A method for detecting the yaw of a ship from optical remote sensing images

Abstract

The application discloses a kind of optical remote sensing image's ship yaw detection method, it is related to remote sensing technical field, comprising the following specific steps: S1: remote sensing image data acquisition;S2: image data processing;S3: route planning and import;S4: ship detection;S5: yaw early warning;S6: route correction.The optical remote sensing image's ship yaw detection method, first by multispectral scanning identification obtains multiple spectral bands of scanning image, and using data correction, image classification is carried out remote sensing image feature extraction, then after data conversion, data storage and database establishment operation data information is completely imported into detection equipment inside, then combined with the image data of navigation area realizes route planning, and completes route import, according to ship positioning information in the process of navigation, yaw prediction is realized by remote sensing image data comparison, and timely make early warning when ship has yaw tendency, then generate correction route to correct the actual route of ship route.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of remote sensing, in particular to a ship deviation detection method for optical remote sensing images. BACKGROUND

[0002] Optical remote sensing technology is a technology for obtaining target and environmental information at a distance in the ultraviolet to infrared optical band. An optical remote sensing system generally consists of a remote sensor, a remote sensing platform, information transmission and information processing equipment, etc. As an important earth observation technology, optical remote sensing images have the characteristics of high spectral, high spatial and high temporal resolution, and are mainly used in reconnaissance, surveillance, missile early warning and weather forecasting fields.

[0003] For example, the application with the application number CN202110677403.X discloses a ship automatic matching method based on optical satellite remote sensing images. Similar to the above-mentioned ship automatic matching method based on optical satellite remote sensing images, the following problems still exist: although the ship can be automatically matched, only ship detection and matching can be performed, when the ship driving route gradually deviates, early warning and intelligent intervention cannot be achieved, so that the effect of correcting the ship route cannot be achieved, and the detected ship has the risk of driving out of the navigation area.

[0004] Therefore, in view of the above, the existing structure and defects are studied and improved, and a ship deviation detection method for optical remote sensing images is proposed. SUMMARY

[0005] In view of the deficiencies of the prior art, the present application provides a ship deviation detection method for optical remote sensing images, which solves the problems raised in the background art.

[0006] To achieve the above purpose, the present application is realized by the following technical scheme, a ship deviation detection method for optical remote sensing images, comprising the following specific steps:

[0007] S1: remote sensing image data acquisition

[0008] S11: multi-spectral scanning: using spectroscopy and photoelectric technology to record and send the information (pixels) of the spectral reflection energy of several to dozens of wavebands at a scanned point at the same time, and forming a scanning image of the same waveband by scanning pixels to obtain a scanning image of multiple spectral bands;

[0009] S12: data correction: geometric correction of remote sensing images, accurate matching and superimposing of the same ground object elements in the images, maps or data sets with geometric accuracy obtained by different sensors, using images as basic data sets, correcting other images based on a scene image;

[0010] S13: Remote sensing image classification: using image segmentation technology to segment the corrected image, and according to the minimum unit segmentation of image classification, homogeneous image objects (polygon) are obtained to realize higher level remote sensing image classification and target feature extraction;

[0011] S2: Image data processing:

[0012] S21: Data conversion: using an image scanner to scan the polygon of the remote sensing image, and using the digital data obtained by scanning to convert the image data into digital data, realizing analog / digital (A / D) conversion;

[0013] S22: Data storage: storing the digital data of the remote sensing image on a general digital computer and a detection device that can read out the CCT general carrier;

[0014] S23: Database establishment: using a computer system to integrate and induce the remote sensing image digital data stored in the detection device, forming a remote sensing image database, and constantly improving and updating the internal data of the database.

[0015] S3: Route planning and import:

[0016] S31: Route planning: retrieve the river remote sensing image data in the database to obtain the relevant image data of the navigation area, and the computer system analyzes and calculates the image data according to the relevant algorithm to plan all feasible navigation routes;

[0017] S32: Route data import: import the route data of the navigation route of the navigation area into the system of the ship detection device;

[0018] S4: Ship detection:

[0019] S41: Determine the planned route: input the starting point and ending point of the navigation into the ship detection device before navigation, after the best route analysis and feasibility analysis of the system, the detection device will provide all feasible navigation routes in the recommended order according to the best route, and manually select one of them as the planned route;

[0020] S42: Ship positioning: the ship travels along the river according to the planned route, and in the navigation process, the ship is positioned in real time by the cloud satellite, and the real-time remote sensing image data of the ship and its surrounding area is collected by the remote sensing image collection device according to the positioning information;

[0021] S43: Image data comparison: the computer system processes the real-time remote sensing image data, retrieves the remote sensing image data in the database, matches them with the real-time remote sensing image data, and realizes remote sensing image data comparison by using computer algorithm, so as to obtain the size, direction, navigation direction and other information of the detected ship.

[0022] S5: yaw early warning:

[0023] S51: yaw prediction: the detection device predicts the actual navigation route of the navigation ship according to the collected information related to the navigation ship, and compares the actual route with the planned route to predict whether the ship has a yaw tendency;

[0024] S52: early warning: for the ship with yaw tendency, the detection device will warn the yaw through the voice broadcast system, and intelligently intervene before the ship yaw to avoid the ship from completely yawing and leaving the navigation area;

[0025] S6: route correction:

[0026] S61: deviation correction route planning: after the detection device feeds back the yaw information to the ship, the deviation degree of the ship from the planned route can be calculated and analyzed through the internal algorithm of the system, and the deviation correction route back to the planned route is planned;

[0027] S62: return to the planned route: the deviation correction route is fed back to the ship user through the voice broadcast system, the actual route of the ship is corrected, and the ship is prompted to return to the planned route again.

[0028] Further, the geometric correction in S12 includes orthographic correction, RPC model correction and polynomial model correction, and the orthographic correction is to select some ground control points on the photograph, use the DEM data of the original photograph and the tilt and projection difference of the corrected image, resample the image into an orthographic image, splice and inlay several orthographic images, and cut out the image within a certain range after color balance processing.

[0029] Further, the RPC model correction is to replace the complex strict physical correction model with the RPC model for remote sensing image processing.

[0030] Further, the polynomial model correction is to use a polynomial to express and calculate the coordinate relationship between the images before and after correction, which is mainly applied to the case where the ground objects are relatively flat.

[0031] Further, the segmentation basis of the image segmentation technology in S13 includes the spectral information of the ground object itself and the spatial information of the ground object, and the spatial information includes shape, texture, area, size and other elements.

[0032] Further, the remote sensing image data comparison in S42 includes the following steps: respectively rasterizing the matched remote sensing image data in the database and the real-time remote sensing image data, comparing the differences between the two groups of raster units, and combining the image data resolution to obtain the difference information of the two groups of image data.

[0033] Further, the voice broadcast system in S52 can convert the text information formed by the detection device into voice information.

[0034] The application provides a ship deviation detection method for optical remote sensing images, which has the following beneficial effects: the ship deviation detection method for optical remote sensing images first obtains scanning images of multiple spectral bands through multi-spectral scanning identification, and performs remote sensing image ground object extraction by using data correction and image classification, then completely imports data information into the detection device through the operations of data conversion, data storage and database establishment, then realizes route planning in combination with image data of a navigation area, and completes route import, and in the process of navigation, deviation prediction is realized by comparing remote sensing image data according to ship positioning information, and a warning is given in time when the ship has a deviation tendency, and a deviation correction route is generated to correct the actual route of the ship. DETAILED DESCRIPTION

[0035] The technical solutions in the embodiments of the application will be clearly and completely described in combination with specific embodiments of the application. Obviously, the described embodiments are only some of the embodiments of the application, but not all the embodiments.

[0036] A ship deviation detection method for optical remote sensing images, comprising the following specific steps:

[0037] S1: remote sensing image data acquisition

[0038] S11: multi-spectral scanning: information (pixels) of spectral reflection energy of several to dozens of wave bands on a scanned point is recorded and sent at the same time by using spectroscopy and photoelectric technology, a scanning image is formed by scanning image elements of the same wave band, and scanning images of multiple spectral bands are obtained;

[0039] S12: data correction: the same ground object elements in images, maps or data sets with geometric accuracy obtained by different sensors are accurately matched and superimposed on each other by performing geometric correction on the remote sensing image, and other images are corrected based on a scene image as a basic data set;

[0040] S13: remote sensing image classification: the corrected image is segmented by using image segmentation technology, and the same image objects (map patches) obtained by segmenting the minimum unit of image classification are used to realize higher-level remote sensing image classification and target ground object extraction;

[0041] S2: image data processing:

[0042] S21: data conversion: the map patches of the remote sensing image are scanned by using an image scanner, and digital data obtained in a scanning manner is used to convert the image data into digital data, so as to realize analog / digital (A / D) conversion;

[0043] S22: Data storage: store the digital data of remote sensing images on a general carrier such as CCT, which can be read by general digital computers and detection devices;

[0044] S23: Database establishment: use a computer system to integrate and summarize the remote sensing image digital data stored in the detection device, form a remote sensing image database, and continuously improve and update the internal data of the database.

[0045] S3: Route planning and import:

[0046] S31: Route planning: retrieve the river remote sensing image data in the database to obtain the relevant image data of the navigation area, and the computer system analyzes and calculates the image data according to the relevant algorithm to plan all feasible navigation routes;

[0047] S32: Route data import: import all route data of the navigation area navigation route into the system of the ship detection device;

[0048] S4: Ship detection:

[0049] S41: Determine the planned route: input the starting point and endpoint of the navigation into the ship detection device before navigation, after the system's best route analysis and feasibility analysis, the detection device will provide all feasible navigation routes in the recommended order according to the best route, and manually select one of them as the planned route;

[0050] S42: Ship positioning: the ship travels along the river according to the planned route, and in the navigation process, the ship is positioned in real time by the cloud satellite, and the real-time remote sensing image data of the ship and its surrounding area is collected in real time by the remote sensing image collection device according to the positioning information;

[0051] S43: Image data comparison: the computer system processes the real-time remote sensing image data, retrieves the remote sensing image data in the database, matches it with the real-time remote sensing image data, and uses computer algorithms to compare the remote sensing image data to obtain the size, direction, navigation direction, etc. Information of the detected ship;

[0052] S5: Deviation warning:

[0053] S51: Deviation prediction: the detection device predicts the actual navigation route of the ship according to the collected information, and compares the actual route with the planned route to determine whether the ship has a deviation tendency;

[0054] S52: Early warning: for ships with deviation tendency, the detection device will warn the deviation through the voice broadcast system, and intelligently intervene before the ship deviates to avoid it from completely deviating and leaving the navigation area.

[0055] S6: Course correction:

[0056] S61: Course correction planning: After the detection device feeds back the expected deviation information to the ship, the system internally calculates and analyzes the deviation of the ship from the planned course, and plans a correction course to return to the planned course.

[0057] S62: Return to the planned course: The correction course is fed back to the user of the ship through the voice broadcast system, and the actual course of the ship is corrected to make the ship return to the planned course again.

[0058] The geometric correction in S12 includes orthographic correction, RPC model correction and polynomial model correction, and the orthographic correction is achieved by selecting some ground control points on the photos, using the DEM data of the original photos and the tilt and projection difference of the corrected images, resampling the images into orthographic images, splicing and inlaying a plurality of orthographic images, and cutting out the images within a certain range after color balance processing.

[0059] The RPC model correction is to replace the complex strict physical correction model with the RPC model for remote sensing image processing.

[0060] The polynomial model correction is to express and calculate the coordinate relationship between the corresponding points of the images before and after correction by using a polynomial, and is mainly applied to the case where the ground objects are relatively flat.

[0061] The segmentation basis of the image segmentation technology in S13 includes the spectral information of the ground objects and the spatial information of the ground objects, and the spatial information includes shape, texture, area, size and other elements.

[0062] The remote sensing image data comparison in S42 includes the following steps: rasterizing the matched remote sensing image data in the database and the real-time remote sensing image data respectively, comparing the differences between the two groups of raster cells, and combining the image data resolution to obtain the difference information of the two groups of image data.

[0063] The voice broadcast system in S52 can convert the text information formed by the detection device into voice information.

[0064] The optical remote sensing image ship deviation detection method includes the following specific steps when used:

[0065] S1: Remote sensing image data acquisition

[0066] S11: Multispectral scanning: using spectroscopy and photoelectric technology to record and transmit the information of several to dozens of spectral reflection energies (pixels) at a scanned point, and forming a scanning image of the same waveband by combining the scanning pixels of the same waveband, to obtain a scanning image of multiple spectral bands;

[0067] S12: data correction: geometric correction is performed on the remote sensing image, the same ground feature elements in the images, maps or data sets with geometric accuracy obtained by different sensors are accurately matched and overlaid with each other, the image is used as the basic data set, other images are corrected based on one image, the geometric correction includes ortho correction, RPC model correction and polynomial model correction, and the ortho correction is performed by selecting some ground control points on the photograph, using the DEM data of the original photograph and the tilt and projection difference of the corrected image, resampling the image into an ortho image, splicing and inlaying a plurality of ortho images, and cutting out the image within a certain range after color balance processing; the RPC model correction is performed by using the RPC model to replace the complex strict physical correction model for remote sensing image processing; the polynomial model correction is performed by using a polynomial to express and calculate the coordinate relationship between the corresponding points of the images before and after correction, and is mainly applied to the case where the ground features are relatively flat;

[0068] S13: remote sensing image classification: the corrected image is segmented by using an image segmentation technology, and a homogenous image object (polygon) is obtained by segmenting the minimum unit of image classification, so as to realize higher level remote sensing image classification and target ground feature extraction, wherein the segmentation of the image segmentation technology in S13 is based on the spectral information of the ground features and the spatial information of the ground features, and the spatial information includes shape, texture, area, size and the like;

[0069] S2: image data processing:

[0070] S21: data conversion: the polygons of the remote sensing image are scanned by using an image scanner, and digital data is obtained in a scanning manner, so as to convert the image data into digital data and realize analog / digital (A / D) conversion;

[0071] S22: data storage: the digital data of the remote sensing image is converted and stored on a general carrier such as CCT which can be read by general digital computers and detection devices;

[0072] S23: database establishment: the remote sensing image digital data stored in the detection device is integrated and summarized by using a computer system, a remote sensing image database is formed, and the internal data of the database is constantly improved and updated.

[0073] S3: route planning and import:

[0074] S31: route planning: the remote sensing image data of the river channel in the database is called, the relevant image data of the navigable area is obtained, the computer system analyzes and calculates the image data according to the relevant algorithm, and all feasible navigation routes are planned;

[0075] S32: Route data import: Import all route data of the navigation route of the navigation area into the system of the ship detection device;

[0076] S4: Ship detection:

[0077] S41: Determine the planned route: Before sailing, input the starting point and endpoint of the sailing into the ship detection device. After the best route analysis and feasibility analysis of the system, the detection device will provide all feasible sailing routes in the recommended order according to the best route, and manually select one of them as the planned route;

[0078] S42: Ship positioning: The ship travels along the river according to the planned route. During the sailing process, the ship is positioned in real time by the cloud satellite, and the real-time remote sensing image data of the ship and its surrounding area is collected by the remote sensing image collection device according to the positioning information;

[0079] S43: Image data comparison: The computer system processes the real-time remote sensing image data, retrieves the remote sensing image data in the database, matches it with the real-time remote sensing image data, and uses computer algorithms to compare the remote sensing image data to obtain the size, direction, sailing direction, etc. Information of the detected ship. The remote sensing image data comparison in S42 includes the following steps: rasterizing the matched remote sensing image data in the database and real-time remote sensing image data, comparing the differences between the two sets of raster cells, and combining the image data resolution to obtain the difference information between the two sets of image data;

[0080] S5: Deviation warning:

[0081] S51: Deviation prediction: The detection device predicts the actual sailing route of the ship according to the collected information related to the sailing ship, and compares the actual route with the planned route to determine whether the ship has a deviation tendency;

[0082] S52: Early warning: For ships with deviation tendency, the detection device will warn the deviation through the voice broadcast system, and intelligently intervene before the ship deviates completely to avoid it from deviating completely and leaving the navigation area. The voice broadcast system in S52 can convert the text information formed by the detection device into voice information;

[0083] S6: Route correction:

[0084] S61: Correction route planning: After the detection device feeds back the predicted deviation information to the ship, it can calculate and analyze the deviation degree of the ship from the planned route through the internal algorithm of the system, and plan a correction route back to the planned route;

[0085] S62: Return to the planned route: the deviation route is fed back to the user of the ship through the voice broadcast system, the actual route of the ship is corrected, and the ship is prompted to return to the planned route again.

[0086] The ship deviation detection method of the optical remote sensing image first obtains a plurality of spectral band scanning images through multi-spectral scanning identification, and uses data correction and image classification to extract ground objects from the remote sensing image. Then, through the operations of data conversion, data storage and database establishment, the data information is completely imported into the detection equipment. Then, combined with the image data of the navigation area, the route planning is realized, and the route import is completed. According to the ship positioning information during the navigation process, the deviation prediction is realized by comparing the remote sensing image data, and the early warning is made in time when the ship deviates. Then, the correction route is generated to correct the actual route of the ship.

[0087] The above is only the preferred specific implementation of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can make equivalent replacement or change according to the technical solution and the inventive concept of the present application within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A method for detecting a yaw of a ship from optical remote sensing imagery, characterized in that, The method comprises the following specific steps: S1: remote sensing image data acquisition S11: multi-spectral scanning: using spectroscopy and photoelectric technology to record and send the information of spectral reflection energy of several to dozens of wave bands at a scanned point, i.e. a pixel, and to form a frame of scanning image by scanning pixels of the same wave band, and to obtain scanning images of multiple spectral wave bands; S12: data correction: geometric correction of remote sensing images, accurate matching and superimposing of the same ground object elements in images, maps or data sets with geometric accuracy obtained by different sensors, using images as basic data sets, correcting other images based on a scene image; S13: remote sensing image classification: using image segmentation technology to segment the corrected image, and obtaining homogeneous image objects, i.e. patches, based on the minimum unit of image classification segmentation, so as to realize higher level remote sensing image classification and target ground object extraction; S2: image data processing S21: data conversion: scanning the patches of remote sensing images by using an image scanner, and obtaining digital data in a scanning manner, so as to convert image data into digital data and realize analog / digital conversion; S22: data storage: storing the digital data of remote sensing images on a CCT universal carrier which can be read by general digital computers and detection equipment; S23: database establishment: using a computer system to integrate and summarize the remote sensing image digital data stored in the detection equipment, forming a remote sensing image database, and constantly improving and updating the internal data of the database; S3: route planning and import S31: route planning: calling the river remote sensing image data in the database, obtaining relevant image data of the navigation area, and analyzing and calculating the image data by the computer system according to relevant algorithms to plan all feasible navigation routes; S32: route data import: importing the route data of the navigation routes of the navigation area into the system of the ship detection equipment; S4: ship detection S41: determining the planned route: inputting the starting point and the ending point of the navigation into the ship detection equipment before navigation, after the best route analysis and feasibility analysis of the system, the detection equipment will provide all feasible navigation routes in the recommended order according to the best route, and a navigation route is manually selected and determined as the planned route; S42: ship positioning: the ship travels along the river according to the planned route, and in the navigation process, the ship is positioned in real time by the cloud satellite, and the real-time remote sensing image data of the ship and its surrounding area are collected in real time by the remote sensing image collection equipment according to the positioning information; S43: image data comparison: the computer system processes the real-time remote sensing image data, calls the remote sensing image data in the database, matches the real-time remote sensing image data with the remote sensing image data, and realizes remote sensing image data comparison by using computer algorithms, so as to obtain the size, direction and navigation direction information of the detected ship; S5: deviation warning S51: deviation prediction: the detection equipment predicts the actual navigation route of the navigation ship according to the collected information of the navigation ship, and compares the actual route with the planned route to predict whether the ship has a deviation tendency; S52: early warning broadcast: for the ship with the tendency of yaw, the detection device will warn the yaw through the voice broadcast system, and intelligently intervene before the ship yaw, so as to avoid the complete yaw and sailing out of the navigation area; S6: route correction: S61: correction route planning: after the detection device feeds back the yaw information to the ship, the deviation degree of the ship from the planned route can be calculated and analyzed through the internal algorithm of the system, and the correction route is planned to return to the planned route; S62: return to the planned route: the correction route is fed back to the ship user through the voice broadcast system, and the actual route of the ship is corrected to make the ship return to the planned route again.

2. The method for detecting the yaw of a ship from optical remote sensing images according to claim 1, characterized in that, The geometric correction in S12 includes orthographic correction, RPC model correction and polynomial model correction, and the orthographic correction is to select some ground control points on the photograph, use the DEM data of the original photograph and the tilt and projection difference of the corrected image, resample the image into an orthographic image, splice and inlay several orthographic images, and cut out the image in a certain range after color balance processing.

3. A method of detecting the yaw of a vessel from optical remote sensing imagery according to claim 2, characterized in that, The RPC model correction is to replace the complex strict physical correction model with the RPC model for remote sensing image processing.

4. The method of claim 2, wherein the method further comprises: The polynomial model correction is to express and calculate the coordinate relationship between the corresponding points of the images before and after correction by using a polynomial, mainly applied to the case of relatively flat ground objects.

5. The method of claim 1, wherein the method further comprises: The image segmentation technology in S13 is segmented according to the spectral information of the ground object itself and the spatial information of the ground object, and the spatial information includes shape, texture, area and size elements.

6. The method of claim 1, wherein the method further comprises: The remote sensing image data comparison in S42 includes the following steps: the matched remote sensing image data in the database and the real-time remote sensing image data are rasterized respectively, then the difference between the two groups of raster units is compared, and the difference information of the two groups of image data is obtained by combining the image data resolution.

7. The method of claim 1, wherein the method further comprises: The voice broadcast system in S52 can convert the text information formed by the detection device into voice information.