CN113205029B - Real-time detection method for airborne synthetic aperture radar sea surface ship - Google Patents
Real-time detection method for airborne synthetic aperture radar sea surface ship Download PDFInfo
- Publication number
- CN113205029B CN113205029B CN202110456528.XA CN202110456528A CN113205029B CN 113205029 B CN113205029 B CN 113205029B CN 202110456528 A CN202110456528 A CN 202110456528A CN 113205029 B CN113205029 B CN 113205029B
- Authority
- CN
- China
- Prior art keywords
- target
- image
- pixel
- detection
- detection window
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Abstract
The invention provides a real-time detection method for a sea surface ship by an airborne synthetic aperture radar, which comprises the following steps: respectively taking each line of pixel points of the DBS image containing the target as a center to construct a bar block diagram; setting a detection window, and detecting a block diagram through the detection window by taking a pixel point as a center to obtain a detection window diagram; respectively calculating the pixel gray distribution characteristics of the detection window image and the strip block image, and detecting a target pixel point according to the difference of the pixel gray distribution characteristics of the detection window image and the strip block image; and searching target pixel points to obtain a target contour range. The real-time detection method for the sea surface ship of the airborne synthetic aperture radar disclosed by the invention has the advantages of high data processing speed, no manual intervention in the whole process, adaptability to the real-time detection of multi-scale targets and weak ship targets and high real-time detection rate of the targets, and meets the actual application requirements of real-time detection of the sea surface ship targets of the airborne SAR.
Description
Technical Field
The invention relates to the technical field of electronic detection, in particular to a real-time detection method for an airborne synthetic aperture radar sea surface ship.
Background
The sea surface ship target is a key target for marine monitoring, and ship detection has important significance for marine traffic, fishery monitoring and national defense safety. Compared with the application of the satellite-borne SAR, the airborne SAR has wide application prospect in daily sea area patrol and emergency monitoring for dealing with sudden events due to the characteristics of strong data timeliness, flexible operation, high resolution and the like.
However, the development of the current airborne SAR real-time detection technology in the field of marine applications is still relatively delayed due to the following reasons: 1) The SAR system and the image interpretation technology thereof have strong specialization; 2) The software and hardware conditions of an aircraft platform, a radar system, a measurement and control system, even an operation airspace and the like required by the development of airborne SAR data acquisition and real-time application are difficult to guarantee. Therefore, the current SAR-based target detection application of sea vessels still focuses on target detection for the conventional SAR image afterwards, i.e., relevant detection work is carried out by utilizing an onboard or satellite-borne SAR image.
The ship information obtained by adopting a non-real-time after-event detection method is often lack of timeliness, can only be used for marine ship information statistics and analysis application, is difficult to be used for real-time monitoring of marine ships or guiding field enforcement, and does not have the advantage of full-time and all-weather monitoring of SAR. In addition, a moving target detection mode based on an airborne SAR image or an airborne MMTI mode cannot simultaneously consider the targets of a static ship and a moving ship, a large number of targets can be omitted in a detection result, the traditional airborne SAR image has relatively narrow swath and limited sea area coverage capacity, detection based on the traditional SAR image is difficult to support large-range and wide-area target detection and search, and the defects limit popularization and application of the SAR technology.
Disclosure of Invention
Technical problem to be solved
Aiming at the technical problems in the prior art, the invention provides a real-time detection method for a sea surface ship of an airborne synthetic aperture radar, which is used for at least partially solving the technical problems.
(II) technical scheme
The invention provides a real-time detection method for a sea surface ship by an airborne synthetic aperture radar, which comprises the following steps: respectively taking each line of pixel points of the DBS image containing the target as a center to construct a bar block diagram; setting a detection window, taking a pixel point as a center, and obtaining a detection window image through a detection bar block image of the detection window; respectively calculating the pixel gray distribution characteristic values of the DBS image, the detection window image and the bar block image; calculating according to the DBS image and the pixel gray distribution characteristic value of the bar block image to obtain a target detection threshold, wherein the target detection threshold is set in a self-adaptive mode according to the DBS image and the bar block image corresponding to the image line where the detected pixel point is located; detecting a target pixel point according to the difference between the pixel gray level distribution characteristic value of the detection window image and a target detection threshold value; searching a target pixel point to obtain a target contour range; wherein, setting the detection window includes: calculating the pixel gray level distribution characteristics of the DBS image to obtain a first characteristic value; and adaptively setting the size of the detection window according to the first characteristic value.
Optionally, respectively centering on each line of pixel points of the DBS image containing the target, constructing the bar graph includes: and constructing a bar block diagram with the height less than or equal to the dimension of the detection window and the width equal to the width of the DBS image.
Optionally, detecting a target pixel point according to a difference between a pixel gray scale distribution characteristic value of the detection window image and a target detection threshold includes: calculating the pixel gray distribution characteristic of the detection window image to obtain a second characteristic value; and according to the second characteristic value and the target detection threshold, performing target judgment on each pixel point to obtain a detection binary image, wherein the gray value of the pixel point in the detection binary image is a first gray value or a second gray value.
Optionally, the setting the scale of the target detection window according to the first feature value includes: and setting the scale of the target detection window to be an odd number of pixel units, wherein the scale of the target detection window is positively correlated with the first characteristic value.
Optionally, performing target judgment on each pixel point according to the second characteristic value and the target detection threshold, and obtaining a detection binary image includes: calculating the pixel gray distribution characteristic of a block image corresponding to the image row where the current detected pixel point is located to obtain a third characteristic value; obtaining a target detection threshold corresponding to the image row where the current detected pixel point is located according to the first characteristic value and the third characteristic value, wherein the target detection threshold is greater than the third characteristic value, and the ratio of the target detection threshold to the third characteristic value is a constant; comparing the second characteristic value with the target detection threshold, wherein pixel points corresponding to the second characteristic value which is greater than the target detection threshold are target pixel points, and pixel points corresponding to the second characteristic value which is less than or equal to the target detection threshold are clutter backgrounds; and setting the gray value of the target pixel point as a first gray value, and setting the gray value of the pixel point of the clutter background as a second gray value to obtain a detection binary image.
Optionally, searching for a target pixel point, and obtaining a target contour range includes: searching target pixel points point by point, and acquiring the minimum external rectangle of the target pixel points; and expanding the four sides of the minimum external rectangle until any one side does not contain other target pixel points, wherein the expanded external rectangle forms a target outline range.
Optionally, setting a detection window, centering on a pixel point, and detecting the bar graph through the detection window includes: and sliding the detection window along the transverse direction or the longitudinal direction to detect the bar block diagram.
Optionally, the calculating the pixel gray scale distribution characteristic of the DBS image includes: and calculating the pixel gray scale distribution characteristics of the sea area DBS image and/or calculating the pixel gray scale distribution characteristics of the land area DBS image and/or calculating the pixel gray scale distribution characteristics of the airspace DBS image.
Optionally, the method for real-time detection of sea-surface vessels by airborne synthetic aperture radar further comprises: and carrying out false alarm filtering on the target contour, and rejecting targets with the difference with a preset target scale larger than a target filtering threshold value.
(III) advantageous effects
The invention provides a real-time detection method for an airborne synthetic aperture radar sea surface ship. The real-time detection method for the airborne synthetic aperture radar sea surface ship disclosed by the invention has the advantages of high data processing speed and no manual intervention in the whole process, and meets the actual application requirement of real-time detection of the airborne SAR sea surface ship target.
The invention adopts a method for comparing image pixel gray distribution characteristics capable of quantitatively describing image information complexity, the measurement of the image information accords with the visual sense of human eyes, and simultaneously, the method is sensitive to the catastrophe points in the image and does not depend on the prior knowledge of target scale and the like, so that the method can meet the detection of multi-scale ship targets and weak ship targets under the background of certain complexity on the premise of ensuring the real-time property. The method has strong adaptability in the aspect of real-time detection of multi-scale targets and targets of weak ships, and the real-time detection rate of the targets is high.
According to the invention, the traditional SAR image is replaced by the DBS image of the airborne SAR, the characteristic that the DBS image can image the static and moving ship targets simultaneously is fully considered, and the method is more effective particularly for the moving ship target, so that the consideration treatment of the sea surface static and moving ship targets is realized. The problem that a large number of targets are missed in the detection result is avoided.
Compared with the traditional airborne SAR image, the airborne DBS image adopted by the invention has lower resolution, and under the condition of the same image scale, the sea area covered by the DBS image is far larger than that covered by the SAR image, so that the ship target detection based on the DBS image can support large-range and wide-area target detection and search.
Drawings
The above and other objects, features and advantages of the present invention will become more apparent from the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:
fig. 1 schematically shows an application scenario of a real-time detection method for sea-surface vessels by airborne synthetic aperture radar according to an embodiment of the present invention;
FIG. 2 schematically illustrates a flow chart of a method for real-time detection of a sea-surface vessel by an airborne synthetic aperture radar in accordance with an embodiment of the present invention;
FIG. 3 schematically illustrates a target sliding window detection scheme according to an embodiment of the invention;
FIG. 4 is a flow chart of a method for obtaining the size of a detection window and a bar graph according to an embodiment of the present invention;
FIG. 5 is a flow chart of a method for obtaining a target pixel according to an embodiment of the present invention;
FIG. 6 is a flow chart that schematically illustrates a target contour range acquisition method, in accordance with an embodiment of the present invention;
fig. 7 schematically shows a flow chart of a method for real-time detection of a sea-surface vessel by an airborne synthetic aperture radar according to another embodiment of the invention.
[ description of reference ]
100-DBS images
110-first moving vessel
120-second motion vessel
130-first stationary vessel
140-second stationary vessel
310-bar graph
320-center pixel
330-detection Window
340-center pixel row
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. Features of the embodiments illustrated in the description may be freely combined to form new embodiments without conflict, and each claim may be individually referred to as an embodiment or features of the claims may be combined to form a new embodiment, and in the drawings, the shape or thickness of the embodiment may be enlarged and simplified or conveniently indicated. Further, elements or implementations not shown or described in the drawings are of a form known to those of ordinary skill in the art. Additionally, although examples may be provided herein of parameters including particular values, it should be appreciated that the parameters need not be exactly equal to the respective values, but may approximate the respective values within acceptable error margins or design constraints.
Unless a technical obstacle or contradiction exists, the above-described various embodiments of the present invention may be freely combined to form further embodiments, which are within the scope of the present invention.
Although the present invention has been described in connection with the accompanying drawings, the embodiments disclosed in the drawings are intended to be illustrative of preferred embodiments of the present invention and should not be construed as limiting the invention. The dimensional proportions in the figures are merely schematic and are not to be understood as limiting the invention.
Although a few embodiments of the present general inventive concept have been shown and described, it would be appreciated by those skilled in the art that changes may be made in these embodiments without departing from the principles and spirit of the general inventive concept, the scope of which is defined in the claims and their equivalents.
Fig. 1 schematically shows an application scenario of the method for detecting a sea vessel by using an airborne synthetic aperture radar according to an embodiment of the present invention in real time. It should be noted that fig. 1 is only an example of an application scenario in which the embodiment of the present invention may be applied to help those skilled in the art understand the technical content of the present invention, and does not mean that the embodiment of the present invention may not be applied to other devices, systems, environments or scenarios.
According to an embodiment of the present invention, as shown in fig. 1, the method for real-time detection of sea-surface vessels by an airborne synthetic aperture radar according to an embodiment of the present invention may be used for real-time detection of sea-surface vessel targets of an airborne SAR system, for example, where the DBS image 100 includes 4 targets, which are the first moving vessel 110, the second moving vessel 120, and the first stationary vessel 130 and the second stationary vessel 140, respectively. The vessels may or may not be the same or similar in size, and the DBS image may be, for example, an 8bit grayscale image. The invention can ensure the real-time processing of each target data in the image by taking the airborne SAR Doppler Beam Sharpening (DBS) image as the detection input so as to achieve the real-time monitoring of various targets on the sea surface.
It can be understood that the detection method of the present invention is not limited to detecting sea targets, and when the method is applied to real-time detection of land targets or airspace targets, the airborne SAR sea DBS image in the solution of the present invention may be replaced by land DBS images or airspace DBS images. When the method is applied to the detection of the non-real-time SAR sea vessel target, the airborne SAR sea area DBS image in the scheme of the invention can be replaced by the traditional SAR image.
In order to realize the real-time monitoring of various targets on the sea surface, the invention designs an airborne synthetic aperture radar sea surface ship real-time detection method shown in fig. 2.
According to an embodiment of the present invention, as shown in fig. 2, the method for real-time detection of a sea-surface vessel by an airborne synthetic aperture radar includes:
s210, respectively taking each line of pixel points of the DBS image containing the target as a center, and constructing a bar block diagram.
S220, setting a detection window, taking the pixel point as the center, and obtaining a detection window image through a detection window detection bar block image.
S230, respectively calculating the pixel gray distribution characteristics of the DBS image, the detection window image and the bar block image, calculating a target detection threshold value of the image line of the current detected pixel point according to the pixel gray distribution characteristic values of the bar block image corresponding to the DBS image and the image line of the current detected pixel point, adaptively setting the target detection threshold value according to the DBS image and the current bar block image, and detecting the target pixel point according to the pixel gray distribution characteristic value of the detection window image and the target detection threshold value of the current image line.
S240, searching the target pixel point to obtain a target contour range.
Fig. 3 schematically shows a target sliding window detection scheme according to an embodiment of the invention.
According to the embodiment of the invention, the pixel gray distribution characteristics of the DBS image can be represented by processing the global variance weighting information entropy of the generated and input single sea area DBS image in real time through an onboard SAR system, and the scale sizes of a target detection window and a bar graph of a ship are adaptively set according to the calculated global variance weighting information entropy. As shown in fig. 3, for example, the width and height of the DBS image containing the target vessel are W and H, respectively, and the target detection window and the height of the bar graph may be set to be the same. The detection window scale and the size of the bar graph may be calculated according to steps S410 to S420 in fig. 4.
And S410, statistically calculating the global variance weighting information entropy of the whole DBS image by using formulas (1) to (3) according to the DBS image input in real time and the width and height information thereof.
P s ＝N s /N (3)
Wherein E is the global variance weighting information entropy of the DBS image, s is the gray level of the image, and the range of the value range of s is, for example, 0 ≦ s ≦ 255;
is the mean value of the gray levels of the entire image, s i Is the gray level of the ith pixel in the figure, and N is the total pixel number of the image, and the value size is W multiplied by H shown in figure 3; p s As the probability of the s-th gray level appearing in the entire image, N s Is the number of pixels in the image at that level of gray scale.
S420, based on the calculated global variance weighting information entropy E, the dimension (unit: pixel) of the detection window of the image is set according to the formula (4) in a self-adaptive mode.
The detection window is, for example, a square detection window, R is a side length of the square detection window, and E is a global variance weighting information entropy of the DBS image. The side length of the square detection window may be, for example, 5 pixel units when the global variance weighting information entropy is, for example, less than 5000, and may be, for example, 9 pixel units when the global variance weighting information entropy is, for example, greater than or equal to 5000 and, at the same time, less than or equal to 10000, and may be, for example, 13 pixel units when the global variance weighting information entropy is, for example, greater than 10000.
It can be understood that, in the detection method according to the embodiment of the present invention, since the tile map is created with each row of pixel points as a center, and for example, the target detection window is set to have the same height as the tile map, and the upper and lower rows of pixel points of each row need to be detected, the scale of the target detection window may be set to be an odd number of pixel units, where the scale of the target detection window is positively correlated with the value of the global variance weighting information entropy of the DBS image. The values in equation (4) are merely examples and do not limit the present invention to use other values of pixel units and global variance weighting information entropy values for DBS images. When the first line or the last line of pixel points is taken as the center, because no other pixel points are arranged above or below the line of pixel points, and when a similar DBS image is detected with a bar block diagram with the upper and lower boundaries less than R/2, namely, the upper part (upper boundary) or the lower part (lower boundary) of the image center line can be used for the condition that the number of image lines is insufficient for statistics, the statistical calculation can be carried out according to the actual number of lines.
Fig. 5 schematically shows a flowchart of a method for acquiring a target pixel point according to an embodiment of the present invention.
According to the embodiment of the present invention, as shown in fig. 5, for example, the gray scale distribution characteristics of the pixels of the detection window and the bar graph may be compared through steps S510 to S520, so as to obtain the target pixel point. For example, the local variance weighting information entropy of a bar graph having a set sliding window size as high and a set width as wide as a sliding window size is calculated on a DBS image line by line, and a target detection threshold corresponding to each image line is adaptively set based on the calculated value.
Specifically, in step S510, according to formulas (1) to (3), the local variance weighting information entropy E of the bar block graph with the current image line l as the center line, the detection window dimension R as the height, and the width W of the DBS image as the width is calculated line by line from top to bottom in the DBS graph l . Wherein l is the number of image line, l is more than or equal to 1 and less than or equal to H.
S520, weighting information entropy E based on local variance of each block image obtained by calculation l The detection threshold Th corresponding to the image line l is adaptively calculated according to the formula (5) and the formula (6) l 。
Th l ＝k×E l (5)
k＝1000/E+1.0 (6)
Wherein Th l For detecting the threshold, k is a weight coefficient, E is a global variance weighting information entropy of the DBS image, and E is l The information entropy is weighted for the variance of the patch map.
It can be understood that for higher processing speed requirements, the block map local variance weighting information entropy used for adaptively calculating the image line detection threshold in the scheme of the present invention can be replaced by the image global variance weighting information entropy. This replacement can omit the process of calculating the local variance weighting information entropy of the slice block image line by line, greatly shorten the image processing time, but also reduce the detection performance.
Fig. 6 schematically shows a flowchart of a target contour range acquisition method according to an embodiment of the present invention.
According to the embodiment of the present invention, as shown in fig. 6, for example, the target pixel point may be searched through steps S610 to S620 to obtain the target contour range. For example, on the DBS image, a line-by-line pixel-by-pixel sliding scanning detection method is adopted, a set detection window is slid, the local variance weighted information entropy of the image in the sliding window with each pixel as the center is calculated, and compared with the threshold of the line where the local variance weighted information entropy is located, if the local variance weighted information entropy is larger than the threshold, the target is determined, otherwise, the target is a clutter, the whole image processing is completed, and a target detection binary image is formed.
Specifically, S610 uses a square detection window with adaptively set R as a side length, performs sliding processing on a DBS image as shown in fig. 3 through a sliding window with pixel by pixel and line by line detection order from left to right and top to bottom, and calculates, for each pixel, a local variance weighting information entropy E of a DBS image block covered by the detection window with a current pixel as a center according to formulas (1) to (3) i 。
It can be understood that, when the sliding window is used for the local variance weighting information entropy calculation in the solution of the present invention, the sliding manner is not limited to sliding horizontally row by row and pixel by pixel in an image row manner, and may also be replaced by vertical, for example, in an image column manner.
S620, comparing the local variance weighting information entropy E i And a detection threshold Th l Size of (E) i Greater than Th l The pixel point of (2) is a target pixel point, for example, the pixel value is set to 255 (white), E i Less than or equal to Th l The pixel of (2) is a clutter background pixel, for example, the pixel value is set to 0 (black). After the comparison of the pixel points of the whole DBS image is completed, a frame based on the DBS image can be obtainedAnd (4) carrying out variance weighted information entropy on the ship target detection binary image.
According to the embodiment of the invention, based on the target detection binary image, for example, a method of expanding search based on the target point (with a pixel value of 255) can be adopted to determine the scale range of each ship target in the image. For example, a line-by-line and point-by-point searching method may be adopted on the target detection binary image to sequentially search for white target points in the image. And aiming at the searched target point, taking the current target point as the center, taking a rectangle formed by 8 adjacent pixel points surrounding the point as an expansion basis, respectively expanding four sides by a pixel distance in four directions of up, down, left and right, detecting whether the four sides of the new rectangle contain a new white target point after expansion, continuing to expand the sides of the rectangle containing the new target point outwards, and stopping expansion until the four sides are expanded. The rectangle formed by the method is the minimum external rectangle of the target, and the pixel point coordinates of the upper left corner and the lower right corner of the rectangle are recorded. And searching and expanding all target pixel points, and obtaining the contour range of all targets.
It can be understood that, when the target point is searched and expanded by adopting a line-by-line and point-by-point searching method in the embodiment of the present invention, the target area which has been expanded is removed in the subsequent search, so as to avoid repeated processing.
According to the embodiment of the invention, after the external rectangles of the targets are obtained, the external rectangles of the preset ship can be subjected to scale-based false alarm filtering within a preset scale range of the target, the targets which are obviously inconsistent in scale are removed, and the pixel point coordinates of the external rectangles at the upper left point and the lower right point in the DBS image are sequentially output aiming at the reserved targets, so that the final target contour range is obtained. Filtering out the target which is not in scale conformity based on a preset ship target scale range, wherein the preset ship target scale range can be a ship target scale range which is expected to be detected, and comprises a minimum scale and a maximum scale. In practical applications, the target dimension can be converted to an image pixel dimension according to the resolution of the DBS image.
Fig. 7 schematically shows a flow chart of a method for real-time detection of a sea-surface vessel by an airborne synthetic aperture radar according to another embodiment of the invention.
According to an embodiment of the present invention, taking real-time detection of an onboard SAR marine surface vessel target as an example, as shown in fig. 7, a real-time detection method for an onboard SAR marine surface vessel target includes, for example:
and S710, calculating the global variance weighting information entropy of the sea area DBS image generated by the airborne SAR system in real time, and adaptively setting the dimension of the target detection window of the square ship according to the calculated value.
S720, calculating the local variance weighted information entropy of the bar block graph with the current image line as the center, the set target detection window as the height and the DBS image width as the width line by line, and adaptively setting the target detection threshold of each image line based on the calculated value.
And S730, sliding on the DBS image by using a target detection window in a line-by-line and pixel-by-pixel sliding scanning detection mode, calculating the local variance weighting information entropy of the image in the sliding window with each pixel as the center, comparing the local variance weighting information entropy with the detection threshold of the image line where the pixel is located, judging the local variance weighting information entropy as a target when the local variance weighting information entropy is larger than the threshold, and judging the local variance weighting information entropy as a clutter if the local variance weighting information entropy is not larger than the threshold until the whole image is processed to form a detection binary image.
And S740, performing pixel-by-pixel target boundary definition processing on the detected binary image by adopting a target point external expansion searching boundary method, and recording target boundary pixel coordinates.
And S750, filtering out the non-conforming targets based on a preset ship target scale range, and outputting boundary coordinate information conforming to the targets.
For other details of the detection method according to the another embodiment, reference may be made to the description of the above embodiment, and details are not repeated herein.
In summary, the embodiment of the present invention provides a method for real-time detection of a sea-surface vessel by an airborne synthetic aperture radar. On the basis of image domain detection, a real-time wide area detection method with small operand, high processing speed and high sensitivity to weak and small targets is provided, a scheme of local parameter self-adaption and sliding window processing based on a variance weighted information entropy algorithm is designed, the real-time performance and the self-adaption of data processing in the ship target detection processing process are guaranteed, and the real-time automation of the airborne SAR sea surface ship detection processing process is realized. In the process design, the application requirements of simultaneously monitoring a static target and a moving target in practical application are fully considered, a DBS (direct sequence database) graph which has less calculation amount and high generation speed and can simultaneously reflect the static target and the moving target in an image is taken as input, and the real-time processing of data and the comprehensiveness of a detection object are ensured; in the aspect of algorithm design, a variance weighted information entropy method is used as a method for detecting the ship target in the DBS diagram, the advantages of small calculated amount, flexible algorithm and high target sensitivity are fully exerted, local detection parameters of the target are adaptively set by using global and local statistical information of the DBS diagram, and the adaptivity and result robustness in the process of detecting the ship target on the sea surface are improved.
It should be understood that the specific order or hierarchy of steps in the processes disclosed is an example of exemplary approaches. Based upon design preferences, it is understood that the specific order or hierarchy of steps in the processes may be rearranged without departing from the scope of the present disclosure. The accompanying method claims present elements of the various steps in a sample order, and are not intended to be limited to the specific order or hierarchy.
In the foregoing detailed description, various features are grouped together in a single embodiment for the purpose of streamlining the disclosure. This method of disclosure is not to be interpreted as reflecting an intention that the claimed embodiments of the subject matter require more features than are expressly recited in each claim. Rather, as the following claims reflect, invention lies in less than all features of a single disclosed embodiment. Thus, the following claims are hereby expressly incorporated into the detailed description, with each claim standing on its own as a separate preferred embodiment of the invention.
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. Furthermore, to the extent that the term "includes" is used in either the detailed description or the claims, such term is intended to be inclusive in a manner similar to the term "comprising" as "comprising" is interpreted when employed as a transitional word in a claim. Furthermore, any use of the term "or" in the specification of the claims is intended to mean a "non-exclusive or".
Claims (9)
1. A real-time detection method for an airborne synthetic aperture radar sea surface vessel is characterized by comprising the following steps:
respectively taking each line of pixel points of the DBS image containing the target as a center to construct a bar block diagram;
setting a detection window, and detecting the bar block image through the detection window by taking a pixel point as a center to obtain a detection window image;
respectively calculating the pixel gray distribution characteristic values of the DBS image, the detection window image and the bar block image;
calculating according to the DBS image and the pixel gray distribution characteristic value of the bar block image to obtain a target detection threshold, wherein the target detection threshold is set in a self-adaptive mode according to the DBS image and the bar block image;
detecting a target pixel point according to the difference between the pixel gray distribution characteristic value of the detection window image and the target detection threshold value;
searching the target pixel point to obtain a target contour range;
wherein the setting the detection window includes:
calculating the pixel gray scale distribution characteristics of the DBS image to obtain a first characteristic value;
and adaptively setting the scale of the detection window according to the first characteristic value.
2. The method for real-time vessel inspection by airborne synthetic aperture radar according to claim 1, wherein the step of constructing a bar graph with each line of pixel points of the DBS image containing the target as a center comprises:
and constructing the bar block graph with the height less than or equal to the dimension of the detection window and the width equal to the width of the DBS image.
3. The method of claim 1, wherein the detecting a target pixel point according to the difference between the pixel gray scale distribution characteristic value of the detection window map and the target detection threshold value comprises:
calculating the pixel gray distribution characteristic of the detection window image to obtain a second characteristic value;
and performing target judgment on each pixel point according to the second characteristic value and the target detection threshold value to obtain a detection binary image, wherein the gray value of the pixel point in the detection binary image is a first gray value or a second gray value.
4. The method of claim 1, wherein the scaling the target detection window according to the first characteristic value comprises:
setting the scale of the target detection window to be an odd number of pixel units, wherein the scale of the target detection window is positively correlated with the first characteristic value.
5. The method for detecting sea-surface vessels by using the airborne synthetic aperture radar according to claim 3, wherein the step of performing target judgment on each pixel point according to the second characteristic value and the target detection threshold value to obtain a detection binary image comprises:
calculating pixel gray level distribution characteristics of the bar block image corresponding to the image row where the current detected pixel point is located to obtain a third characteristic value;
obtaining the target detection threshold corresponding to the image line where the current detected pixel point is located according to the first characteristic value and the third characteristic value, wherein the target detection threshold is greater than the third characteristic value, and the ratio of the target detection threshold to the third characteristic value is a constant;
comparing the second characteristic value with the target detection threshold, wherein pixel points corresponding to the second characteristic value which is greater than the target detection threshold are the target pixel points, and pixel points corresponding to the second characteristic value which is less than or equal to the target detection threshold are clutter backgrounds;
and setting the gray value of the target pixel point as the first gray value, and setting the gray value of the pixel point of the clutter background as the second gray value to obtain the detection binary image.
6. The method of claim 1, wherein the searching the target pixel points to obtain a target profile range comprises:
searching the target pixel point by point to obtain a minimum external rectangle of the target pixel point;
and expanding the four sides of the minimum external rectangle until any one side does not contain other target pixel points, wherein the expanded external rectangle forms the target outline range.
7. The method of claim 1, wherein the setting of the detection window, centering on a pixel point, and the detecting the bar graph through the detection window comprises:
and sliding the detection window along the transverse direction or the longitudinal direction to detect the bar block diagram.
8. The method for real-time sea-surface vessel detection by airborne synthetic aperture radar according to claim 1, wherein said calculating the pixel gray scale distribution characteristics of the DBS image comprises:
and calculating the pixel gray scale distribution characteristics of the sea DBS image and/or calculating the pixel gray scale distribution characteristics of the land DBS image and/or calculating the pixel gray scale distribution characteristics of the spatial DBS image.
9. The method of claim 1, further comprising:
and carrying out false alarm filtering on the target contour range, and rejecting targets with the difference with a preset target scale larger than a target filtering threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110456528.XA CN113205029B (en) | 2021-04-26 | 2021-04-26 | Real-time detection method for airborne synthetic aperture radar sea surface ship |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110456528.XA CN113205029B (en) | 2021-04-26 | 2021-04-26 | Real-time detection method for airborne synthetic aperture radar sea surface ship |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113205029A CN113205029A (en) | 2021-08-03 |
| CN113205029B true CN113205029B (en) | 2023-04-07 |
Family
ID=77028796
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110456528.XA Active CN113205029B (en) | 2021-04-26 | 2021-04-26 | Real-time detection method for airborne synthetic aperture radar sea surface ship |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113205029B (en) |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009144754A1 (en) * | 2008-05-29 | 2009-12-03 | Telespazio S.P.A. | Target detection in a sar-imaged sea area |
| CN101750883B (en) * | 2008-12-11 | 2011-02-09 | 北京大学 | Method and device for detecting angular point of screened image |
| CN105787902B (en) * | 2016-03-22 | 2018-11-27 | 天津大学 | Utilize the image denoising method of block sorting detection noise |
| CN110490848B (en) * | 2019-08-02 | 2022-09-30 | 上海海事大学 | Infrared target detection method, device and computer storage medium |
-
2021
- 2021-04-26 CN CN202110456528.XA patent/CN113205029B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113205029A (en) | 2021-08-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chen et al. | Ship detection from coastal surveillance videos via an ensemble Canny-Gaussian-morphology framework | |
| Ai et al. | A new CFAR ship detection algorithm based on 2-D joint log-normal distribution in SAR images | |
| CN102279973B (en) | Sea-sky-line detection method based on high gradient key points | |
| AU2012384518B2 (en) | Infrared image based early detection of oil spills in water | |
| CN109427055B (en) | Remote sensing image sea surface ship detection method based on visual attention mechanism and information entropy | |
| CN111666854B (en) | High-resolution SAR image vehicle target detection method fusing statistical significance | |
| CN112766221B (en) | Ship direction and position multitasking-based SAR image ship target detection method | |
| Espedal | Detection of oil spill and natural film in the marine environment by spaceborne SAR | |
| Wang et al. | Superpixel-based LCM detector for faint ships hidden in strong noise background SAR imagery | |
| Zhao et al. | SAR ship detection based on end-to-end morphological feature pyramid network | |
| Kong et al. | Lightweight algorithm for multi-scale ship detection based on high-resolution SAR images | |
| CN112435249B (en) | Dynamic small target detection method based on circumferential scanning infrared search system | |
| CN113205029B (en) | Real-time detection method for airborne synthetic aperture radar sea surface ship | |
| CN113205139A (en) | Unmanned ship water sphere detection method based on density clustering | |
| CN112101250A (en) | Method for detecting offshore ship target based on context semantic perception | |
| Yang et al. | A novel ship detection method based on sea state analysis from optical imagery | |
| CN116343057A (en) | Ship target detection and identification method combining SAR (synthetic aperture radar) with optical image | |
| CN116403114A (en) | Remote sensing image target recognition method and system based on deep learning | |
| CN114879180A (en) | Seamless situation perception method for real-time fusion of unmanned ship-borne multi-element multi-scale radar | |
| CN112686222B (en) | Method and system for detecting ship target by satellite-borne visible light detector | |
| Pappas et al. | Superpixel-guided CFAR detection of ships at sea in SAR imagery | |
| CN114373135A (en) | Ship target detection method based on local significance characteristic measurement | |
| Wu et al. | Ship recognition from high resolution remote sensing imagery aided by spatial relationship | |
| Smith et al. | The statistical characterization of the sea for the segmentation of maritime images | |
| CN113033577B (en) | Marine target feature extraction method based on variance correction model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |