CN106780525B - Optical remote sensing image ship direction feature extraction method based on coordinate rotation minimum circumscribed rectangle - Google Patents
Optical remote sensing image ship direction feature extraction method based on coordinate rotation minimum circumscribed rectangle Download PDFInfo
- Publication number
- CN106780525B CN106780525B CN201611016618.2A CN201611016618A CN106780525B CN 106780525 B CN106780525 B CN 106780525B CN 201611016618 A CN201611016618 A CN 201611016618A CN 106780525 B CN106780525 B CN 106780525B
- Authority
- CN
- China
- Prior art keywords
- ship
- circumscribed rectangle
- target
- points
- minimum
- 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.)
- Active
Links
- 238000000605 extraction Methods 0.000 title claims abstract description 19
- 230000003287 optical effect Effects 0.000 title abstract description 11
- 238000000034 method Methods 0.000 claims abstract description 32
- 238000007781 pre-processing Methods 0.000 claims abstract description 5
- 230000011218 segmentation Effects 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 3
- 230000003044 adaptive effect Effects 0.000 claims description 2
- 230000003628 erosive effect Effects 0.000 claims 1
- 238000011160 research Methods 0.000 abstract description 2
- 238000004364 calculation method Methods 0.000 description 5
- 230000007547 defect Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000002093 peripheral effect Effects 0.000 description 3
- 230000007797 corrosion Effects 0.000 description 2
- 238000005260 corrosion Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- 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
Abstract
And (4) ship direction feature extraction of the optical remote sensing image based on the minimum circumscribed rectangle of coordinate rotation. The research provides the ship direction feature extraction of the optical remote sensing image based on the coordinate rotation minimum circumscribed rectangle, and the ship direction feature extraction method adopts a method of rotating a coordinate axis and efficiently and accurately obtains the direction features of the orientation, the length, the width and the like of a target by extracting the minimum circumscribed rectangle of the ship. Firstly, image preprocessing is carried out to obtain a single-pixel edge of a target, and the single-pixel edge is used as an effective reference area for extracting target direction features. Secondly, aiming at the obtained edge contour image, carrying out initial judgment on the target orientation by adopting a quadrant division strategy, wherein the step aims to limit the search range of the minimum circumscribed rectangle and improve the algorithm efficiency; and thirdly, under the limitation of the target initial judgment range in the previous step, searching for a minimum circumscribed rectangle in the range, and performing initial judgment on the heading of the ship based on the minimum circumscribed rectangle. And finally, extracting ship direction related features such as the length, the width, the orientation and the like of the target by using the extracted minimum circumscribed rectangle.
Description
Technical Field
The invention relates to a coordinate rotation minimum circumscribed rectangle-based optical remote sensing image ship direction feature extraction method.
Background
In recent years, with the rapid development of the air-to-air platform-mounted technology, the space remote sensing technology is also continuously perfected and advanced, and the research and technical development of ship detection and monitoring by using the acquired remote sensing image are more and more emphasized in the field of ocean remote sensing.
The optical remote sensing image ship direction feature extraction is that based on optical remote sensing image data, an image processing method is used for extracting features related to ship directions, and the technology can be matched with an automatic detection technology to achieve automatic and effective extraction of remote sensing data sea ship information. The system is applied to the civil field, and can support shipping monitoring, marine fishery supervision, water traffic control and the like. The application in the military field can be used for: dynamic monitoring of ships at foreign military ports, early warning of departure of key ship targets and the like.
In the extraction of remote sensing features of a sea surface ship, the orientation of the ship is key information for automatically interpreting a ship target of remote sensing data. The current technology related to ship direction extraction mostly acquires ship directions by traversing a target area and searching long and short axes. The traditional method for acquiring ship direction characteristics by calculating the long axis and the short axis of the connected region is easily interfered by the integrity of the mark calculation of the connected region and a small number of false alarm points around the target, and the method for searching the long axis and the short axis by traversing the target region has large calculation amount and low efficiency.
The scheme provides a coordinate rotation minimum circumscribed rectangle-based method for extracting ship direction features of optical remote sensing images, and the method can effectively solve the problems of connected domain mark integrity and peripheral false alarm point interference in the traditional method by considering the integrity of the periphery of a target. In addition, the method adopts a strategy of rotating coordinate axes, and the orientation of the target bow is primarily judged, and the angle of the rotating traversal range is limited to [0,45 degrees ] after the primary judgment, so that the complexity and the time of calculation are greatly reduced, and the efficiency is improved.
Disclosure of Invention
According to one aspect of the invention, the method for extracting the ship direction feature of the remote sensing image based on the coordinate rotation minimum circumscribed rectangle is provided, and is used for extracting the ship direction feature, and comprises the following steps:
A) performing image pre-processing, comprising:
carrying out self-adaptive threshold segmentation on the target slice to obtain a binary image of the gray level input image;
median filtering is carried out on the binary image, interference noise points of the peripheral region of the target ship body are removed, and the noise points are prevented from seriously interfering the image processing;
carrying out binary corrosion and binary expansion on the filtered image to achieve the purpose of filling holes in the target ship body, and then carrying out binary expansion operation to expand the binary area of the target ship body by one circle to prepare for extracting edges;
subtracting the area binary image of the target hull from the binary image of the target hull with the edge extended by adopting a binary morphology method to obtain a single-pixel edge contour image of the target hull;
B) aiming at the obtained edge image, quickly determining an initial angle of a coordinate based on a method for counting a single-pixel edge contour of a binary connected region of a target hull in 4 quadrants of a coordinate axis, and performing initial judgment on the orientation of the target hull;
C) under the limitation of the initial judgment result of the orientation of the target ship body in the step B), making a minimum circumscribed rectangle of the search range and calculating the ship direction based on the minimum circumscribed rectangle;
D) and extracting ship direction related features including the length, width and orientation of the ship by using the acquired minimum circumscribed rectangle.
Drawings
Fig. 1 is a general flowchart of the ship direction feature extraction of a remote sensing image based on a coordinate rotation minimum bounding rectangle.
FIG. 2 is a flow chart for fast acquisition of a minimum bounding rectangle.
Fig. 3(a) to 3(c) are schematic diagrams of initial determination of the orientation of the target.
Fig. 4 is a drawing illustrating the quadrant division of the initial judgment of the rectangle.
FIG. 5 is a diagram of minimum bounding rectangle validation and bow initialization for a limited search range.
Detailed description of the preferred embodiments
The invention aims to provide an optical remote sensing image ship direction feature extraction method based on the minimum circumscribed rectangle of coordinate rotation aiming at the defects of the prior art, so as to efficiently and accurately acquire ship direction related information in a remote sensing image.
Aiming at the situations, the invention adopts an extraction method of ship direction characteristics of the optical remote sensing image based on the coordinate rotation minimum circumscribed rectangle. There is no report of such a method.
The invention provides a method for extracting ship direction characteristics of a ship optical remote sensing image based on a coordinate rotation minimum circumscribed rectangle. The method comprises the following specific steps:
the first step is as follows: image pre-processing
And carrying out self-adaptive threshold segmentation on the target slice to obtain a binary image of the gray-scale input image, and carrying out binary median filtering on the binary image to remove scattered interference noise points in the target image. And then filling holes and expanding edges, and obtaining a single-pixel edge profile of the target binary connected region by adopting a binary morphology method.
The second step is that: ship direction initial judgment based on quadrant division
Firstly, initial judgment is carried out on the orientation of a target, the initial angle of the coordinate is rapidly determined by counting the single-pixel edge contour (obtained in the previous step) of a target binary connected region in 4 quadrants of a coordinate axis, and the subsequent search range is shortened. And then, preliminarily judging the target orientation by judging the number of target points in the two areas of the bow and the stern.
The third step: quick acquisition of minimum ship external rectangle within initial judgment limit range
After the initial direction judgment of the previous step, the minimum circumscribed rectangle is determined in the limited search range through coordinate rotation, the coordinate system is rotated for 1 degree from the initial angle every time, scanning is performed for 45 times in total, the circumscribed rectangle with the minimum area is found out, and the rotation angle at the moment is recorded, and the rotation angle at the moment is the ship direction required by people. FIG. 2 is a flow chart of fast minimum bounding rectangle acquisition.
The fourth step: direction feature extraction based on minimum circumscribed rectangle
The heading of the ship can be determined according to the main shaft direction of the minimum circumscribed rectangle by the minimum circumscribed rectangle obtained from the single-pixel edge of the target ship body in the last step; based on the length and width of the minimum bounding rectangle found, the length and width of the vessel can be determined directly.
How to implement the method provided by the present invention is described below, and fig. 1 is a flow chart diagram of a method according to an embodiment of the present invention, the method including:
the first step is as follows: image pre-processing
Step (1.1) adaptive threshold segmentation: firstly, self-adaptive threshold segmentation is carried out on the gray level image to obtain a binary image, so that the edge contour of the target ship body can be conveniently extracted and prepared.
And (1.2) median filtering: and performing median filtering on the binary image to remove interference noise points in a target peripheral region and prevent the noise points from seriously interfering image processing.
Step (1.3) hole filling and edge enlargement: carrying out binary corrosion and binary expansion on the filtered image to achieve the purpose of filling holes in the ship body area; and then, carrying out binary expansion operation to expand the binary region of the target hull by one circle to prepare for extracting the edge contour.
And (1.4) extracting edges: and (3) subtracting the target area binary image from the target binary image with the edge extended by adopting a binary morphology method to obtain a single-pixel edge contour image of the target hull.
The second step is that: ship direction initial judgment based on quadrant division
And (2.1) initially judging the orientation of the target: quadrant division is carried out on the target contour image: and subtracting the pixel point with the minimum abscissa from the pixel point with the maximum abscissa value to obtain the width of the rectangle, and subtracting the pixel point with the minimum ordinate from the pixel point with the maximum ordinate value to obtain the height of the rectangle. The circumscribed rectangle is divided into four regions with the center line as a boundary. The preliminary orientation of the target can be counted based on quadrant division: counting the points of the target contour in the quadrants, if the points in the first quadrant and the third quadrant are more than the points in the second quadrant and the fourth quadrant, the ship body faces to the right, otherwise, the ship body faces to the left, as shown in fig. 3(a) -3 (c). And vice versa.
And (2.2) quickly determining the initial angle: each quadrant is divided into two areas with the size of 45 degrees, the four quadrants are divided into 8 areas with the same size, and the areas above the abscissa axis are respectively set as four areas a, b, c and d from right to left. If the target ship body faces right, counting points with the target contour pixel value of 1 in the area 1, if the number of points falling in the area b is more than a, setting the initial angle to be 0, otherwise setting the initial angle to be 45 degrees; if the ship is facing left and the number of points falling in the area c is more than d, the initial angle is 0, otherwise it is 45 °. As shown in fig. 4.
The third step: quick acquisition of minimum ship external rectangle within initial judgment limit range
And (3.1) determining the minimum bounding rectangle in a limited search range through coordinate rotation: and after the initial angle range of the target is obtained in the last step, the coordinate system is rotated by 1 degree from the initial angle every time, if the target ship body faces right, the target ship body rotates clockwise, and if not, the target ship body rotates anticlockwise. Scanning is carried out for 45 times in total, and the circumscribed rectangle after each scanning is obtained. And finding out the circumscribed rectangle with the minimum area and recording the rotation angle at the moment. The rotation angle at this time is the ship direction required by people, because when the circumscribed rectangle is the minimum, the central line of the circumscribed rectangle coincides with the central axis of the target ship body, and the circumscribed rectangle is tangent to the edge of the target ship body.
And (3.2) primarily judging the orientation of the bow: the minimum external rectangle obtained in the above steps is divided into two perpendicular line segments for two parallel long sides, so that the minimum external rectangle is divided into three areas, namely, the target ship body is divided into a bow part, a stern part and a middle part, the orientation of the bow part is initially judged by counting the number of pixel points of the outlines of the two edges of the ship body 1 and 3, namely, the side where the area with the less pixel points of 1 is positioned is determined as the side of the bow part. As shown in fig. 5.
The fourth step: direction feature extraction based on minimum circumscribed rectangle
And (4.1) acquiring the direction characteristics of the bow of the ship: the direction of the bow is obtained through the direction calculated in the minimum external rectangle obtaining process, and in the process of obtaining the external rectangle, the external rectangle obtained every time can correspond to a corresponding angle, and the angle of the minimum external rectangle is recorded when the minimum external rectangle is found out by comparing the size of the external rectangle, namely the direction angle of the bow. And further giving the determination of the ship heading through the initial judgment of the ship bow.
Step (4.2): vessel length and width feature acquisition: the length of the long and short sides of the minimum bounding rectangle is the length and width of the ship.
Compared with the existing detection method, the invention has the following advantages:
(1) and extracting the direction by adopting the target edge contour. The target contour is the most essential image characteristic for measuring the target direction, the method uses the extracted target edge contour to replace the direct region extraction of the target, can directly correlate the essence of the direction information, reduces the influence of holes generated by segmentation on the direction estimation, avoids the inaccuracy of the direction estimation caused by inaccurate segmentation of the region information, and is beneficial to more accurately extracting the direction characteristic information of the target.
(2) And adopting a target direction estimation strategy of a minimum bounding rectangle. The overall form of the target is considered, the defect that errors are easy to occur due to the fact that local information is considered in the traversal target extraction method is avoided, for example, result deviation caused by interference at irregular details and the like is avoided, and the direction feature of the target can be extracted with high accuracy.
(3) The algorithm efficiency is improved by adopting the direction initial judgment. The target direction initial judgment method based on quadrant division is provided, the quadrant initial judgment of the bow is firstly carried out before the minimum external rectangle is determined by rotating a coordinate axis, the method reduces the search range, the defect of large calculation amount in full-angle search is avoided, and the calculation efficiency is effectively improved.
(4) The direction-related features are directly solved by the minimum circumscribed rectangle. The long axis and the short axis of the target can be directly obtained by directly using the long edge and the short edge of the minimum circumscribed rectangle, so that the complex process of independently obtaining the long axis and the short axis of the ship by using a full-image traversal method in the prior art is avoided; meanwhile, the minor axis of the ship is extracted by using the minimum circumscribed rectangle, so that the orthogonal verticality of the minor axis is effectively ensured, the influence of noise, incomplete segmentation and other reasons on the extraction of the minor axis in the traditional method is avoided, and the method has higher robustness.
Claims (2)
1. A remote sensing image ship direction feature extraction method based on coordinate rotation minimum circumscribed rectangle is used for extracting ship direction features and comprises the following steps:
A) performing image pre-processing, comprising:
adaptive threshold segmentation is performed on the ship slices to obtain a binary image of the grayscale input image,
median filtering is carried out on the binary image, interference noise points in the circumference of the ship are removed, the serious interference of the noise points on the image processing is prevented,
performing binary erosion and binary expansion on the filtered image to fill holes in the ship, performing binary expansion to expand the binary area of the ship by one circle to prepare for extracting edges,
using a binary morphology method, subtracting the ship area binary image from the ship binary image of the edge extension to obtain a single-pixel edge contour image of the ship,
B) aiming at the obtained edge image, based on a method for counting the single-pixel edge contour of a binary connected region of a ship in four quadrants of a coordinate axis, the method quickly determines the initial angle of the coordinate and carries out initial judgment on the orientation of the ship, and comprises the following steps:
B1) the ship orientation is judged initially, including carrying out quadrant division to ship profile image, include: subtracting the pixel point with the minimum abscissa from the pixel point with the maximum abscissa value to obtain the width of the circumscribed rectangle, and simultaneously subtracting the pixel point with the minimum ordinate from the pixel point with the maximum ordinate value to obtain the height of the circumscribed rectangle; dividing the external connecting rectangle into four areas by taking the central line as a boundary; based on the quadrant division, the preliminary orientation of the ship can be counted, points of the ship outline in the quadrants are counted, if the points falling in the quadrant like the first limit and the third limit are more than the points falling in the quadrant like the second limit and the fourth limit, the ship faces to the right, otherwise the ship faces to the left,
B2) rapidly determining the initial angle, namely continuously dividing each quadrant in the step B1) into two areas with the size of 45 degrees, dividing the four quadrants into 8 areas with the same size, and setting the areas above the abscissa axis as four areas a, B, c and d from right to left; if the ship faces right according to the step B1), counting the number of points with the ship contour pixel value of 1 in the first quadrant, if the number of points with the ship contour pixel value of 1 in the area B is more than the number of points with the ship contour pixel value of 1 in the area a, the initial angle is 0, otherwise, the initial angle is 45 degrees; if it is found from step B1) that the number of points at which the contour pixel value of the ship facing left and falling on area c is 1 is greater than the number of points at which the contour pixel value of the ship falling on area d is 1, the initial angle is 0, otherwise 45,
C) and under the limitation of the initial judgment result of the orientation of the ship in the step B), making a minimum circumscribed rectangle of the search range and calculating the ship direction based on the minimum circumscribed rectangle, wherein the specific steps are as follows.
C1) Determining the minimum bounding rectangle within the limited search range by coordinate rotation: after the initial angle range of the ship is obtained in the step B2), the coordinate system is rotated by 1 degree from the initial angle every time, if the ship faces right, the ship rotates clockwise, otherwise, the ship rotates anticlockwise for 45 times, the circumscribed rectangle after each scanning is obtained, the circumscribed rectangle with the smallest area is found out, and the rotating angle at the moment is recorded, the rotating angle at the moment is the ship direction required by people, because when the circumscribed rectangle is smallest, the central line of the circumscribed rectangle coincides with the central axis of the ship, and the circumscribed rectangle is tangent to the edge of the ship;
C2) the bow is towards first judgement, includes: for the minimum circumscribed rectangle determined in step C1), two perpendicular line segments are made for the two parallel long sides, thereby equally dividing the minimum circumscribed rectangle into three regions, i.e., dividing the ship into three parts: a bow part, a stern part and a middle part, and counting the number of points of which the pixel points of the edge contour in the bow 1 and the stern are 1, and determining the part with the less number of the pixel points of the edge contour being 1 as the actual bow,
D) and extracting ship direction related features including the length, width and orientation of the ship by using the acquired minimum circumscribed rectangle.
2. The method according to claim 1, characterized in that said step D) comprises:
D1) obtaining the direction of the bow from the direction calculated in the process of obtaining the minimum bounding rectangle, wherein, by using the corresponding rotation angle of the bounding rectangle obtained in each time in the step C1), the angle of the minimum bounding rectangle is recorded while the minimum bounding rectangle is determined, the angle is taken as the direction of the ship, and the direction of the bow is determined by combining the bow orientation determined in the step C2),
D2) the lengths of the long and short sides of the minimum bounding rectangle are determined as the length and width of the ship, respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611016618.2A CN106780525B (en) | 2016-11-18 | 2016-11-18 | Optical remote sensing image ship direction feature extraction method based on coordinate rotation minimum circumscribed rectangle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611016618.2A CN106780525B (en) | 2016-11-18 | 2016-11-18 | Optical remote sensing image ship direction feature extraction method based on coordinate rotation minimum circumscribed rectangle |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106780525A CN106780525A (en) | 2017-05-31 |
| CN106780525B true CN106780525B (en) | 2020-08-14 |
Family
ID=58969044
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611016618.2A Active CN106780525B (en) | 2016-11-18 | 2016-11-18 | Optical remote sensing image ship direction feature extraction method based on coordinate rotation minimum circumscribed rectangle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106780525B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109034173A (en) * | 2017-06-08 | 2018-12-18 | 北京君正集成电路股份有限公司 | Target object choosing method and device |
| CN108460367A (en) * | 2018-03-28 | 2018-08-28 | 北方工业大学 | Target detection method based on prejudgment mechanism |
| CN108596223A (en) * | 2018-04-11 | 2018-09-28 | 珠海博明视觉科技有限公司 | A method of automatically generating object data set |
| CN108830862B (en) * | 2018-06-08 | 2021-11-30 | 江南大学 | Crab orientation identification method based on image segmentation |
| CN109034153B (en) * | 2018-07-20 | 2022-04-26 | 泰康保险集团股份有限公司 | Image fidelity method and device applied to image rotation |
| CN109657713A (en) * | 2018-12-11 | 2019-04-19 | 武汉大学 | A kind of multiple-factor road network method and system based on crowd-sourced road net data |
| CN110223302B (en) * | 2019-05-08 | 2021-11-19 | 华中科技大学 | Ship multi-target detection method based on rotation region extraction |
| CN112184744B (en) * | 2020-11-29 | 2021-03-30 | 惠州高视科技有限公司 | Display screen edge defect detection method and device |
| CN113534795A (en) * | 2021-06-30 | 2021-10-22 | 深圳市银星智能科技股份有限公司 | Equipment searching method and self-moving equipment |
| CN114047865B (en) * | 2021-11-11 | 2023-04-25 | 四川长虹教育科技有限公司 | Graphic control method |
| CN114281076B (en) * | 2021-12-13 | 2024-02-09 | 烟台杰瑞石油服务集团股份有限公司 | Covering and moving operation method of robot |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001357401A (en) * | 2000-06-12 | 2001-12-26 | Matsushita Electric Works Ltd | Picture processing method |
| CN102496018A (en) * | 2011-12-08 | 2012-06-13 | 方正国际软件有限公司 | Document skew detection method and system |
| CN104077777A (en) * | 2014-07-04 | 2014-10-01 | 中国科学院大学 | Sea surface vessel target detection method |
| CN104484883A (en) * | 2014-12-24 | 2015-04-01 | 河海大学常州校区 | Video-based three-dimensional virtual ship positioning and track simulation method |
| CN105809705A (en) * | 2016-03-30 | 2016-07-27 | 广东工业大学 | Patch element positioning identification method based on minimum enclosing rectangle |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7602995B2 (en) * | 2004-02-10 | 2009-10-13 | Ricoh Company, Ltd. | Correcting image distortion caused by scanning |
-
2016
- 2016-11-18 CN CN201611016618.2A patent/CN106780525B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001357401A (en) * | 2000-06-12 | 2001-12-26 | Matsushita Electric Works Ltd | Picture processing method |
| CN102496018A (en) * | 2011-12-08 | 2012-06-13 | 方正国际软件有限公司 | Document skew detection method and system |
| CN104077777A (en) * | 2014-07-04 | 2014-10-01 | 中国科学院大学 | Sea surface vessel target detection method |
| CN104484883A (en) * | 2014-12-24 | 2015-04-01 | 河海大学常州校区 | Video-based three-dimensional virtual ship positioning and track simulation method |
| CN105809705A (en) * | 2016-03-30 | 2016-07-27 | 广东工业大学 | Patch element positioning identification method based on minimum enclosing rectangle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106780525A (en) | 2017-05-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106780525B (en) | Optical remote sensing image ship direction feature extraction method based on coordinate rotation minimum circumscribed rectangle | |
| CN108229433B (en) | Method for detecting ship landing on shore based on straight-line segment detection and shape characteristics | |
| CN101609504B (en) | Method for detecting, distinguishing and locating infrared imagery sea-surface target | |
| CN104077777A (en) | Sea surface vessel target detection method | |
| CN115456970B (en) | Method for identifying carrier line of topside-cut ship based on dimension-reduction regression | |
| CN110428425B (en) | Sea-land separation method of SAR image based on coastline vector data | |
| CN108846844B (en) | Sea surface target detection method based on sea antenna | |
| CN107679535A (en) | A kind of pointer-type water meter automatic indication recognition system and method based on template matches | |
| CN106651880B (en) | Offshore moving target detection method based on multi-feature fusion thermal infrared remote sensing image | |
| CN106803078B (en) | SAR image ship target segmentation method | |
| CN108288268B (en) | Method for detecting ship landing in high-resolution remote sensing image based on projection analysis | |
| CN106682674B (en) | A kind of pointer instrument automatic reading method based on intelligent inspection robot | |
| CN112884746A (en) | Character defect intelligent detection algorithm based on edge shape matching | |
| CN106709523B (en) | Optical remote sensing image ship identification method based on S-HOG characteristics | |
| CN113298775B (en) | Self-priming pump double-sided metal impeller appearance defect detection method, system and medium | |
| CN113177593B (en) | Fusion method of radar point cloud and image data in water traffic environment | |
| CN117115196B (en) | Visual detection method and system for cutter abrasion of cutting machine | |
| CN109325958B (en) | Method for detecting offshore ship based on contour refinement and improved generalized Hough transform | |
| CN111105390B (en) | Improved sea-sky-line detection and evaluation method | |
| CN109934817B (en) | Method for detecting malformation of external contour of fruit body | |
| CN109918968B (en) | Ship target detection method | |
| CN111105419A (en) | Vehicle and ship detection method and device based on polarized SAR image | |
| CN114187195A (en) | 2D (two-dimensional) forward-looking sonar image denoising method | |
| Wu et al. | Ship recognition from high resolution remote sensing imagery aided by spatial relationship | |
| CN112990064A (en) | Dial pointer identification method based on color segmentation and probability model |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Bi Fukun Inventor after: Han Jianhong Inventor after: Zhang Xu Inventor after: Chen Jing Inventor after: Ge Xianjun Inventor after: Ma Xiao Inventor before: Bi Fukun Inventor before: Zhang Xu Inventor before: Chen Jing Inventor before: Ge Xianjun Inventor before: Ma Xiao |
|
| CB03 | Change of inventor or designer information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |