CN106709426B - Ship target detection method based on infrared remote sensing image - Google Patents
Ship target detection method based on infrared remote sensing image Download PDFInfo
- Publication number
- CN106709426B CN106709426B CN201611076345.0A CN201611076345A CN106709426B CN 106709426 B CN106709426 B CN 106709426B CN 201611076345 A CN201611076345 A CN 201611076345A CN 106709426 B CN106709426 B CN 106709426B
- Authority
- CN
- China
- Prior art keywords
- image
- ship
- target
- ship target
- remote sensing
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a ship target detection method based on an infrared remote sensing image, which comprises the following steps: s1: carrying out water-bank separation on the infrared remote sensing image, and segmenting a sea area image containing a ship target; s2: carrying out contrast enhancement processing on the sea area image obtained after water-bank separation, highlighting a ship target in the image, and selecting a candidate area of the ship target according to the highlighted ship target; s3: and determining a suspected ship target from the candidate area, and extracting the ship target according to the personalized characteristics of the ship. Under the condition of complex background interference of wind waves, cloud background, ship trails, solar flares, islands and the like, the ship target can be accurately detected in real time.
Description
Technical Field
The invention relates to a target detection and identification technology in the field of infrared remote sensing images, in particular to a ship target detection method based on infrared remote sensing images.
Background
Although the ship target detection and identification technology based on the infrared remote sensing image is relatively late in research and start, with the continuous improvement of the load resolution, the high-resolution infrared image has outstanding advantages and broad prospects in the aspect of ship identification. At present, the detection of a ship target based on an infrared remote sensing image under a complex background is still a difficult point; on one hand, the data volume is very large due to the large ocean area and the very rich ship information on the sea surface, and the existing detection method based on ship histogram trailing property and the detection method based on C-V threshold segmentation are time-consuming and not beneficial to real-time processing; on the other hand, due to the interference of wind waves, cloud backgrounds, ship trails, solar flares, islands and other factors, more false alarms are often caused during ship detection.
Disclosure of Invention
The invention aims to solve the technical problem of how to accurately detect a ship target in real time under the condition of complex background interference of wind waves, cloud backgrounds, ship trails, solar flares, islands and the like.
In order to solve the problems, the invention provides a ship target detection method based on an infrared remote sensing image, which comprises the following steps:
s1: carrying out water-bank separation on the infrared remote sensing image, and segmenting a sea area image containing a ship target;
s2: carrying out contrast enhancement processing on the sea area image obtained after water-bank separation, highlighting a ship target in the image, and selecting a candidate area of the ship target according to the highlighted ship target;
s3: and determining a suspected ship target from the candidate area, and extracting the ship target according to the personalized characteristics of the ship.
According to an embodiment of the present invention, in step S1, the image is divided into an ocean part, a land part and a sea-land mixed part according to an image entropy statistic result of frequencies within a set frequency range in an image frequency spectrum of the infrared remote sensing image, a coastline is determined through edge extraction, and the sea area image is segmented.
According to an embodiment of the present invention, the step S1 includes the steps of:
s11: dividing the infrared remote sensing image into a plurality of image small blocks, performing frequency spectrum calculation on each image small block, calculating the image entropy of the frequency within a set frequency range, and performing accumulation statistics to obtain the total value of the image entropy of each image small block;
s12: comparing the total image entropy value of each image small block with a preset upper threshold value and a preset lower threshold value, judging the image small block of which the total image entropy value is smaller than the preset lower threshold value as an ocean part, and judging the image small block of which the image entropy is larger than the preset upper threshold value as a land part and the rest image small blocks as an ocean-land mixed part;
s13: and extracting the boundary of the sea-land change of the image small block as a seed point, performing region growth, drawing a sea-land line to realize water-land separation, and segmenting the sea area image containing the ship target.
According to an embodiment of the invention, the set frequency range is 4-20 Hz.
According to an embodiment of the present invention, in step S2, by performing Top-Hat transform and Bottom-Hat transform on the sea area image obtained after water-shore separation, the gray contrast between the ship target and the ocean background is enhanced, and the ship target is highlighted.
According to an embodiment of the present invention, the determining the ship suspected target from the candidate area in step S3 includes the following steps:
s31: performing Top-Hat transformation processing on the image of the ship target candidate region;
s32: carrying out gray morphological reconstruction on the image after the transformation processing;
s33: and performing threshold segmentation on the reconstructed image to determine the suspected ship target.
According to an embodiment of the present invention, the extracting of the optimal ship target according to the ship personalized features in the step S3 includes the following steps:
s34: counting the pixel area of each communication area aiming at the image segmented by the threshold, reserving the communication areas with the pixel areas within a target communication setting range, and removing the rest communication areas, wherein the target communication setting range is determined according to the size of a ship target;
s35: extracting the framework of the reserved communication area, calculating the length-width ratio of the framework, reserving the framework which accords with the set range of the target length-width ratio, and removing the rest frameworks, wherein the set range of the target length-width ratio is determined according to the length-width ratio of the target framework of the ship;
s36: calculating the gray level change of the reserved framework, eliminating the frameworks which have unchanged gray levels and the framework lengths of which are greater than a preset length value, and reserving the rest target frameworks;
s37: morphological reconstruction is performed according to the target skeleton retained in the step S36 and the image segmented by the threshold value in the step S34, so that a ship target is extracted.
According to an embodiment of the present invention, in the step S34, the morphological dilation is performed on the threshold-segmented image, and then the pixel area of each connected region is counted.
After the technical scheme is adopted, compared with the prior art, the invention has the following beneficial effects:
1) reducing the detection range of the ship target by water bank segmentation, then carrying out denoising processing on the segmented image, inhibiting ocean background noise, enhancing the ship target, finally removing false alarms according to the characteristics of the ship target, extracting the real ship target, reducing the false alarm rate of ship detection, and adapting to the ship target detection in a complex environment;
2) the gray level and texture feature difference of the water bank area is not obvious, the interference of the ship shadow is added, the complete water bank outline is difficult to extract, image spectrum statistics is carried out on the block area, the image is divided into three parts, namely smooth, complex and mixed parts, based on priori knowledge, the smooth part is seawater, the complex part is land, the seed points of the mixed part are calculated for regional growth, the water bank is completely divided, the land and the sea can be quickly divided, and the ship detection efficiency is improved;
3) according to the characteristics of the ship target, the real ship target is detected by methods such as connected region judgment, skeleton extraction, length-width ratio judgment, gray judgment and the like, and the detection method has strong adaptability.
Drawings
Fig. 1 is a schematic flow chart of a ship detection method based on infrared remote sensing images according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of water bank splitting according to an embodiment of the present invention;
FIGS. 3a and 3b are schematic diagrams of different image patches and image spectra thereof according to an embodiment of the present invention;
FIG. 4 is an effect diagram of an original image and a segmented image;
FIG. 5 is a diagram of background suppression and the effect of performing gray scale reconstruction on the result of background suppression;
FIG. 6 is a schematic flow diagram of ship target detection validation;
fig. 7 is a graph of segmentation and recognition effects under different background interference conditions.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein, but rather construed as limited to the embodiments set forth herein.
Referring to fig. 1, in one embodiment, the method for detecting the ship target based on the infrared remote sensing image comprises the following steps:
s1: carrying out water-bank separation on the infrared remote sensing image, and segmenting a sea area image containing a ship target;
s2: carrying out contrast enhancement processing on the sea area image obtained after water-bank separation, highlighting a ship target in the image, and selecting a candidate area of the ship target according to the highlighted ship target;
s3: and determining a suspected ship target from the candidate area, and extracting the ship target according to the personalized characteristics of the ship.
In step S1, the infrared remote sensing image may be obtained by an existing infrared remote sensing image acquisition device, but is not limited thereto, and may be obtained by other channels. Firstly, water-bank separation is carried out on the infrared remote sensing image, ocean and land are primarily segmented, the image of the sea area containing the ship target is segmented, and the detection range of the ship target is narrowed. The sea area image contains ship target information and ocean background noise, so that the ocean background noise needs to be further eliminated.
In one embodiment, in step S1, the ocean and land are initially segmented through statistical analysis of the image frequency spectrum, specifically, the image is divided into an ocean part, a land part and a sea-land mixed part according to the image entropy statistical result of the frequency in the set frequency band range in the image frequency spectrum of the infrared remote sensing image, and the coastline is determined through edge extraction to segment the sea area image. The gray level and texture feature difference of the water bank area is not obvious, and the interference of the ship shadow is added, so that the complete water bank outline is difficult to extract, the image is divided into three parts of smoothness, complexity and mixture by carrying out image spectrum statistics on the image, based on priori knowledge, the smooth part is seawater, the complex part is land, and the water bank can be completely segmented by carrying out edge extraction according to the preliminary segmentation.
Referring to fig. 2, in one embodiment, step S1 further includes the steps of:
s11: dividing the infrared remote sensing image into a plurality of image small blocks, performing frequency spectrum calculation on each image small block, calculating the image entropy of the frequency within a set frequency range, and performing accumulation statistics to obtain the total value of the image entropy of each image small block;
s12: comparing the total image entropy value of each image small block with a preset upper threshold value and a preset lower threshold value, judging the image small block of which the total image entropy value is smaller than the preset lower threshold value as an ocean part, and judging the image small block of which the image entropy is larger than the preset upper threshold value as a land part and the rest image small blocks as an ocean-land mixed part;
s13: and extracting the boundary of the sea-land change of the image small block as a seed point, performing region growth, drawing a sea-land line to realize water-land separation, and segmenting the sea area image containing the ship target.
Specifically, in step S11, the infrared remote sensing image may be divided into 50 × 50 image patches, an image frequency spectrum of each image patch is obtained through calculation, and the image entropy of each image patch frequency within the set frequency band range is calculated and accumulated to obtain the total image entropy value of each image patch. Preferably, the frequency range is set to be 4-20 Hz, and the frequency spectrum in the frequency range is more suitable for sea and land image analysis.
In step S12, the total value of the image entropy of each image patch is compared with a preset upper threshold and a preset lower threshold, referring to the land image in fig. 3a and the sea image in fig. 3b, the entropy of the land image is relatively large, and the entropy of the sea image is relatively small, based on this, the image patch whose total value of the image entropy is smaller than the preset lower threshold is determined as the sea portion, the image patch whose image entropy is larger than the preset upper threshold is determined as the land portion, and the rest of the image patches are determined as the sea-land mixed portion. The preset upper threshold and the preset lower threshold may be set according to a plurality of experimental experience accumulations, and are not particularly limited.
In step S13, the boundary of sea-land change of the image patch is extracted as a seed point, for example, the image is binarized, the boundary of 0-1 change is extracted as a seed point for region growth, the boundary of 0-1 change may refer to the boundary between the image patch of the sea part and the image patch of the land part, or the sea-land boundary line in the image patch of the sea-land mixed part, after the region growth, the sea-land line is accurately drawn, water-land separation is realized, and the sea-land image including the ship target is segmented, see the original image and the segmented image in fig. 4. The region growing according to the seed points is a conventional image processing technique, and is not described herein again.
In step S2, the sea area image obtained after the water-shore separation is subjected to contrast enhancement processing to highlight the ship target in the image, and a candidate area of the ship target is selected based on the highlighted ship target.
In one embodiment, in step S2, due to interference of ocean background noise such as solar flare, wind wave, ship trail, etc., a high false alarm rate may be generated when performing image segmentation recognition on a ship target, and by performing Top-Hat transform (a mode of subtracting an image after an open operation from an original image and enhancing the image) and Bottom-Hat transform (a mode of subtracting an image after a close operation from an original image and enhancing the image) on an ocean image obtained after water-bank separation, the gray contrast of the ship target and an ocean background is enhanced, background noise is suppressed, and the ship target is highlighted, so as to determine a candidate area of the ship target, and further reduce a ship detection range, where the candidate area may be an area as small as possible including the highlighted ship target. Fig. 5 is an effect diagram (left) of background suppression of an image with complex ocean background noise after water bank separation and an effect diagram (right) of gray scale reconstruction after background suppression.
WTH(x)＝(f-fg)(x) (1)
BTH(x)＝(fg-f)(x) (2)
Wherein, the formula (1) is a Top-Hat transformation formula, the formula (2) is a Bottom-Hat transformation formula, f is image gray, fg is background gray, and x is a ship target.
In step S3, a suspected ship target is determined from the candidate area, and a ship target is extracted according to the personalized ship features. The suspected target of the ship is determined firstly, so that the complexity of personalized feature analysis can be reduced.
Referring to fig. 6, in one embodiment, the determining of the ship suspected target from the candidate area in step S3 further includes the steps of:
s31: performing Top-Hat transformation processing on the image of the ship target candidate region;
s32: carrying out gray morphological reconstruction on the image after the transformation processing;
s33: and performing threshold segmentation on the reconstructed image to determine the suspected ship target.
The threshold segmentation can be image binary segmentation, and whether the suspected ship target exists is judged through binary gray scale.
In one embodiment, with continued reference to fig. 6, the extracting of the optimal ship target from the ship personalized features in step S3 further comprises the steps of:
s34: counting the pixel area of each communication area aiming at the image segmented by the threshold, reserving the communication areas with the pixel areas within a target communication setting range, and removing the rest communication areas, wherein the target communication setting range is determined according to the size of a ship target;
s35: extracting the framework of the reserved communication area, calculating the length-width ratio of the framework, reserving the framework which accords with the set range of the target length-width ratio, and removing the rest frameworks, wherein the set range of the target length-width ratio is determined according to the length-width ratio of the target framework of the ship; target aspect ratio set ranges such as, but not limited to, less than 5;
s36: calculating the gray level change of the reserved framework, eliminating the frameworks which have unchanged gray levels and the framework lengths of which are greater than a preset length value, and reserving the rest target frameworks; a preset length value such as, but not limited to, greater than 30;
s37: morphological reconstruction is performed according to the target skeleton retained in the step S36 and the image segmented by the threshold value in the step S34, so that a ship target is extracted.
Referring to fig. 7, morphological expansion is performed on the image after threshold segmentation, connected regions with sizes which do not meet requirements are removed, skeleton extraction is performed on the connected regions which meet the requirements, aspect ratio and gray level change are calculated, false targets which do not meet the requirements are removed, the false alarm rate is reduced, and real ship targets are identified. The detection method has local adaptability, accurately extracts the ship target and reduces the false alarm rate of ship detection.
In one embodiment, in step S34, the morphological dilation is performed on the threshold-segmented image, and then the pixel area of each connected region is counted.
Although the present invention has been described with reference to the preferred embodiments, it is not intended to limit the scope of the claims, and those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention.
Claims (6)
1. A ship target detection method based on infrared remote sensing images is characterized by comprising the following steps:
s1: carrying out water-bank separation on the infrared remote sensing image, dividing the image into an ocean part, a land part and a sea-land mixed part according to an image entropy statistical result of the frequency in a set frequency range in an image frequency spectrum of the infrared remote sensing image, determining a coastline through edge extraction, and segmenting the sea area image containing a ship target; the step S1 includes the steps of:
s11: dividing the infrared remote sensing image into a plurality of image small blocks, performing frequency spectrum calculation on each image small block, calculating the image entropy of the frequency within a set frequency range, and performing accumulation statistics to obtain the total value of the image entropy of each image small block;
s12: comparing the total image entropy value of each image small block with a preset upper threshold value and a preset lower threshold value, judging the image small block of which the total image entropy value is smaller than the preset lower threshold value as an ocean part, and judging the image small block of which the image entropy is larger than the preset upper threshold value as a land part and the rest image small blocks as an ocean-land mixed part;
s13: extracting the boundary of sea-land change of the image small block as a seed point, performing region growth, drawing a sea-shore line to realize water-shore separation, and segmenting the sea area image containing the ship target;
s2: carrying out contrast enhancement processing on the sea area image obtained after water-bank separation, highlighting a ship target in the image, and selecting a candidate area of the ship target according to the highlighted ship target;
s3: and determining a suspected ship target from the candidate area, and extracting the ship target according to the personalized characteristics of the ship.
2. The infrared remote sensing image-based ship target detection method as claimed in claim 1, wherein the set frequency range is 4-20 Hz.
3. The method for detecting the ship target based on the infrared remote sensing image as claimed in claim 1, wherein in the step S2, the gray contrast between the ship target and the ocean background is enhanced by performing Top-Hat transformation and Bottom-Hat transformation on the sea area image obtained after the water-bank separation, so as to highlight the ship target.
4. The infrared remote sensing image-based ship target detection method as claimed in claim 1, wherein the step S3 of determining the ship suspected target from the candidate area comprises the steps of:
s31: performing Top-Hat transformation processing on the image of the ship target candidate region;
s32: carrying out gray morphological reconstruction on the image after the transformation processing;
s33: and performing threshold segmentation on the reconstructed image to determine the suspected ship target.
5. The method for detecting ship targets based on infrared remote sensing images as claimed in claim 4, wherein the step of extracting optimal ship targets according to the individual ship features in the step S3 comprises the following steps:
s34: counting the pixel area of each communication area aiming at the image segmented by the threshold, reserving the communication areas with the pixel areas within a target communication setting range, and removing the rest communication areas, wherein the target communication setting range is determined according to the size of a ship target;
s35: extracting the framework of the reserved communication area, calculating the length-width ratio of the framework, reserving the framework which accords with the set range of the target length-width ratio, and removing the rest frameworks, wherein the set range of the target length-width ratio is determined according to the length-width ratio of the target framework of the ship;
s36: calculating the gray level change of the reserved framework, eliminating the frameworks which have unchanged gray levels and the framework lengths of which are greater than a preset length value, and reserving the rest target frameworks;
s37: morphological reconstruction is performed according to the target skeleton retained in the step S36 and the image segmented by the threshold value in the step S34, so that a ship target is extracted.
6. The infrared remote sensing image-based ship target detection method as claimed in claim 5, wherein in step S34, morphological dilation is performed on the threshold segmented image, and then statistics is performed on pixel areas of the respective connected regions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611076345.0A CN106709426B (en) | 2016-11-29 | 2016-11-29 | Ship target detection method based on infrared remote sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611076345.0A CN106709426B (en) | 2016-11-29 | 2016-11-29 | Ship target detection method based on infrared remote sensing image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106709426A CN106709426A (en) | 2017-05-24 |
| CN106709426B true CN106709426B (en) | 2020-04-14 |
Family
ID=58934098
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611076345.0A Active CN106709426B (en) | 2016-11-29 | 2016-11-29 | Ship target detection method based on infrared remote sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106709426B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109255301A (en) * | 2018-08-14 | 2019-01-22 | 北京航空航天大学 | A kind of infrared remote sensing image naval vessel detection false-alarm elimination method based on FPGA |
| CN109583319B (en) * | 2018-11-07 | 2020-12-15 | 北京市遥感信息研究所 | Ship detection method based on sea surface infrared remote sensing image |
| CN109859178B (en) * | 2019-01-18 | 2020-11-03 | 北京航空航天大学 | FPGA-based infrared remote sensing image real-time target detection method |
| CN110109189B (en) * | 2019-05-17 | 2021-11-30 | 中国人民解放军92942部队 | Infrared control method and device for offshore target |
| CN110133668B (en) * | 2019-05-17 | 2020-12-08 | 中国人民解放军92942部队 | Infrared detection method and device for offshore targets |
| CN110728635B (en) * | 2019-09-10 | 2023-07-07 | 中国科学院上海技术物理研究所 | Contrast enhancement method for dark and weak target |
| CN111105419B (en) * | 2019-10-09 | 2023-04-18 | 中国船舶重工集团公司第七0九研究所 | Vehicle and ship detection method and device based on polarized SAR image |
| CN111178230B (en) * | 2019-12-26 | 2022-03-15 | 武汉大学 | Intelligent extraction method for river beach in remote sensing image |
| CN113192059A (en) * | 2021-05-24 | 2021-07-30 | 北京市遥感信息研究所 | On-orbit ship detection method based on infrared remote sensing image |
| CN115222758B (en) * | 2022-09-21 | 2023-01-10 | 北京九章星图科技有限公司 | Real-time detection method for ship moving target of low-resolution wide-area sequence remote sensing image |
| CN115661434B (en) * | 2022-10-17 | 2023-05-02 | 中国人民解放军61540部队 | Night marine ship automatic identification method, system, electronic equipment and medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104732215A (en) * | 2015-03-25 | 2015-06-24 | 广西大学 | Remote-sensing image coastline extracting method based on information vector machine |
| CN105335978A (en) * | 2015-10-28 | 2016-02-17 | 上海斐讯数据通信技术有限公司 | Image edge detection method and system |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663348B (en) * | 2012-03-21 | 2013-10-16 | 中国人民解放军国防科学技术大学 | Marine ship detection method in optical remote sensing image |
| CN104077777B (en) * | 2014-07-04 | 2017-01-11 | 中国科学院大学 | Sea surface vessel target detection method |
| CN105303526B (en) * | 2015-09-17 | 2017-11-07 | 哈尔滨工业大学 | A kind of Ship Target Detection method based on coastline data and spectrum analysis |
| CN105184804B (en) * | 2015-10-12 | 2018-10-12 | 上海海大技术转移有限公司 | Small targets detection in sea clutter method based on Airborne IR camera Aerial Images |
-
2016
- 2016-11-29 CN CN201611076345.0A patent/CN106709426B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104732215A (en) * | 2015-03-25 | 2015-06-24 | 广西大学 | Remote-sensing image coastline extracting method based on information vector machine |
| CN105335978A (en) * | 2015-10-28 | 2016-02-17 | 上海斐讯数据通信技术有限公司 | Image edge detection method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106709426A (en) | 2017-05-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106709426B (en) | Ship target detection method based on infrared remote sensing image | |
| CN109191432B (en) | Remote sensing image cloud detection method based on domain transformation filtering multi-scale decomposition | |
| CN107274401B (en) | High-resolution SAR image ship detection method based on visual attention mechanism | |
| CN106650812B (en) | A kind of urban water-body extracting method of satellite remote-sensing image | |
| CN103778627B (en) | A kind of marine site based on SAR image oil spilling detection method | |
| CN109598729B (en) | Ship target detection method based on SRM segmentation and hierarchical line segment characteristics | |
| CN109117802A (en) | Ship Detection towards large scene high score remote sensing image | |
| CN102676633A (en) | Method for automatically counting bacterial colonies | |
| CN110889843B (en) | SAR image ship target detection method based on maximum stable extremal region | |
| Hou et al. | SAR image ship detection based on visual attention model | |
| CN110147716B (en) | SAR image internal wave region detection method based on combination of frequency domain and airspace | |
| CN108335294A (en) | The power distribution room abnormality image-recognizing method of complex condition | |
| Han | Watershed segmentation algorithm based on morphological gradient reconstruction | |
| CN105184804A (en) | Sea surface small target detection method based on airborne infrared camera aerially-photographed image | |
| CN108038856B (en) | Infrared small target detection method based on improved multi-scale fractal enhancement | |
| CN109002777A (en) | A kind of infrared small target detection method towards complex scene | |
| Ren et al. | Effective SAR sea ice image segmentation and touch floe separation using a combined multi-stage approach | |
| CN110310263B (en) | SAR image residential area detection method based on significance analysis and background prior | |
| CN110111355B (en) | Moving vehicle tracking method capable of resisting strong shadow interference | |
| CN111767856A (en) | Infrared small target detection algorithm based on gray value statistical distribution model | |
| CN110853050A (en) | SAR image river segmentation method, device and medium | |
| Zhai et al. | Segmentation-based ship detection in harbor for SAR images | |
| CN110472472B (en) | Airport detection method and device based on SAR remote sensing image | |
| Baorong et al. | Research of the infrared ship target recognition technology based on the complex background | |
| Chang et al. | A novel ship detection method based on gabor feature |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 201109 Minhang District, Shanghai Road, No. 1777 spring Applicant after: Shanghai Spaceflight Institute of TT&C And Telecommunication Address before: 200080 Shanghai city Hongkou District street Xingang Tianbao Road No. 881 Applicant before: Shanghai Spaceflight Institute of TT&C And Telecommunication |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |