CN110782447A - Multi-motion ship target detection method based on earth static orbit satellite optical remote sensing image - Google Patents
Multi-motion ship target detection method based on earth static orbit satellite optical remote sensing image Download PDFInfo
- Publication number
- CN110782447A CN110782447A CN201911027653.8A CN201911027653A CN110782447A CN 110782447 A CN110782447 A CN 110782447A CN 201911027653 A CN201911027653 A CN 201911027653A CN 110782447 A CN110782447 A CN 110782447A
- Authority
- CN
- China
- Prior art keywords
- image
- remote sensing
- orbit satellite
- optical remote
- image sequence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000003287 optical effect Effects 0.000 title claims abstract description 30
- 238000001514 detection method Methods 0.000 title claims abstract description 12
- 230000003068 static effect Effects 0.000 title claims description 7
- 238000000034 method Methods 0.000 claims abstract description 39
- 238000012545 processing Methods 0.000 claims abstract description 23
- 238000012216 screening Methods 0.000 claims abstract description 9
- 230000003595 spectral effect Effects 0.000 claims abstract description 6
- 238000001228 spectrum Methods 0.000 claims description 32
- 238000001914 filtration Methods 0.000 claims description 10
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000012935 Averaging Methods 0.000 claims description 4
- 238000003672 processing method Methods 0.000 claims description 2
- 238000011426 transformation method Methods 0.000 claims description 2
- 230000002708 enhancing effect Effects 0.000 abstract description 2
- 230000000007 visual effect Effects 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000011248 coating agent Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/262—Analysis of motion using transform domain methods, e.g. Fourier domain methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
-
- 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
- G06T2207/20048—Transform domain processing
- G06T2207/20056—Discrete and fast Fourier transform, [DFT, FFT]
-
- 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
- G06T2207/20172—Image enhancement details
- G06T2207/20201—Motion blur correction
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Mathematical Physics (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a method for detecting a multi-motion ship target by using an optical remote sensing image of a geostationary orbit satellite. The method comprises the steps of inputting an optical remote sensing image sequence of the geostationary orbit satellite, cutting an image, denoising the image, enhancing the image, extracting a significant map by a spectral residual error method, binarizing the image, performing expansion processing and screening a target, and can realize accurate and rapid detection of the trail of the marine multi-motion ship target by using the optical remote sensing image of the geostationary orbit satellite.
Description
Technical Field
The invention relates to the field of satellite remote sensing image detection, in particular to a multi-motion ship target detection method based on an earth static orbit satellite optical remote sensing image.
Background
The geostationary orbit optical remote sensing satellite is positioned about 36000km above the equator and has many characteristics which are not possessed by low orbit optical remote sensing satellites. Taking a high-resolution fourth satellite as an example, the high-resolution fourth satellite is a geostationary orbit earth observation remote sensing satellite emitted in 2015 years in China, and is used as a geosynchronous orbit optical remote sensing satellite with the highest resolution in the world at present, and has the following characteristics: 1) the device is positioned above the equator, so that the region in the coverage area can be continuously observed, and the data timeliness is good; 2) the single-spectrum shortest repeated imaging time is 5s, and the time resolution is high; 3) the single-scene image coverage area reaches 500km multiplied by 500km, and the observation range is wide; 4) the space resolution of the sub-satellite points of the carried area array staring camera is better than 50m multiplied by 50 m. The high-resolution fourth satellite can carry out comprehensive observation combining large-range real-time continuous maneuvering imaging and high-time resolution imaging to obtain dynamic change data of the region of interest.
Because the coating color of a large ship is usually darker, the optical remote sensing image of the geostationary orbit satellite is usually not high in resolution, for example, the resolution of the high-resolution four-numbered satellite remote sensing image is 50m, a ship target can not be seen in the geostationary orbit satellite remote sensing image usually, and only a trail formed by the ship in the motion process can be seen. At present, the mainstream optical remote sensing image ship target detection method mainly comprises a ship target detection algorithm based on gray level statistical characteristics, deep learning and a visual attention mechanism. The method based on the gray statistical characteristics utilizes the characteristic that the gray of the ship or the trail thereof is obviously larger than the gray of the sea surface to carry out detection; detecting by extracting the texture and geometric characteristics of the target based on a deep learning method; the method based on the visual attention mechanism extracts a visual saliency map through a visual saliency model, and then screens salient regions in the saliency map to extract regions of interest.
However, because the interference of dense cloud, islands and the like in the geostationary orbit satellite remote sensing image is more, the texture and geometric characteristics of the ship trail are few, and the ship trail continuously changes along with the movement speed, direction, wind power and wind direction of the ship, so that the method based on the gray level statistical characteristics and the deep learning is hardly applicable to the ship detection of the geostationary orbit satellite. The finding of a proper method for detecting the multi-motion ship target by using the optical remote sensing image of the geostationary orbit satellite is an urgent problem to be solved.
Disclosure of Invention
The invention provides a method for rapidly detecting a plurality of moving ship targets on the sea by using an optical remote sensing image of an earth static orbit satellite.
The method comprises the following steps:
step S101: inputting an optical remote sensing image sequence of a geostationary orbit satellite;
step S102: cutting the remote sensing image sequence according to the size of the attention area;
step S103: denoising the cut image sequence;
step S104: carrying out image enhancement on the image sequence subjected to denoising treatment;
step S105: extracting a saliency map from the image sequence after image enhancement by using a spectral residual error method;
step S106: carrying out binarization on the extracted image sequence saliency map;
step S107: performing expansion processing on the binarized image sequence saliency map;
step S108: and screening the image sequence after the expansion processing according to the geometric characteristics of the target.
In step S103, a median filtering method, a neighborhood averaging method, or a wiener filtering method may be used to perform denoising processing on the image sequence.
In step S104, the de-noised image sequence may be image enhanced by using a gray scale transformation method, a histogram processing method, or a non-linear gray scale stretching method.
In step S104, the two-dimensional discrete fourier transform is performed on the image data matrix after the image enhancement processing to obtain a frequency domain image sequence, a magnitude spectrum and a phase spectrum of the image are obtained from a frequency spectrum of the frequency domain image sequence, and a logarithm is taken from the magnitude spectrum to obtain a logarithm spectrum.
In step S105, the logarithmic spectrum obtained in step S104 is smoothed by a local average filter, and a spectrum residual is obtained.
And (4) performing two-dimensional discrete Fourier transform and Gaussian filtering on the spectrum residual in the step (105) and the phase spectrum obtained in the step (104).
In step S108, the target geometric features include: number of pixel points, length, width, aspect ratio.
According to the technical scheme, compared with the prior art, the method for detecting the multi-motion ship target of the optical remote sensing image of the geostationary orbit satellite comprises the steps of inputting the optical remote sensing image sequence of the geostationary orbit satellite, cutting the image, denoising the image, enhancing the image, extracting the saliency map by a spectral residual error method, binarizing the image, performing expansion processing and screening the target, accurately and quickly detecting the trail of the multi-motion ship target on the sea by using the optical remote sensing image of the geostationary orbit satellite, and being low in calculation complexity, simple in algorithm and easy to implement engineering.
Drawings
Fig. 1 is a flowchart of a method for detecting a marine multi-motion ship target based on an optical remote sensing image of an earth stationary orbit satellite according to an embodiment of the present invention.
Fig. 2 is a cut original optical remote sensing image of a high-resolution four-satellite according to an embodiment of the present invention.
Fig. 3 is a remote sensing image of fig. 2 after median filtering and nonlinear gray scale enhancement.
Fig. 4 is a saliency map of fig. 3 extracted by the spectral residual method.
Fig. 5 is a three-dimensional view of the saliency map of fig. 4 with a semi-transparent plane as the binarization threshold.
Fig. 6 is an image of fig. 4 after the binarization process.
Fig. 7 is an image of fig. 6 after the expansion process.
Fig. 8 is the image of fig. 7 after geometric feature screening.
Fig. 9 is an image of fig. 3 labeled results after screening.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to specific embodiments and the accompanying drawings. It should be noted that in the drawings or description, the same drawing reference numerals are used for similar or identical parts. Implementations not depicted or described in the drawings are of a form known to those of ordinary skill in the art. Additionally, while exemplifications of parameters including particular values may be provided herein, it is to be understood that the parameters need not be exactly equal to the respective values, but may be approximated to the respective values within acceptable error margins or design constraints. Directional phrases used in the embodiments, such as "upper," "lower," "front," "rear," "left," "right," and the like, refer only to the orientation of the figure. Accordingly, the directional terminology used is intended to be in the nature of words of description rather than of limitation.
The invention provides a method for detecting a multi-motion ship target by using an optical remote sensing image of a geostationary orbit satellite, wherein the embodiment takes the actual remote sensing image of a satellite with a high resolution of four as an example.
The method comprises the following steps:
step S101: inputting an optical remote sensing image sequence of the geostationary orbit satellite.
Step S102: cutting the remote sensing image sequence according to the size of the attention area;
as shown in fig. 2, the clipped original optical remote sensing image of the high-resolution four-size satellite is used for reducing subsequent data processing amount and improving data processing efficiency.
Step S103: denoising the cut image sequence;
the common denoising method for remote sensing images comprises the following steps: median filtering, neighborhood averaging, wiener filtering, and the like. The embodiment adopts a median filtering method, and has the advantages of effectively removing impulse noise and protecting image edges, as shown in fig. 3.
Step S104: carrying out image enhancement on the image sequence subjected to denoising treatment;
because the trail of the moving ship in the geostationary orbit satellite image belongs to a weak target, and the gray level is usually much smaller than that of cloud and land, the ratio of the trail of the moving ship to the sea background needs to be improved through image enhancement processing, and the improvement of the significance of the subsequent trail is facilitated. Commonly used image enhancement methods are: grayscale transformation, histogram processing, or non-linear grayscale stretching, etc. The embodiment adopts a nonlinear gray level stretching method, selects the stretching gravity center by utilizing the image gray level mean value, and has the characteristics of self-adapting stretching contrast ratio and effectively improving the target-background gray level ratio.
And carrying out image enhancement processing on the image sequence matrix E subjected to noise reduction processing to obtain an image data matrix G.
And (3) carrying out two-dimensional discrete Fourier transform on the data matrix G to obtain a frequency domain image sequence as shown in formula (1):
f(u，v，k)＝F(G(x，y，k))＝A(u，v，k)e
-jP(u，v，k)(1)
wherein F (G (x, y, k)) represents fourier transform. f (u, v, k) represents a value of a frequency spectrum coordinate (u, v) of a k-th image in the image sequence after Fourier transform, a magnitude spectrum A (u, v, k) and a phase spectrum P (u, v, k) of the image can be obtained from the frequency spectrum, and a logarithmic spectrum can be obtained by taking the logarithm of the magnitude spectrum as shown in formula (2):
L(u，v，k)＝log(A(u，v，k)) (2)
step S105: extracting a saliency map from the image sequence after image enhancement by using a spectral residual error method;
using a local averaging filter h
n(u, V, k) smoothing the log spectrum L (u, V, k) to obtain a general log spectrum V (u, V, k), and defining a difference between the log spectrum L (u, V, k) and the general log spectrum V (u, V, k) as a spectrum residual R (u, V, k), as shown in equations (3) and (4):
V(u，v，k)＝L(u，v，k)*h
n(u，v，k) (3)
R(u，v，k)＝L(u，v，k)-V(u，v，k) (4)
wherein, denotes convolution, h
n(u, v, k) is defined as an n mean filtered convolution kernel. And (3) performing two-dimensional inverse discrete Fourier transform and Gaussian filtering on the spectrum residual R (u, v, k) and the phase spectrum P (u, v, k), so as to reconstruct an image, wherein the significance of each pixel in the original image is represented by the formula (5):
S(x，y，k)＝g(x，y)*|F
-1(exp(R(u，v，k)+iP(u，v，k)))|
2(5)
wherein, F
-1(exp (R (u, v, k) + iP (u, v, k))) represents the inverse Fourier transform, and g (x, y) is a Gaussian filter, in order to improve the saliency effect of the image, as shown in FIG. 4, where the three-dimensional view is as shown in FIG. 5, where the semi-transparent plane is the binarization threshold.
Step S106: carrying out binarization on the extracted image sequence saliency map;
and (3) carrying out binarization processing on the extracted saliency map data matrix S:
wherein, B (x, y, k) represents the coordinate value of (x, y) of the k-th image after binarization processing in the saliency map sequence, F
THRepresenting the threshold of the binarization processing of the image sequence, and the binarized image is shown in fig. 6.
Step S107: performing expansion processing on the binarized image sequence saliency map;
and (3) performing expansion processing on the remote sensing image sequence subjected to binarization processing:
where r represents a set of all displacements of the dilated binarized image b (k) by the structural element T, and the dilated image is shown in fig. 7.
Step S108: and screening the image sequence after the expansion processing according to the geometric characteristics of the target to obtain the detection result of the multi-motion ship target.
Among the common geometric features are: pixel dot number, length, width, aspect ratio, and the like. In the embodiment, the characteristics such as the number of pixel points and the length-width ratio are adopted, and on the basis of considering the difference between the ship target and the debris clouds and the island reef, a proper characteristic is selected for screening, as shown in fig. 8.
As shown in fig. 9, the screening results in fig. 8 are labeled in fig. 3, and are the final detection results of the multi-motion ship target in this embodiment.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the present invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (7)
1. A multi-motion ship target detection method based on an earth static orbit satellite optical remote sensing image is characterized by comprising the following steps:
step S101: inputting an optical remote sensing image sequence of a geostationary orbit satellite;
step S102: cutting the remote sensing image sequence according to the size of the attention area;
step S103: denoising the cut image sequence;
step S104: carrying out image enhancement on the image sequence subjected to denoising treatment;
step S105: extracting a saliency map from the image sequence after image enhancement by using a spectral residual error method;
step S106: carrying out binarization on the extracted image sequence saliency map;
step S107: performing expansion processing on the binarized image sequence saliency map;
step S108: and screening the image sequence after the expansion processing according to the geometric characteristics of the target.
2. The method for detecting the multi-motion ship target by the earth static orbit satellite optical remote sensing image as claimed in claim 1, wherein in step S103, a median filtering method, a neighborhood averaging method or a wiener filtering method can be adopted to perform denoising processing on the image sequence.
3. The method for detecting the multi-motion ship target by the geostationary orbit satellite optical remote sensing image as recited in claim 1, wherein the de-noised image sequence can be image-enhanced by a gray scale transformation method, a histogram processing method or a non-linear gray scale stretching in step S104.
4. The method for detecting the multi-motion ship target by the earth static orbit satellite optical remote sensing image as recited in claim 1, wherein in step S104, the image data matrix after the image enhancement processing is subjected to two-dimensional discrete fourier transform to obtain a frequency domain image sequence, a magnitude spectrum and a phase spectrum of the image are obtained from a frequency spectrum of the frequency domain image sequence, and a logarithmic spectrum is obtained by logarithmizing the magnitude spectrum.
5. The method for detecting the multi-motion ship target by the earth-still-orbit satellite optical remote sensing image as recited in claim 4, wherein in step S105, the log spectrum obtained in step S104 is smoothed by a local average filter, and a spectrum residual is obtained.
6. The method for detecting the multi-motion ship target by the geostationary orbit satellite optical remote sensing image according to claim 5, wherein the two-dimensional inverse discrete Fourier transform and Gaussian filter are performed on the spectrum residual in the step S105 and the phase spectrum obtained in the step S104.
7. The method for detecting the multi-motion ship target by the earth-still-orbit satellite optical remote sensing image as recited in claim 1, wherein in step S108, the geometric features of the target comprise: number of pixel points, length, width, aspect ratio.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911027653.8A CN110782447A (en) | 2019-10-25 | 2019-10-25 | Multi-motion ship target detection method based on earth static orbit satellite optical remote sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911027653.8A CN110782447A (en) | 2019-10-25 | 2019-10-25 | Multi-motion ship target detection method based on earth static orbit satellite optical remote sensing image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110782447A true CN110782447A (en) | 2020-02-11 |
Family
ID=69386859
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911027653.8A Pending CN110782447A (en) | 2019-10-25 | 2019-10-25 | Multi-motion ship target detection method based on earth static orbit satellite optical remote sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110782447A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111967508A (en) * | 2020-07-31 | 2020-11-20 | 复旦大学 | Time series abnormal point detection method based on saliency map |
| CN112446440A (en) * | 2021-01-29 | 2021-03-05 | 江苏德劭信息科技有限公司 | Multi-sensor target tracking method of robot based on MSR-CNN |
| CN113112481A (en) * | 2021-04-16 | 2021-07-13 | 北京理工雷科电子信息技术有限公司 | Mixed heterogeneous on-chip architecture based on matrix network |
| CN115861839A (en) * | 2022-12-06 | 2023-03-28 | 平湖空间感知实验室科技有限公司 | Method and system for detecting weak and small target of geostationary orbit and electronic equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108805904A (en) * | 2018-05-25 | 2018-11-13 | 中国空间技术研究院 | A kind of moving ship detection and tracking based on satellite sequence image |
| CN106384344B (en) * | 2016-08-30 | 2019-02-22 | 中国科学院长春光学精密机械与物理研究所 | A kind of remote sensing image surface vessel target detection and extracting method |
| CN109815871A (en) * | 2019-01-16 | 2019-05-28 | 中国科学院电子学研究所 | The detection of target naval vessel and tracking based on remote sensing image |
| CN109816606A (en) * | 2019-01-18 | 2019-05-28 | 中国科学院电子学研究所 | A method of target following is carried out using Optical remote satellite |
-
2019
- 2019-10-25 CN CN201911027653.8A patent/CN110782447A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106384344B (en) * | 2016-08-30 | 2019-02-22 | 中国科学院长春光学精密机械与物理研究所 | A kind of remote sensing image surface vessel target detection and extracting method |
| CN108805904A (en) * | 2018-05-25 | 2018-11-13 | 中国空间技术研究院 | A kind of moving ship detection and tracking based on satellite sequence image |
| CN109815871A (en) * | 2019-01-16 | 2019-05-28 | 中国科学院电子学研究所 | The detection of target naval vessel and tracking based on remote sensing image |
| CN109816606A (en) * | 2019-01-18 | 2019-05-28 | 中国科学院电子学研究所 | A method of target following is carried out using Optical remote satellite |
Non-Patent Citations (1)
| Title |
|---|
| 冯冲: "光学遥感图像的舰船检测方法研究", 《中国优秀硕士学位论文信息科技辑》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111967508A (en) * | 2020-07-31 | 2020-11-20 | 复旦大学 | Time series abnormal point detection method based on saliency map |
| CN112446440A (en) * | 2021-01-29 | 2021-03-05 | 江苏德劭信息科技有限公司 | Multi-sensor target tracking method of robot based on MSR-CNN |
| CN112446440B (en) * | 2021-01-29 | 2021-04-16 | 江苏德劭信息科技有限公司 | Multi-sensor target tracking method of robot based on MSR-CNN |
| CN113112481A (en) * | 2021-04-16 | 2021-07-13 | 北京理工雷科电子信息技术有限公司 | Mixed heterogeneous on-chip architecture based on matrix network |
| CN113112481B (en) * | 2021-04-16 | 2023-11-17 | 北京理工雷科电子信息技术有限公司 | Hybrid heterogeneous on-chip architecture based on matrix network |
| CN115861839A (en) * | 2022-12-06 | 2023-03-28 | 平湖空间感知实验室科技有限公司 | Method and system for detecting weak and small target of geostationary orbit and electronic equipment |
| CN115861839B (en) * | 2022-12-06 | 2023-08-29 | 平湖空间感知实验室科技有限公司 | Weak and small target detection method and system for geostationary orbit and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110782447A (en) | Multi-motion ship target detection method based on earth static orbit satellite optical remote sensing image | |
| CN101661611B (en) | Realization method based on bayesian non-local mean filter | |
| Wang et al. | Generative adversarial network-based restoration of speckled SAR images | |
| CN104504652A (en) | Image denoising method capable of quickly and effectively retaining edge and directional characteristics | |
| CN104715474B (en) | High resolution synthetic aperture radar linearity building object detecting method based on Based On Method of Labeling Watershed Algorithm | |
| CN107705313A (en) | A kind of remote sensing images Ship Target dividing method | |
| CN107273903B (en) | UUV offshore visible light image sea-sky-line extraction method based on LSD improvement | |
| CN110059640A (en) | The in-orbit recognition methods of sea ship based on Optical remote satellite near-infrared spectral coverage | |
| CN104036461B (en) | A kind of Infrared Complex Background suppressing method based on Federated filter | |
| Pandey et al. | Enhancing the quality of satellite images using fuzzy inference system | |
| Ghircoias et al. | Contour lines extraction and reconstruction from topographic maps | |
| Liang et al. | Research on airport runway FOD detection algorithm based on texture segmentation | |
| Fawwaz et al. | The edge detection enhancement on satellite image using bilateral filter | |
| Cheon et al. | A modified steering kernel filter for AWGN removal based on kernel similarity | |
| Jiao et al. | Infrared dim small target detection method based on background prediction and high-order statistics | |
| Zhu et al. | Cloud removal for optical images using SAR structure data | |
| Ye et al. | On linear and nonlinear processing of underwater, ground, aerial and satellite images | |
| Stolojescu-Crisan et al. | Denoising and inpainting SONAR images | |
| Wan et al. | Removing thin cloud on single remote sensing image based on SWF | |
| Mittal et al. | Modified watershed segmentation with denoising of medical images | |
| Shi et al. | Ship targets detection based on visual attention | |
| Kaur et al. | Study of Image Enhancement Techniques in Image Processing: A | |
| Saied et al. | Digital Elevation Model Enhancement using CNN-Based Despeckled SAR Images | |
| McLaughlin et al. | Modified deconvolution using wavelet image fusion | |
| Wang et al. | Design of wavelet denoising and image enhancement algorithm based on MATLAB |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210707 Address after: 100190 No. 19 West North Fourth Ring Road, Haidian District, Beijing Applicant after: Aerospace Information Research Institute,Chinese Academy of Sciences Address before: 100190 No. 19 West North Fourth Ring Road, Haidian District, Beijing Applicant before: Institute of Electronics, Chinese Academy of Sciences |
|
| TA01 | Transfer of patent application right | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200211 |
|
| RJ01 | Rejection of invention patent application after publication |