CN111127506B - Marine moving target comprehensive detection method based on sequence image - Google Patents
Marine moving target comprehensive detection method based on sequence image Download PDFInfo
- Publication number
- CN111127506B CN111127506B CN201911137190.0A CN201911137190A CN111127506B CN 111127506 B CN111127506 B CN 111127506B CN 201911137190 A CN201911137190 A CN 201911137190A CN 111127506 B CN111127506 B CN 111127506B
- Authority
- CN
- China
- Prior art keywords
- target
- delta
- cloud
- white point
- cloud layer
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/207—Analysis of motion for motion estimation over a hierarchy of resolutions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention relates to a comprehensive detection method of a marine moving target based on a sequence image, belonging to the field of remote sensing satellite image processing and application; shooting the sea surface; judging whether a cloud layer exists in the single-frame image or not; step two, setting a cloud layer gradient detection function delta (delta); when delta (delta) is 1, entering a step III; when delta (delta) is 0, entering a step four; sequentially carrying out cloud removal, hole filling and target enhancement treatment on the image; step five, entering a step five; fourthly, shooting multi-frame continuous images, determining a target white point by adopting a clustering method, and carrying out cloud removal and target enhancement treatment on other interference target points; step five, when the target white point and the wake coexist, finishing target ship detection and determination of the target ship movement direction, and calculating the speed of the target ship; according to the invention, the movement direction and the movement state of the target are judged through multi-frame detection, so that the influence of most of marine cloud layers on the detection false alarm of the target is solved, and the detection probability is improved.
Description
Technical Field
The invention belongs to the field of remote sensing satellite image processing and application, and relates to a marine moving target comprehensive detection method based on a sequence image.
Background
CN201810513672.0 is a method for detecting and tracking a moving ship based on satellite sequence images. The method provided by the invention does not provide a solution to cloud interference, but various cloud interference is common in the sea surface optical image, the cloud-free opportunity is less, and the applicability of the method is restricted.
A method for quickly detecting sea surface motion ship of static orbit remote sensing satellite in volume 37 and 8 of 2015 of electronic and information journal. The document proposes a moving ship detection algorithm using sequential images, but the method does not mention a false alarm removal method in a cloud state.
CN201310256096.3 is a ship detection method of optical remote sensing image with cloud layer interference. The method is suitable for images with medium resolution (about 10 m) by taking geometric features as main judgment basis, but is oriented to high-orbit optical satellite images with 50m resolution, and the method is constrained due to the loss of the geometric features of the target.
A method for detecting targets of marine ships and warships resistant to crushing cloud interference is provided in the 12 th stage of the 32 nd volume of the 2010 of computer engineering and science. The method is characterized in that the radon transformation is used for detecting the ship wake so as to eliminate the false alarm of the broken cloud, but the ship wake is difficult to embody by a high-orbit optical satellite with low resolution, and the method is constrained.
A novel real-time detection method for a marine moving target is provided in the 14 th stage of volume 43 of computer engineering and application 2007, a method for detecting the target on an imaging sequence in a visible light range is provided, and a rapid anti-interference target detection is realized by using a deformed time difference method.
A ship detection method based on wavelet direction filtering and provided with cloud layer remote sensing images is provided in volume 48 and phase 1 of telecommunication technology 2008. The method for eliminating cloud interference and finally realizing target detection in the remote sensing image with cloud interference is not suitable for cloud breaking false alarm elimination by fusing wavelet decomposition of the image and directional filtering of the detection direction.
University of northeast newspaper: weak target detection algorithm research under cloud layer background based on OTSU segmentation in Nature science edition, 6 th month, 41 nd stage 2. The floating cloud interference in the background is removed by using the maximum inter-class variance OTSU segmentation algorithm, and the calculation is simple and convenient. But this method is not suitable for cloud-crushing false alarm rejection.
Disclosure of Invention
The invention solves the technical problems that: the method for comprehensively detecting the marine moving target based on the sequence images overcomes the defects of the prior art, judges the moving direction and the moving state of the target through multi-frame detection, solves the problem that most of marine cloud layers affect the target detection false alarm, and improves the detection probability.
The solution of the invention is as follows:
a comprehensive detection method of an offshore moving target based on a sequence image comprises the following steps:
shooting the sea surface through a high-orbit optical remote sensing satellite; judging whether a cloud layer exists in the single-frame image or not; entering a second step when a cloud layer exists; entering a fifth step when the cloud layer does not exist;
step two, setting a cloud layer gradient detection function delta (delta); judging the cloud layer condition according to the value of a cloud layer gradient detection function delta (delta); when the cloud layer gradient detection function delta (delta) is 1, entering a step III; when the cloud layer gradient detection function delta (delta) is 0, entering a step four;
sequentially carrying out cloud removal, hole filling and target enhancement treatment on the image; realizing the weakening of cloud layer; step five, entering a step five;
fourthly, shooting a plurality of frames of continuous images, and judging the motion directions of all target points according to the plurality of frames of images by adopting a clustering method; determining a target white point, and carrying out cloud removal and target enhancement treatment on the rest interference target points; step five, entering a step five;
step five, judging whether a target white point and a trail exist; when the target white point and the wake are simultaneously present, the detection of the target ship and the determination of the movement direction of the target ship are completed, and the speed of the target ship is calculated; otherwise, returning to the step one.
In the above-mentioned method for comprehensive detection of a marine moving object based on sequence images, in the second step, a cloud layer gradient detection function Δ (δ) is:
Δ(δ)＝Δ(X+)·Δ(X-)·Δ(Y+)·Δ(Y-)
wherein X is the X-direction coordinate of the detected pixel in the image;
y is the Y-direction coordinate of the detected pixel in the image;
delta (X+) and delta (X-) are the image gray scale gradient values of the pixel at two adjacent positions in the X direction;
delta (Y+) and delta (Y-) are the image gray scale gradient values of the pixel at two adjacent positions in the Y direction;
the cloud layer gradient detection function delta (delta) is calculated by the following steps:
setting a threshold value a according to experience; let Δ (δ) =1 when the cloud gradient detection function Δ (δ) is greater than a; otherwise Δ (δ) =0.
In the above-mentioned comprehensive detection method of marine moving target based on sequence image, in the second step, when the cloud layer gradient detection function delta (delta) is 1, the cloud layer is continuously distributed, and there is no interference item to the target white point and wake; when the cloud layer gradient detection function delta (delta) is 0, cloud layers are distributed in a cloud crushing mode, and the cloud crushing mode is an interference target point of target white points and trails.
In the above-mentioned method for comprehensively detecting the marine moving target based on the sequence image, in the fourth step, the specific method for determining the target white point is as follows:
judging the motion directions of all target points according to the multi-frame images; when the motion direction of one target point is inconsistent with that of the other target points; the target point is a target white point; the other target points are interference target points formed by the broken cloud.
In the above-mentioned comprehensive detection method for marine moving target based on sequence image, in the fifth step; the judgment method for the trail comprises the following steps:
s1, when a target white point and a wake in a spike shape exist and a wake tip points to the target white point; the target white point is the target ship; the pointed direction is the movement direction of the target ship;
s2, when a target white point and a linear-shaped tail trace exist and one end of the linear tail trace points to the target white point, the target white point is the target ship; the direction pointing to the target white point along the linear wake is the movement direction of the target ship.
In the above-mentioned comprehensive detection method for a marine moving object based on a sequence image, in the step S1, when the wake is in a spike shape, the included angle of the spike is 32 ° -39 °, and the track is judged to be valid.
In the above-mentioned method for comprehensive detection of a moving object at sea based on a sequence image, in the step S2, when the trail is in a linear shape, and the trail length is 3 times or more the target white point length, the trail is judged to be valid.
The above-mentioned comprehensive detection method of the marine moving target based on the sequence image, the calculation method of the target ship speed is: and (5) performing shape-advancing association calculation according to the continuous multi-frame images to obtain the speed of the target ship.
Compared with the prior art, the invention has the beneficial effects that:
(1) The invention adopts a brand new processing flow to meet the requirements of high orbit optical satellite sequence image ship target detection. The sequence image of the high orbit optical satellite is not a video image, the frame frequency is as low as 3-5 minutes/frame, in order to rapidly acquire the target motion state, only 5-10 frames are generally continuously acquired, a large number of samples cannot be provided, and the traditional video image moving target detection means cannot be effectively used. The high time resolution characteristic of the high orbit optical satellite is beneficial to realizing continuous tracking and monitoring of a sea surface moving target. The detection of the sea surface moving target can be realized through the effective processing of the sequence images of the same area;
(2) The invention improves the detection rate of the ship targets and reduces the false alarm rate caused by interference such as broken clouds and the like. Because of the longer distance from the earth and the poorer resolution of the high-orbit satellite, the state of the moving target on the sea surface is similar to that of broken clouds and reefs, and the false alarm rate of the common detection algorithm is high. The static and dynamic combined comprehensive treatment flow effectively reduces the occurrence of false alarms;
(3) The invention integrates various detection means such as single-frame image ship trail detection, target detection under the condition of reducing cloud layer interference by the sequence image, sequence image moving target detection and the like. The wake detection is used for effectively confirming the moving ship targets, the influence of cloud layers and broken clouds can be effectively reduced by the sequence image gradient change detection and clustering method, and the sequence image moving target detection can be used for predicting the state of the moving targets in an estimated mode;
(4) The method is used for processing the high orbit satellite sequence image, can be used for ship target detection in important areas, and provides important reference data for industries such as national defense construction, shipping management, fishery management and the like.
Drawings
FIG. 1 is a flow chart of the comprehensive detection of the moving object at sea;
FIG. 2 is a schematic diagram of the morphology of a moving ship target with a linear trail according to the present invention;
fig. 3 is a schematic diagram of the morphology of a sports ship target of the kelvin trail of the present invention.
Detailed Description
The invention is further illustrated below with reference to examples.
The invention provides a comprehensive detection method of an offshore moving target facing to a high-orbit optical satellite sequence image. The method considers the data characteristics of high orbit optical satellite sequence images, combines a mature static image ship detection means and an image processing technology, adopts a static-dynamic combination mode to improve the detection probability of sea surface moving targets, classifies cloud areas when sea surface cloud layer interference is processed, and reduces the false alarm rate. The method aims at the sequence image which is preprocessed, and preprocessing content comprises the following steps: geometric correction, radiation correction and pixel level registration between sequential images, sea-land segmentation, sea-island segmentation of each frame image. Wherein the radiation correction needs to take into account dynamic range adjustment problems in marine contexts in addition to correcting systematic errors; pixel level registration mainly considers eliminating imaging position errors caused by track perturbation and attitude shake in the imaging process.
As shown in fig. 1, the method for comprehensively detecting the marine moving target based on the sequence image mainly comprises the following steps:
shooting the sea surface through a high-orbit optical remote sensing satellite; judging whether a cloud layer exists in the single-frame image or not; entering a second step when a cloud layer exists; entering a fifth step when the cloud layer does not exist; because other types of wake are not obvious in the high-orbit optical remote sensing satellite, the motion direction of the moving target can be directly locked by only using a single frame image through wake detection, and then the moving target can be rapidly detected and motion state estimated by searching in the sequence image along the motion direction of the target.
And step two, sea surface moving object detection for eliminating cloud layer interference. For the area covered by the cloud layer, the cloud layer cannot be directly detected, and the cloud layer is classified first and then is respectively processed according to the characteristics of the cloud layer and the target. The specific flow is as follows: firstly, classifying cloud cover characteristics, wherein a region which is continuously changed in gradient and does not have a target bulge in a cloud range is a continuous thick cloud cover and certainly resembles a target region, and cloud removal and hole filling treatment is directly adopted for the region without target detection operation; the method comprises the steps that a region with continuous gradient change and target bulge is a region with continuous thick cloud and suspected target, and target enhancement and cloud layer inhibition treatment are adopted for the region; judging the overall movement direction of the crushed clouds by utilizing multi-frame images and adopting a clustering method, and selecting target points inconsistent with the overall movement direction of the clouds from the cluster as suspected targets; and for a single chip cloud (or independent suspected moving ship target) area, detecting whether the form and the gray level of the single chip cloud (or independent suspected moving ship target) area are changed by utilizing multi-frame images so as to eliminate suspected possibility. Setting a cloud layer gradient detection function delta (delta); judging the cloud layer condition according to the value of a cloud layer gradient detection function delta (delta); when the cloud layer gradient detection function delta (delta) is 1, entering a step III; when the cloud layer gradient detection function delta (delta) is 0, entering a step four; the cloud gradient detection function delta (delta) is:
Δ(δ)＝Δ(X+)·Δ(X-)·Δ(Y+)·Δ(Y-)
wherein X is the X-direction coordinate of the detected pixel in the image;
y is the Y-direction coordinate of the detected pixel in the image;
delta (X+) and delta (X-) are the image gray scale gradient values of the pixel at two adjacent positions in the X direction;
delta (Y+) and delta (Y-) are the image gray scale gradient values of the pixel at two adjacent positions in the Y direction;
the cloud layer gradient detection function delta (delta) is calculated by the following steps:
the detection threshold of delta (delta) can be determined by computational or machine learning or other equivalent means. If the calculation method is adopted, a certain numerical calculation method is utilized to determine a threshold value, and if the threshold value is defined as the 4 th power of the mean value of the gray gradient matrix of the image in the selected area; if the machine learning method is adopted, the support vector machine model can be learned based on the marked training sample with the intense/gentle change: in the test stage, determining a category with intense/gentle change by utilizing gradient characteristics of the model and the current sample, wherein the intensity of the change is higher than a threshold value; the gradual change is lower than the threshold value. Setting a threshold value a; let Δ (δ) =1 when the cloud gradient detection function Δ (δ) is greater than a; otherwise Δ (δ) =0. When the cloud layer gradient detection function delta (delta) is 1, the cloud layers are continuously distributed, and interference items on target white points and trails do not exist in the cloud layers; when the cloud layer gradient detection function delta (delta) is 0, cloud layers are distributed in a cloud crushing mode, and the cloud crushing mode is an interference target point of target white points and trails.
Sequentially carrying out cloud removal, hole filling and target enhancement treatment on the image; realizing the weakening of cloud layer; step five, entering a step five;
fourthly, shooting a plurality of frames of continuous images, and judging the motion directions of all target points according to the plurality of frames of images by adopting a clustering method; the specific method for determining the target white point comprises the following steps: judging the motion directions of all target points according to the multi-frame images; when the motion direction of one target point is inconsistent with that of the other target points; the target point is a target white point; if the motion direction of the target in the multi-frame image accords with the motion characteristics of a general ship (the motion direction between adjacent image frames is stable, the motion track has no folding angle larger than 90 degrees) and the brightness of the target in the image has no obvious change, the suspected target can be confirmed to be the moving ship target, and the track is marked. The other target points are interference target points formed by the broken cloud. Cloud removal and target enhancement processing are carried out on other interference target points; step five, entering a step five;
step five, judging whether a target white point and a trail exist; the judgment method for the trail comprises the following steps:
s1, when a target white point and a wake in a spike shape exist and a wake tip points to the target white point; the target white point is the target ship; the pointed direction is the movement direction of the target ship; when the wake is in a sharp wave shape, the included angle of the sharp wave is 32-39 degrees, and the track is judged to be effective, as shown in fig. 3.
S2, when a target white point and a linear-shaped tail trace exist and one end of the linear tail trace points to the target white point, the target white point is the target ship; the direction pointing to the target white point along the linear wake is the movement direction of the target ship. When the trace is in a linear shape, the trace length is 3 times or more the target white point length, and the trace is judged to be valid as shown in fig. 2. When the target white point and the wake are simultaneously present, the detection of the target ship and the determination of the movement direction of the target ship are completed, and the speed of the target ship is calculated; the calculation method of the target ship speed comprises the following steps: and (5) performing shape-advancing association calculation according to the continuous multi-frame images to obtain the speed of the target ship. Otherwise, returning to the step one.
Although the present invention has been described in terms of the preferred embodiments, it is not intended to be limited to the embodiments, and any person skilled in the art can make any possible variations and modifications to the technical solution of the present invention by using the methods and technical matters disclosed above without departing from the spirit and scope of the present invention, so any simple modifications, equivalent variations and modifications to the embodiments described above according to the technical matters of the present invention are within the scope of the technical matters of the present invention.
Claims (7)
1. A comprehensive detection method of an offshore moving target based on a sequence image is characterized in that: the method comprises the following steps:
shooting the sea surface through a high-orbit optical remote sensing satellite; judging whether a cloud layer exists in the single-frame image or not; entering a second step when a cloud layer exists; entering a fifth step when the cloud layer does not exist;
step two, setting a cloud layer gradient detection function delta (delta); judging the cloud layer condition according to the value of a cloud layer gradient detection function delta (delta); when the cloud layer gradient detection function delta (delta) is 1, entering a step III; when the cloud layer gradient detection function delta (delta) is 0, entering a step four;
the cloud gradient detection function delta (delta) is:
Δ(δ)＝Δ(X+)·Δ(X-)·Δ(Y+)·Δ(Y-)
wherein X is the X-direction coordinate of the detected pixel in the image;
y is the Y-direction coordinate of the detected pixel in the image;
delta (X+) and delta (X-) are the image gray scale gradient values of the pixel at two adjacent positions in the X direction;
delta (Y+) and delta (Y-) are the image gray scale gradient values of the pixel at two adjacent positions in the Y direction;
the cloud layer gradient detection function delta (delta) is calculated by the following steps:
setting a threshold value a according to experience; let Δ (δ) =1 when the cloud gradient detection function Δ (δ) is greater than a; otherwise Δ (δ) =0;
sequentially carrying out cloud removal, hole filling and target enhancement treatment on the image; realizing the weakening of cloud layer; step five, entering a step five;
fourthly, shooting a plurality of frames of continuous images, and judging the motion directions of all target points according to the plurality of frames of images by adopting a clustering method; determining a target white point, and carrying out cloud removal and target enhancement treatment on the rest interference target points; step five, entering a step five;
step five, judging whether a target white point and a trail exist; when the target white point and the wake are simultaneously present, the detection of the target ship and the determination of the movement direction of the target ship are completed, and the speed of the target ship is calculated; otherwise, returning to the step one.
2. The method for comprehensively detecting the marine moving target based on the sequence image according to claim 1, wherein the method comprises the following steps: in the second step, when the cloud layer gradient detection function delta (delta) is 1, the cloud layer is continuously distributed, and the cloud layer does not have interference items on target white points and trails; when the cloud layer gradient detection function delta (delta) is 0, cloud layers are distributed in a cloud crushing mode, and the cloud crushing mode is an interference target point of target white points and trails.
3. The method for comprehensively detecting the marine moving target based on the sequence image according to claim 2, wherein the method comprises the following steps: in the fourth step, the specific method for determining the target white point is as follows:
judging the motion directions of all target points according to the multi-frame images; when the motion direction of one target point is inconsistent with that of the other target points; the target point is a target white point; the other target points are interference target points formed by the broken cloud.
4. The method for comprehensively detecting the marine moving target based on the sequence image according to claim 3, wherein the method comprises the following steps of: in the fifth step; the judgment method for the trail comprises the following steps:
s1, when a target white point and a wake in a spike shape exist and a wake tip points to the target white point; the target white point is the target ship; the pointed direction is the movement direction of the target ship;
s2, when a target white point and a linear-shaped tail trace exist and one end of the linear tail trace points to the target white point, the target white point is the target ship; the direction pointing to the target white point along the linear wake is the movement direction of the target ship.
5. The method for comprehensively detecting the marine moving target based on the sequence image according to claim 4, wherein the method comprises the following steps: in the step S1, when the wake is in a sharp wave shape, the included angle of the sharp wave is 32-39 degrees, and the track is judged to be effective.
6. The method for comprehensively detecting the marine moving target based on the sequence image according to claim 5, wherein the method comprises the following steps: in the step S2, when the trail is in a linear shape, the trail length is 3 times or more the target white point length, and the trail is judged to be valid.
7. The method for comprehensively detecting the marine moving target based on the sequence image according to claim 6, wherein the method comprises the following steps: the calculation method of the target ship speed comprises the following steps: and (5) performing shape-advancing association calculation according to the continuous multi-frame images to obtain the speed of the target ship.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911137190.0A CN111127506B (en) | 2019-11-19 | 2019-11-19 | Marine moving target comprehensive detection method based on sequence image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911137190.0A CN111127506B (en) | 2019-11-19 | 2019-11-19 | Marine moving target comprehensive detection method based on sequence image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111127506A CN111127506A (en) | 2020-05-08 |
| CN111127506B true CN111127506B (en) | 2023-06-06 |
Family
ID=70495800
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911137190.0A Active CN111127506B (en) | 2019-11-19 | 2019-11-19 | Marine moving target comprehensive detection method based on sequence image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111127506B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113393571B (en) * | 2021-06-15 | 2023-06-27 | 北京市遥感信息研究所 | Cloud-free satellite image generation method and device |
| CN113393497B (en) * | 2021-07-07 | 2022-08-09 | 中国人民解放军海军航空大学 | Ship target tracking method, device and equipment of sequence remote sensing image under condition of broken clouds |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101246545B (en) * | 2008-02-22 | 2010-07-07 | 华南师范大学 | Possion method for removing cloud from optical remote sensing image |
| WO2016101279A1 (en) * | 2014-12-26 | 2016-06-30 | 中国海洋大学 | Quick detecting method for synthetic aperture radar image of ship target |
| CN109815871B (en) * | 2019-01-16 | 2021-11-16 | 中国科学院空天信息创新研究院 | Target ship detection and tracking method based on optical remote sensing image |
-
2019
- 2019-11-19 CN CN201911137190.0A patent/CN111127506B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN111127506A (en) | 2020-05-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kang et al. | A modified faster R-CNN based on CFAR algorithm for SAR ship detection | |
| Li et al. | Infrared maritime dim small target detection based on spatiotemporal cues and directional morphological filtering | |
| CN102279973B (en) | Sea-sky-line detection method based on high gradient key points | |
| Li et al. | Ship detection and tracking method for satellite video based on multiscale saliency and surrounding contrast analysis | |
| CN109859247B (en) | Near-ground scene infrared small target detection method | |
| CN109359661B (en) | Sentinel-1 radar image classification method based on convolutional neural network | |
| CN111127506B (en) | Marine moving target comprehensive detection method based on sequence image | |
| CN112487912B (en) | Arbitrary direction ship detection method based on improved YOLOv3 | |
| CN113920436A (en) | Remote sensing image marine vessel recognition system and method based on improved YOLOv4 algorithm | |
| CN108986130A (en) | A kind of method for detecting infrared puniness target under Sky background | |
| CN108596952A (en) | Fast deep based on candidate region screening learns Remote Sensing Target detection method | |
| CN108805902A (en) | A kind of space-time contextual target tracking of adaptive scale | |
| Zhang et al. | Nearshore vessel detection based on Scene-mask R-CNN in remote sensing image | |
| Xu et al. | Marine radar oil-spill monitoring through local adaptive thresholding | |
| Yildirim et al. | Ship detection in optical remote sensing images using YOLOv4 and Tiny YOLOv4 | |
| Zhou et al. | Automatic ship detection in SAR Image based on Multi-scale Faster R-CNN | |
| Sun et al. | Nsd-ssd: A novel real-time ship detector based on convolutional neural network in surveillance video | |
| Peng et al. | CourtNet: Dynamically balance the precision and recall rates in infrared small target detection | |
| CN111428624B (en) | Optical remote sensing image on-orbit ship target detection method | |
| Hashmani et al. | A survey on edge detection based recent marine horizon line detection methods and their applications | |
| Li et al. | Overview of deep convolutional neural network approaches for satellite remote sensing ship monitoring technology | |
| CN111275733A (en) | Method for realizing rapid tracking processing of multiple ships based on deep learning target detection technology | |
| Raj et al. | A novel Ship detection method from SAR image with reduced false alarm | |
| Jia et al. | A Fast Progressive Ship Detection Method for Very Large Full-Scene SAR Images | |
| CN113420594A (en) | SAR image ship detection method based on improved Faster R-CNN |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |