10, this row is decided to be to inactive line, 2. scanning next line, get back to step, and wherein N is this row pixel count;
If the number of inactive pixels point is less than N/10, this row gray scale is done to difference after by forward direction, it is the gray-scale value that current pixel point gray-scale value deducts a rear pixel, the absolute value of its difference is difference value, the difference value of every last pixel of row is set to 0, and (this is because for last pixel, there is no pixel, just there is no subtrahend yet, therefore directly the difference value of last point is set to 0 herein) thereafter;
3. investigate difference result, in the result of difference, if difference value is less than 2, be set to 0;
4. judge difference value is whether 0 contiguous pixels number exceedes N/5:
If exceeded, think and occur continuous flat site, be region, sea, get pixel now as region, sea initial seed point, enter step 3);
Otherwise 2. scanning next line, get back to step, until complete the scanning of last column.
3) as shown in Figure 1, according to step 2) region, the sea initial seed point that obtains, utilize the region, region growing algorithm mark sea of prior art:
Due in the remote sensing images of harbour, region, sea is the region more smooth, grey scale change is little, forms so region, sea can be regarded as to the point being closed on by gray-scale value; Using step 2) region, the sea Seed Points that obtains is as the initial seed point of region growing algorithm, utilize region growing algorithm to search the point that all gray-scale values meet the following conditions in adjacent domain: this pixel grey scale with adjoin Seed Points gray scale difference value in 2 and with initial seed point gray scale difference value in 8, the pixel that mark meets this condition is region, sea, and other region (comprising land, harbour, inactive pixels region, cloud layer) is set to land area;
4) to step 3) region, extra large land that obtains carries out binaryzation, is set to 255 by land area gray-scale value, and sea area grayscale codomain is set to 0;
5) binary conversion treatment result is carried out to conventional morphology processing, once corrosion operation and an expansive working, obtains adopting resampling Hai Lu after treatment to cut apart template.
6) to by step 5) Hai Lu that obtains cuts apart template, and, obtain the Hai Lu onesize with former remote sensing images and cut apart template (as shown in Figure 4) based on the relevant resampling disposal route interpolation of elemental area according to prior art.
The present invention is based on the extra large land template automatic division method that sweep trace detects, the sea-surface target that can be applied to various civilian and militaries field detects, and enumerates now an embodiment, to further illustrate the inventive method application.
The present invention is based on the extra large land template automatic division method of sweep trace detection in the application detecting on the Ship Target of sea, it comprises the following steps (as shown in Figure 1):
1) adopt said method to obtain the Hai Lu onesize with former remote sensing images and cut apart template;
2) utilize Hai Lu to cut apart template former harbour remote sensing images are mated, obtain the minimum boundary rectangle of each connected region, it comprises:
1. Hai Lu is cut apart to the part that is labeled as land in template, the land area gray-scale value corresponding at former harbour remote sensing images is set to 0; And Hai Lu is cut apart to the part that is labeled as sea in template, the sea area grayscale value corresponding at former harbour remote sensing images remains unchanged, and obtains thus sea area image;
2. on the area image of sea, adopt the region growing algorithm of known technology to carry out mark to seawater part;
3. by binaryzation, the seawater region gray-scale value of mark is set to 0, other parts gray-scale value is set to 255, obtains the image of sea object;
4. utilize the connected region disposal route of known technology, obtain each candidate's sea-surface target connected region;
5. utilize the minimum boundary rectangle method of searching connected region of known technology, obtain the minimum boundary rectangle of each connected region;
3) the minimum boundary rectangle of each connected region is screened, to determine sea Ship Target, it comprises:
1. the parameter using width, length and the length breadth ratio parameter of minimum boundary rectangle as connected region;
2. according to the priori of Ship Target, the threshold restriction of setting three characteristic parameters of shape of width, length breadth ratio and the connected region of the minimum boundary rectangle of connected region, the sea separate targets connected region that meets above three screening parameter threshold restrictions is sea Ship Target.
The minimum boundary rectangle width of above-mentioned connected region, the yardstick information of length breadth ratio parameter reflection candidate target, the polymerism of the form parameter reflection candidate target region of connected region, form parameter F is defined as follows:
F=||B||
2/4πA
Wherein, B is the girth of connected region, A is the area of connected region, and form parameter F has reflected the compactedness in region to a certain extent, and it does not have dimension, change insensitive to yardstick, rotation, and the span that neither one is fixing, numerical value is larger, and shape is general not compacter regular, this parameter of choose reasonable, just can remove jagged doubtful boats and ships region.
Concrete steps and mode that above-mentioned application the inventive method detects sea-surface target can change to some extent, and every equivalents of carrying out on the basis of technical solution of the present invention and improvement, all should not get rid of outside protection scope of the present invention.
Claims (4)
1. the extra large land template automatic division method detecting based on sweep trace, it comprises the following steps:
1) input High Resolution Visible Light harbour remote sensing images to be detected, according to based on the relevant resampling disposal route of elemental area, carry out resampling processing;
2) for resampling harbour after treatment remote sensing images, adopt region, the sea Seed Points detection method based on sweep trace, detect region, sea Seed Points;
3) according to step 2) region, the sea initial seed point that obtains, utilize region growing algorithm to search the point that all gray-scale values meet the following conditions in adjacent domain: this pixel grey scale with adjoin Seed Points gray scale difference value in 2 and with initial seed point gray scale difference value in 8, the pixel that mark meets this condition is region, sea, and other region is set to land area;
4) to step 3) region, extra large land that obtains carries out binaryzation, is set to 255 by land area gray-scale value, and sea area grayscale codomain is set to 0;
5) binary conversion treatment result is carried out to morphology processing, i.e. once corrosion operation and an expansive working, obtains harbour remote sensing images and adopts resampling Hai Lu after treatment to cut apart template;
6) to by step 5) the resampling Hai Lu after treatment that obtains cuts apart template, and according to based on the relevant resampling disposal route interpolation of elemental area, obtain the Hai Lu onesize with former remote sensing images and cut apart template.
2. a kind of extra large land template automatic division method detecting based on sweep trace as claimed in claim 1, is characterized in that: described step 2) middle region, the sea Seed Points that detects, comprise the following steps:
1. the resampling harbour after treatment remote sensing images of lining by line scan from top to bottom, first scan the first row;
2. the number of inactive pixels point in this row of judgement scanning:
If the number of inactive pixels point is more than or equal to N/10, this row is decided to be to inactive line, 2. scanning next line, get back to step, and wherein N is this row pixel count;
If the number of inactive pixels point is less than N/10, this row gray scale is done to difference after by forward direction, current pixel point gray-scale value deducts the gray-scale value of a rear pixel, and the absolute value of its difference is difference value, and the difference value of every last pixel of row is set to 0;
3. investigate difference result, in the result of difference, if difference value is less than 2, be set to 0;
4. judge difference value is whether 0 contiguous pixels number exceedes N/5:
If exceeded, think and occur continuous flat site, be region, sea, get pixel now as region, sea initial seed point, enter described step 3);
Otherwise 2. scanning next line, get back to step, until complete the scanning of last column.
3. the application of a kind of extra large land template automatic division method detecting based on sweep trace as claimed in claim 1 or 2 in the sea-surface target screening of military field.
4. the application of a kind of extra large land template automatic division method detecting based on sweep trace as claimed in claim 1 or 2 in the sea-surface target screening of civil area.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410317380.1A CN104063870A (en) | 2014-07-04 | 2014-07-04 | Automatic land and sea template segmentation method based on scanning line detection and application thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410317380.1A CN104063870A (en) | 2014-07-04 | 2014-07-04 | Automatic land and sea template segmentation method based on scanning line detection and application thereof |
Publications (1)
Publication Number | Publication Date |
---|---|
CN104063870A true CN104063870A (en) | 2014-09-24 |
Family
ID=51551562
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410317380.1A Pending CN104063870A (en) | 2014-07-04 | 2014-07-04 | Automatic land and sea template segmentation method based on scanning line detection and application thereof |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104063870A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105095846A (en) * | 2014-09-28 | 2015-11-25 | 航天恒星科技有限公司 | Method and system for extracting region growing seed points based on remote sensing images and sea-land segmentation |
CN105654091A (en) * | 2014-11-27 | 2016-06-08 | 航天恒星科技有限公司 | Detection method and apparatus for sea-surface target |
CN106056084A (en) * | 2016-06-01 | 2016-10-26 | 北方工业大学 | Remote sensing image port ship detection method based on multi-resolution hierarchical screening |
CN107145874A (en) * | 2017-05-13 | 2017-09-08 | 复旦大学 | Ship Target Detection and discrimination method in complex background SAR image |
CN108052629A (en) * | 2017-12-19 | 2018-05-18 | 郑州师范学院 | A kind of quick extra large land determination methods based on high accuracy DEM data |
CN108256419A (en) * | 2017-12-05 | 2018-07-06 | 交通运输部规划研究院 | A kind of method for extracting port and pier image using multispectral interpretation |
-
2014
- 2014-07-04 CN CN201410317380.1A patent/CN104063870A/en active Pending
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105095846A (en) * | 2014-09-28 | 2015-11-25 | 航天恒星科技有限公司 | Method and system for extracting region growing seed points based on remote sensing images and sea-land segmentation |
CN105095846B (en) * | 2014-09-28 | 2019-07-30 | 航天恒星科技有限公司 | Region growing seed point extracting method and system towards the segmentation of remote sensing images sea land |
CN105654091A (en) * | 2014-11-27 | 2016-06-08 | 航天恒星科技有限公司 | Detection method and apparatus for sea-surface target |
CN106056084A (en) * | 2016-06-01 | 2016-10-26 | 北方工业大学 | Remote sensing image port ship detection method based on multi-resolution hierarchical screening |
CN106056084B (en) * | 2016-06-01 | 2019-05-24 | 北方工业大学 | Remote sensing image port ship detection method based on multi-resolution hierarchical screening |
CN107145874A (en) * | 2017-05-13 | 2017-09-08 | 复旦大学 | Ship Target Detection and discrimination method in complex background SAR image |
CN108256419A (en) * | 2017-12-05 | 2018-07-06 | 交通运输部规划研究院 | A kind of method for extracting port and pier image using multispectral interpretation |
CN108256419B (en) * | 2017-12-05 | 2018-11-23 | 交通运输部规划研究院 | A method of port and pier image is extracted using multispectral interpretation |
US10325151B1 (en) | 2017-12-05 | 2019-06-18 | Transport Planning and Research Institute Ministry of Transport | Method of extracting image of port wharf through multispectral interpretation |
CN108052629A (en) * | 2017-12-19 | 2018-05-18 | 郑州师范学院 | A kind of quick extra large land determination methods based on high accuracy DEM data |
CN108052629B (en) * | 2017-12-19 | 2021-07-06 | 郑州师范学院 | Rapid sea and land judgment method based on high-precision DEM data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104077777A (en) | Sea surface vessel target detection method | |
CN104063870A (en) | Automatic land and sea template segmentation method based on scanning line detection and application thereof | |
Chen et al. | Ship detection from coastal surveillance videos via an ensemble Canny-Gaussian-morphology framework | |
CN107145874B (en) | Ship target detection and identification method in complex background SAR image | |
KR101856401B1 (en) | Method, apparatus, storage medium, and device for processing lane line data | |
CN102279973B (en) | Sea-sky-line detection method based on high gradient key points | |
CN103942803B (en) | SAR (Synthetic Aperture Radar) image based automatic water area detection method | |
CN103020975A (en) | Wharf and ship segmentation method combining multi-source remote sensing image characteristics | |
CN109117802A (en) | Ship Detection towards large scene high score remote sensing image | |
CN110852992B (en) | Ship trail detection method based on top hat transformation and Radon transformation | |
CN110866926B (en) | Infrared remote sensing image rapid and fine sea-land segmentation method | |
CN105225251A (en) | Over the horizon movement overseas target based on machine vision identifies and locating device and method fast | |
CN105469393A (en) | Shallow water depth multi-temporal remote sensing image inversion method based on decision fusion | |
CN108229342A (en) | A kind of surface vessel target automatic testing method | |
CN106803078B (en) | SAR image ship target segmentation method | |
CN104992172A (en) | Port remote sensing image shore ship detection method based on coastal outburst and different area scanning | |
CN102054275B (en) | Real-time detection method for sea sky line of gray level image | |
CN103500451B (en) | A kind of independent floating ice extracting method for satellite data | |
CN109961065B (en) | Sea surface ship target detection method | |
CN111105390B (en) | Improved sea-sky-line detection and evaluation method | |
CN107704835B (en) | Method for identifying offshore artificial facilities by using spectrum remote sensing images | |
CN109325958A (en) | A kind of offshore ship detection method for refining and improve generalised Hough transform based on profile | |
Medina et al. | Management of dynamic navigational channels using video techniques | |
CN104881670A (en) | Rapid target extraction method used for SAR azimuth estimation | |
CN115639536B (en) | Unmanned ship perception target detection method and device based on multi-sensor fusion |
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: 20140924 |
|
RJ01 | Rejection of invention patent application after publication |