Registration Method, Device and Electronic Equipment for Remote Sensing Images

Abstract

The present invention relates to the technical field of image processing, and provides a registration method, device and electronic device for remote sensing images. The method includes: selecting a target sampling rate from multiple reference image sampling rates, performing resampling processing on the downsampled reference image to obtain a resampled reference image; performing feature point matching between the resampled reference image and the downsampled remote sensing image to obtain multiple pairs of matching points; adjusting the coordinates of the multiple pairs of matching points to obtain multiple pairs of adjusted matching points; adjusting the resolution of the resampled reference image to be consistent with the image with unclear features according to the mean value of the distance ratios between the multiple pairs of adjusted matching points to obtain a standard reference image; performing feature point matching between the image with unclear features and the standard reference image; and performing image geometric correction on the image with unclear features according to the feature point matching result. Through the above method, the registration accuracy of remote sensing images with unclear features is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular, to a method, device, and electronic device for registering remote sensing images. Background Art

[0002] An image with unclear features refers to a remote sensing image in which areas with unclear features of ground objects such as deserts and farmlands in the remote sensing image account for a relatively large proportion. In the related art, the Scale-invariant Feature Transform (SIFT) and its improved methods have rotation and scale invariance, and have been widely used in the automatic registration of remote sensing images, achieving good results. However, for images with unclear features, the SIFT and its improved methods obtain relatively few matching points, or even no matching points, which cannot meet the actual usage requirements. Summary of the Invention

[0003] The present invention provides a method, device, and electronic device for registering remote sensing images, which are used to solve the problem that relatively few or even no matching points are obtained in the feature point matching process of an image with unclear features in the prior art, and improve the registration accuracy of the remote sensing image with unclear features.

[0004] The present invention provides a method for registering remote sensing images, including:

[0005] Performing downsampling processing on the image with unclear features and the reference image respectively to obtain a downsampled remote sensing image and a downsampled reference image;

[0006] Selecting a target sampling rate from multiple reference image sampling rates, performing resampling processing on the downsampled reference image to obtain a resampled reference image; and performing feature point matching between the resampled reference image and the downsampled remote sensing image to obtain multiple pairs of matching points;

[0007] Adjusting the coordinates of multiple pairs of matching points according to the sampling rate of the downsampling processing and the sampling rate of the resampling processing to obtain multiple pairs of adjusted matching points;

[0008] Adjusting the resolution of the resampled reference image to be consistent with that of the image with unclear features according to the mean value of the distance ratios between multiple pairs of adjusted matching points to obtain a standard reference image;

[0009] Performing feature point matching between the image with unclear features and the standard reference image; and performing image geometric correction on the image with unclear features according to the feature point matching result.

[0010] According to the method for registering remote sensing images provided by the present invention, it further includes:

[0011] Obtaining the resolution difference value between the resampled reference image and the downsampled remote sensing image;

[0012] If the resolution difference value is greater than or equal to the resolution difference threshold, a target sampling rate is reselected from multiple reference image sampling rates, and resampling processing is performed on the downsampled reference image.

[0013] According to the remote sensing image registration method provided by the present invention, obtaining the resolution difference value between the resampled reference image and the downsampled remote sensing image includes: obtaining the resolution difference value according to the average value of the distance ratios between multiple pairs of matching points.

[0014] According to the remote sensing image registration method provided by the present invention, it further includes:

[0015] According to a preset sampling rate interval, multiple sampling rates are generated between the maximum sampling rate and the minimum sampling rate of the reference image to obtain multiple reference image sampling rates.

[0016] According to the remote sensing image registration method provided by the present invention, performing downsampling processing on the featureless image and the reference image respectively to obtain a downsampled remote sensing image and a downsampled reference image specifically includes:

[0017] Obtain the image width and image height of the featureless image;

[0018] According to the image width and the image height, as well as a preset downsampled image width and a preset downsampled image height, obtain the sampling rate for downsampling processing;

[0019] According to the sampling rate for downsampling processing, perform downsampling processing on the featureless image and the reference image respectively to obtain a downsampled remote sensing image and a downsampled reference image.

[0020] According to any one of the above remote sensing image registration methods provided by the present invention, performing feature point matching on the resampled reference image and the downsampled remote sensing image to obtain multiple pairs of matching points includes:

[0021] Obtain multiple feature points of the downsampled remote sensing image;

[0022] Obtain a preset range area centered on each feature point to obtain a template window corresponding to each feature point; and set the resampled reference image as the search window;

[0023] According to the template window and the search window, perform gray-scale template matching on the resampled reference image and the downsampled remote sensing image to obtain multiple pairs of matching points.

[0024] According to the remote sensing image registration method provided by the present invention, it further includes:

[0025] If the feature point matching fails, a target sampling rate is reselected from multiple reference image sampling rates, and resampling processing is performed on the downsampled reference image.

[0026] The present invention also provides a registration device for remote sensing images, including:

[0027] A downsampling processing module, configured to perform downsampling processing on the featureless image and the reference image respectively to obtain a downsampled remote sensing image and a downsampled reference image;

[0028] A first matching module, configured to select a target sampling rate from multiple reference image sampling rates, perform resampling processing on the downsampled reference image to obtain a resampled reference image; and perform feature point matching between the resampled reference image and the downsampled remote sensing image to obtain multiple pairs of matching points;

[0029] A coordinate adjustment module, configured to adjust the coordinates of multiple pairs of matching points according to the sampling rate of the downsampling processing and the sampling rate of the resampling processing to obtain multiple pairs of adjusted matching points;

[0030] An image adjustment module, configured to adjust the resolution of the resampled reference image to be consistent with that of the featureless image according to the average value of the distance ratios between multiple pairs of adjusted matching points to obtain a standard reference image;

[0031] A second matching module, configured to perform feature point matching between the featureless image and the standard reference image; and perform image geometric correction on the featureless image according to the feature point matching result.

[0032] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where when the processor executes the computer program, the registration method for remote sensing images as described in any one of the above is implemented.

[0033] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the registration method for remote sensing images as described in any one of the above is implemented.

[0034] The present invention also provides a computer program product, including a computer program, and when the computer program is executed by a processor, the registration method for remote sensing images as described in any one of the above is implemented.

[0035] The remote sensing image registration method, device, and electronic device provided by the present invention perform downsampling on the featureless image and the reference image, so as to reduce the data volume of feature point matching and improve the feature point matching efficiency in the process of subsequently selecting a target sampling rate from multiple reference image sampling rates to obtain the resampled reference image. Based on the sampling rates in the downsampling and resampling processes, multiple pairs of adjusted matching points are obtained to adjust the resolution of the reference image, and a standard reference image with the same resolution as the featureless image is obtained. This enables the acquisition of a standard reference image that can achieve better feature point matching results for the featureless image when the resolution of the featureless image is uncertain. It not only increases the number of matching points obtained by the featureless image in the feature point matching process but also improves the registration accuracy of the featureless image. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] To more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0037] Figure 1 is one of the schematic flowcharts of the remote sensing image registration method provided by the present invention.

[0038] Figure 2 is another schematic flowchart of the remote sensing image registration method provided by the present invention.

[0039] Figure 3 is yet another schematic flowchart of the remote sensing image registration method provided by the present invention.

[0040] Figure 4 is the schematic structural diagram of the remote sensing image registration device provided by the present invention.

[0041] Figure 5 is the schematic structural diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0042] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention in conjunction with the drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in the present invention fall within the scope of protection of the present invention.

[0043] In the description and claims of this application and the above-mentioned drawings, terms such as "first" and "second" are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that such descriptions can be interchanged under appropriate circumstances so that the embodiments can be implemented in an order other than that shown or described in this application. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or modules does not necessarily limit to those steps or modules clearly listed, but may include other steps or modules not clearly listed or inherent to these processes, methods, products, or devices. The naming or numbering of steps in this application does not mean that the steps in the method flow must be executed in the time / logical sequence indicated by the naming or numbering. The named or numbered process steps can change the execution order according to the technical purpose to be achieved, as long as the same or similar technical effects can be achieved. The division of modules in this application is a logical division, and there may be other division methods in actual implementation. For example, multiple modules can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed coupling, direct coupling, or communication connection between each other can be through some interfaces. The indirect coupling or communication connection between units can be in an electrical or other similar form, which is not limited in this application. And the modules or sub-modules described as separate components may or may not be physically separated, may or may not be physical units, or may be distributed to multiple circuit units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this application.

[0044] The present invention provides a registration method for remote sensing images, which is applied to a remote sensing image registration server to solve the problem in the prior art that the number of matching points obtained in the feature point matching process of images with unclear features is small, or even there are no matching points, and at the same time improve the registration accuracy of remote sensing images with unclear features.

[0045] The following combines Figures 1 - 5 to describe the specific content of the present invention.

[0046] Figure 1 is one of the flow schematic diagrams of a registration method for remote sensing images provided by the present invention, including steps S101-S105.

[0047] Step S101, perform downsampling processing on the image with unclear features and the reference image respectively to obtain a downsampled remote sensing image and a downsampled reference image.

[0048] The featureless image in the embodiments of the present invention refers to an area in a remote sensing image where features of ground objects such as deserts and farmlands occupy a relatively large proportion. "Relatively large proportion" means that in the remote sensing image, there is at least more than half of the area that is a featureless area of ground objects. In this embodiment, no specific limitation is placed on the specific proportion of the featureless area of ground objects in the selected featureless image. It can be understood that other types of remote sensing images can also be set as featureless images for subsequent processing according to the actual needs of users.

[0049] Specifically, since in subsequent steps, there may be a situation where a target sampling rate is selected from multiple reference image sampling rates multiple times and the downsampled reference image is resampled. Therefore, repeatedly selecting the sampling rate and resampling the downsampled reference image will increase the time for feature point matching. To improve the feature point matching speed, it is necessary to downsample the featureless image and the reference image according to the sampling rate of the downsampling process based on the size of the featureless image with less obvious feature information to obtain a downsampled remote sensing image and a downsampled reference image. So that the subsequent feature point matching process for the featureless image is carried out on the downsampled remote sensing image.

[0050] In a possible implementation manner, in step S101, the featureless image and the reference image are respectively downsampled to obtain a downsampled remote sensing image and a downsampled reference image, which specifically includes steps S1011 - S1013.

[0051] Step S1011, obtain the image width and image height of the featureless image.

[0052] Step S1012, according to the image width and image height, as well as the preset downsampled image width and preset downsampled image height, obtain the sampling rate of the downsampling process.

[0053] Step S1013, according to the sampling rate of the downsampling process, respectively downsample the featureless image and the reference image to obtain a downsampled remote sensing image and a downsampled reference image.

[0054] Exemplarily, first set the preset downsampled image width or preset downsampled image height of the downsampled remote sensing image to , and the sampling rate during the downsampling process is calculated as follows:

[0055] ;

[0056] Wherein, max represents taking the maximum value, width and height are respectively the image width and image height of the featureless image.

[0057] When takes the value of , the preset downsampled image width is configured as ; when takes the value of , the preset downsampled image height is configured as . The value of S can be set by the user according to the actual application process, and the embodiments of the present invention do not make specific limitations. After the sampling rate S is calculated, the sampling rate S is used to perform downsampling processing on the featureless image and the reference image respectively, generating a downsampled remote sensing image and a downsampled reference image.

[0058] By adopting the above method, during the process of separately performing downsampling processing on the featureless image and the reference image, the reliability of the downsampling processing can be improved. By the downsampling processing, the influence of noise and regions with unclear details in the featureless image is reduced, and then by the resampling processing, the resolution of the reference image is adjusted, making the feature point matching more accurate, thereby increasing the number and quality of the matching points.

[0059] Step S102: Select a target sampling rate from multiple reference image sampling rates, perform resampling processing on the downsampled reference image to obtain a resampled reference image; and perform feature point matching between the resampled reference image and the downsampled remote sensing image to obtain multiple pairs of matching points.

[0060] Specifically, multiple sampling rates are included in the multiple reference image sampling rates. Selecting a target sampling rate from the multiple reference image sampling rates can start from the minimum sampling rate in the multiple reference image sampling rates, select the sampling rate in the set to obtain the resampled reference image; and perform feature point matching between the resampled reference image and the downsampled remote sensing image to obtain multiple pairs of matching points until the feature point matching is successful.

[0061] In a possible implementation manner, as Figure 2 shown, in step S102, performing feature point matching between the resampled reference image and the downsampled remote sensing image to obtain multiple pairs of matching points includes: steps S201 - S203.

[0062] Step S201: Obtain multiple feature points of the downsampled remote sensing image.

[0063] Specifically, feature points can be extracted on the downsampled remote sensing image through fast feature point extraction algorithms such as Harris and FAST. The present invention does not make specific limitations on the manner adopted in the process of extracting feature points of the downsampled remote sensing image, and only gives an example.

[0064] Step S202: Obtain a preset range area centered on each feature point to get a template window corresponding to each feature point; and set the resampled reference image as the search window.

[0065] Step S203: Based on the template window and the search window, perform gray-scale template matching between the resampled reference image and the downsampled remote sensing image to obtain multiple pairs of matching points.

[0066] Specifically, centered on each feature point, obtain the surrounding M M preset range areas as the template window, and set the resampled reference image as the search window. During the sliding process of the template window in the search window, calculate the similarity between the template window and the sub-image below the search window, so as to find the matching position. Repeat the above operations for each feature point, and then obtain multiple pairs of matching points. The similarity between the template window and the sub-image below the search window is calculated through the image gray-scale values.

[0067] By using the method of gray-scale template matching to obtain multiple pairs of matching points between the resampled reference image and the downsampled remote sensing image, the stability and matching success rate of the feature point matching process are improved.

[0068] In a possible implementation manner, in step S102, it further includes: if the feature point matching fails, then reselect a target sampling rate from multiple reference image sampling rates, and perform resampling processing on the downsampled reference image.

[0069] The following gives a specific example of the situation where the feature point matching fails.

[0070] In Figure 3 the shown embodiment, it includes: steps S301 - S302.

[0071] Step S301: Obtain the normalized correlation coefficient between the resampled reference image and the downsampled remote sensing image.

[0072] Step S302: If the normalized correlation coefficient is less than or equal to the normalized correlation coefficient threshold, then reselect a target sampling rate from multiple reference image sampling rates, and perform resampling processing on the downsampled reference image.

[0073] Gray-scale template matching mainly depends on the similarity of gray-scale values. Therefore, it is relatively sensitive to image rotation, scaling, illumination changes, etc. When there are large rotation or scaling differences between the resampled reference image and the downsampled remote sensing image, gray-scale template matching may not be able to accurately find the matching position, resulting in the failure of feature point matching.

[0074] Specifically, in the process of matching feature points between the resampled reference image and the downsampled remote sensing image to obtain multiple pairs of matching points, if the normalized correlation coefficient between the resampled reference image and the downsampled remote sensing image is greater than the normalized correlation coefficient threshold, it is considered that the feature point matching is successful. If the normalized correlation coefficient is less than or equal to the normalized correlation coefficient threshold, it is considered that the feature point matching fails, and a target sampling rate needs to be reselected from multiple reference image sampling rates, and the downsampled reference image is resampled to obtain a new resampled reference image. And continue to match the feature points between the new resampled reference image and the downsampled remote sensing image.

[0075] In a possible implementation manner, the following steps are further included.

[0076] According to the preset sampling rate interval, multiple sampling rates are generated between the maximum sampling rate of the reference image and the minimum sampling rate of the reference image to obtain multiple reference image sampling rates.

[0077] Specifically, the minimum sampling rate and the maximum sampling rate are specifically determined according to the size of the reference image, and it is necessary to ensure that the size of the downsampled reference image is greater than a preset number of pixels, so as to improve the success rate of feature point matching. The maximum sampling rate is determined according to the scale difference between the image with unclear feature information and the reference image.

[0078] By adopting the above method, a reference image sampling rate set is constructed, and multiple sampling rates are generated according to the preset sampling rate interval, so that the geometric correction requirements of images with different resolutions can be adapted, and the adaptability and flexibility of the image geometric correction process are enhanced.

[0079] In a possible implementation manner, in step S102, after obtaining multiple pairs of matching points, the following steps are further included:

[0080] Obtain the resolution difference value between the resampled reference image and the downsampled remote sensing image; if the resolution difference value is greater than or equal to the resolution difference threshold, reselect a target sampling rate from multiple reference image sampling rates, and resample the downsampled reference image.

[0081] Specifically, the inspection step of the feature point matching process is used to detect the feature point matching process of the gray template matching. Subsequently, after the feature point matching is completed, due to the unclear image feature information of the downsampled remote sensing image, it may cause wrong matching in the gray template matching, and it is necessary to further determine whether the feature point matching is successful.

[0082] By adopting the above method, obtaining the resolution difference value between the resampled reference image and the downsampled remote sensing image, and resampling according to the value by reselecting the sampling rate, can improve the registration accuracy of the remote sensing image with unclear features.

[0083] In a possible implementation manner, obtaining a resolution difference value between a resampled reference image and a downsampled remote sensing image includes:

[0084] Obtaining the resolution difference value according to the average value of the distance ratios between multiple pairs of matching points.

[0085] Specifically, since the gray template matching itself does not have scale invariance, only when the resolutions of the two images are relatively consistent can the correct matching be successful. Therefore, the average value of the distance ratios between multiple pairs of matching points can be used to determine whether the feature point matching is successful. The distance ratio of any two pairs of matching points The calculation formula is as follows:

[0086] ;

[0087] Wherein, and are the th pair of matching points and the th pair of matching points, and , represents the distance between the control points and on the downsampled remote sensing image, represents the distance between the control points and on the resampled reference image. Among them, the control point on the downsampled remote sensing image and the control point on the resampled reference image form the th pair of matching points; the control point on the downsampled remote sensing image and the control point on the resampled reference image form the th pair of matching points.

[0088] After all the distance ratios are calculated, calculate the average value of the distance ratios. If the average value is within the preset average value range, it is considered that the feature point matching is successful, and the iterative matching is stopped. Otherwise, it is considered that the feature point matching fails, return to step S102, reselect a target sampling rate from multiple reference image sampling rates, and continue the subsequent steps.

[0089] Exemplarily, the preset average value range can be configured as 0.8 - 1.2.

[0090] Step S103, according to the sampling rate of the downsampling process and the sampling rate of the resampling process, adjust the coordinates of multiple pairs of matching points to obtain multiple pairs of adjusted matching points.

[0091] Obtain the coordinates of the control points on the input downsampled remote sensing image and the resampled reference image. During the feature point matching process, the sampling rate SSampling was performed. The reference image was first sampled using the sampling rate S and then sampled using the sampling rate among multiple reference image sampling rates. If the coordinates of multiple pairs of matching points of the downsampled remote sensing image and the resampled reference image obtained are and respectively, then the coordinates of multiple pairs of adjusted matching points of the original input featureless image and the reference image and are calculated as follows respectively:

[0092] ;

[0093] ;

[0094] where is the set of reference image sampling rates SR ={ SR 1, SR 2, …, SR n}, and it is the p th reference image sampling rate in it.

[0095] Step S104: Adjust the resolution of the resampled reference image to be the same as that of the featureless image according to the mean value of the distance ratios between multiple pairs of adjusted matching points, and obtain a standard reference image.

[0096] Specifically, after the above feature point matching is completed on the downsampled remote sensing image, the obtained matching points are relatively few and the accuracy is relatively low. Therefore, image matching at the same scale is required. First, use multiple pairs of adjusted matching points, and calculate the mean value of the distance ratios according to the distance ratio calculation method in the above steps. This mean value is the resolution difference between the featureless image and the reference image. Then, use this mean value of the distance ratio as the sampling rate to sample the reference image to obtain a standard reference image, so that it has the same resolution as the featureless image.

[0097] Step S105: Perform feature point matching on the featureless image and the standard reference image; and perform image geometric correction on the featureless image according to the feature point matching result.

[0098] Specifically, using the featureless image and the standard reference image for feature point matching can obtain more pairs of matching points with higher accuracy. And use these matching points to construct a correction model, so as to complete the geometric correction of the featureless image.

[0099] In the embodiments of the present invention, the specific manner of performing image geometric correction on the featureless image according to the feature point matching result is not specifically limited.

[0100] In a possible implementation, after obtaining multiple pairs of matching points in the feature point matching performed in any step (including step S102 and step S105), it further includes:

[0101] If the number of matching points obtained by feature point matching is less than the first preset number of matching points, it is determined that the current feature point matching fails. If the number of control points is greater than the first preset number of matching points, methods such as RANSAC are used to eliminate the incorrect matching points among the multiple pairs of matching points. After the elimination is completed, if the number of matching points is less than the second preset number of matching points, it is determined that the current feature point matching fails. After the feature point matching fails, it is necessary to reselect a target sampling rate from multiple reference image sampling rates and perform resampling processing on the downsampled reference image to obtain a new resampled reference image.

[0102] By adopting the above method, the present invention can achieve the following beneficial effects: By performing downsampling processing on the featureless image and the reference image, the data volume of feature point matching is reduced in the process of subsequently selecting a target sampling rate from multiple reference image sampling rates to obtain a resampled reference image. Based on the sampling rates in the downsampling and resampling processes, multiple pairs of adjusted matching points are obtained to adjust the resolution of the reference image, and a standard reference image with the same resolution as the featureless image is obtained. This enables obtaining a standard reference image that can achieve better feature point matching results for the featureless image when the resolution of the featureless image is uncertain. The number of matching points obtained for the featureless image in the feature point matching process is increased, and the registration accuracy of the featureless remote sensing image is improved.

[0103] Figure 4 The structural schematic diagram of a registration device for remote sensing images provided by the present invention includes the following parts:

[0104] A downsampling processing module 410, configured to perform downsampling processing on the featureless image and the reference image respectively to obtain a downsampled remote sensing image and a downsampled reference image;

[0105] A first matching module 420, configured to select a target sampling rate from multiple reference image sampling rates, perform resampling processing on the downsampled reference image to obtain a resampled reference image; and perform feature point matching between the resampled reference image and the downsampled remote sensing image to obtain multiple pairs of matching points.

[0106] A coordinate adjustment module 430, configured to adjust the coordinates of multiple pairs of matching points according to the sampling rate of the downsampling processing and the sampling rate of the resampling processing to obtain multiple pairs of adjusted matching points.

[0107] An image adjustment module 440, configured to adjust the resolution of the resampled reference image to be the same as that of the featureless image according to the average value of the distance ratios between multiple pairs of adjusted matching points to obtain a standard reference image.

[0108] The second matching module 450 performs feature point matching between the images with unclear features and the standard reference images; and performs image geometric correction on the images with unclear features according to the feature point matching results.

[0109] In a possible implementation manner, the device further includes a matching result detection module, configured to obtain a resolution difference value between the resampled reference image and the downsampled remote sensing image; if the resolution difference value is greater than or equal to a resolution difference threshold, reselect a target sampling rate from multiple reference image sampling rates, and perform resampling processing on the downsampled reference image.

[0110] In a possible implementation manner, the matching result detection module is specifically configured to:

[0111] Obtain the resolution difference value according to the mean value of the distance ratios between multiple pairs of matching points.

[0112] In a possible implementation manner, the device further includes a set construction module, configured to construct multiple reference image sampling rates of the reference image, specifically:

[0113] Configure a maximum sampling rate and a minimum sampling rate for the reference image;

[0114] Generate multiple sampling rates between the maximum sampling rate and the minimum sampling rate according to a preset sampling rate interval to obtain multiple reference image sampling rates.

[0115] In a possible implementation manner, the downsampling processing module 410 is specifically configured to:

[0116] Obtain the image width and image height of the image with unclear features,

[0117] Obtain the sampling rate of the downsampling processing according to the image width and image height, and the preset downsampled image width and preset downsampled image height;

[0118] Perform downsampling processing on the image with unclear features and the reference image respectively according to the sampling rate of the downsampling processing to obtain a downsampled remote sensing image and a downsampled reference image.

[0119] In a possible implementation manner, the first matching module 420 is specifically configured to:

[0120] Obtain multiple feature points of the downsampled remote sensing image;

[0121] Taking each feature point as the center, obtain a region within a preset range to obtain a template window corresponding to each feature point; and set the resampled reference image as the search window;

[0122] Perform gray - scale template matching on the resampled reference image and the down - sampled remote sensing image according to the template window and the search window to obtain multiple pairs of matching points.

[0123] In a possible implementation, the first matching module 420 is further configured to:

[0124] Obtain the normalized correlation coefficient between the resampled reference image and the down - sampled remote sensing image;

[0125] If the normalized correlation coefficient is less than or equal to the normalized correlation coefficient threshold, re - select a target sampling rate from multiple reference - image sampling rates and perform resampling processing on the down - sampled reference image.

[0126] Figure 5 Illustrates a schematic physical structure diagram of an electronic device, as Figure 5 shown. The electronic device may include: a processor 510, a communication interface 520, a memory 530, and a communication bus 540. Among them, the processor 510, the communication interface 520, and the memory 530 complete mutual communication through the communication bus 540. The processor 510 can call the logical instructions in the memory 530 to execute a registration method for remote sensing images, and the method includes:

[0127] Perform down - sampling processing on the feature - less obvious image and the reference image respectively to obtain the down - sampled remote sensing image and the down - sampled reference image;

[0128] Select a target sampling rate from multiple reference - image sampling rates, perform resampling processing on the down - sampled reference image to obtain a resampled reference image; and perform feature - point matching between the resampled reference image and the down - sampled remote sensing image to obtain multiple pairs of matching points;

[0129] Adjust the coordinates of multiple pairs of matching points according to the sampling rate of the down - sampling processing and the sampling rate of the resampling processing to obtain multiple pairs of adjusted matching points;

[0130] Adjust the resolution of the resampled reference image to be consistent with that of the feature - less obvious image according to the mean of the distance ratios between multiple pairs of adjusted matching points to obtain a standard reference image;

[0131] Perform feature - point matching between the feature - less obvious image and the standard reference image; and perform image geometric correction on the feature - less obvious image according to the feature - point matching result.

[0132] In addition, when the logical instructions in the above-mentioned memory 530 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

[0133] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute a remote sensing image registration method provided by the above-mentioned various methods. The method includes:

[0134] Performing downsampling processing on the image with unclear features and the reference image respectively to obtain a downsampled remote sensing image and a downsampled reference image;

[0135] Selecting a target sampling rate from multiple reference image sampling rates, performing resampling processing on the downsampled reference image to obtain a resampled reference image; and performing feature point matching between the resampled reference image and the downsampled remote sensing image to obtain multiple pairs of matching points;

[0136] Adjusting the coordinates of multiple pairs of matching points according to the sampling rate of the downsampling processing and the sampling rate of the resampling processing to obtain multiple pairs of adjusted matching points;

[0137] Adjusting the resolution of the resampled reference image to be consistent with the image with unclear features according to the average value of the distance ratios between multiple pairs of adjusted matching points to obtain a standard reference image;

[0138] Performing feature point matching between the image with unclear features and the standard reference image; and performing image geometric correction on the image with unclear features according to the feature point matching result.

[0139] On yet another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements a remote sensing image registration method provided by the above-mentioned various methods. The method includes:

[0140] Downsample the feature-unclear image and the reference image respectively to obtain a downsampled remote sensing image and a downsampled reference image;

[0141] Select a target sampling rate from multiple reference image sampling rates, resample the downsampled reference image to obtain a resampled reference image; and perform feature point matching between the resampled reference image and the downsampled remote sensing image to obtain multiple pairs of matching points;

[0142] Adjust the coordinates of multiple pairs of matching points according to the sampling rate of the downsampling process and the sampling rate of the resampling process to obtain multiple pairs of adjusted matching points;

[0143] Adjust the resolution of the resampled reference image to be the same as that of the feature-unclear image according to the average value of the distance ratios between multiple pairs of adjusted matching points to obtain a standard reference image;

[0144] Perform feature point matching between the feature-unclear image and the standard reference image; and perform image geometric correction on the feature-unclear image according to the feature point matching result.

[0145] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0146] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course also by hardware. Based on such an understanding, the above technical solution, in essence, or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disc, etc., including several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods of each embodiment or some parts of the embodiments.

[0147] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of each embodiment of the present invention.

Claims

1. A registration method for remote sensing images, characterized in that, Including: Downsample the feature-unclear image and the reference image respectively to obtain a downsampled remote sensing image and a downsampled reference image; Select a target sampling rate from multiple reference image sampling rates, resample the downsampled reference image to obtain a resampled reference image; and perform feature point matching between the resampled reference image and the downsampled remote sensing image to obtain multiple pairs of matching points; Adjust the coordinates of multiple pairs of matching points according to the sampling rate of the downsampling process and the sampling rate of the resampling process to obtain multiple pairs of adjusted matching points; Adjust the resolution of the resampled reference image to be consistent with that of the feature-unclear image according to the average value of the distance ratios between multiple pairs of adjusted matching points to obtain a standard reference image; Perform feature point matching between the feature-unclear image and the standard reference image; And perform image geometric correction on the feature-unclear image according to the feature point matching result.

2. The registration method of remote sensing images according to claim 1, characterized in that, Also including: Obtain the resolution difference value between the resampled reference image and the downsampled remote sensing image; If the resolution difference value is greater than or equal to the resolution difference threshold, reselect a target sampling rate from multiple reference image sampling rates and resample the downsampled reference image.

3. The registration method of the remote sensing image according to claim 2, wherein The obtaining the resolution difference value between the resampled reference image and the downsampled remote sensing image includes: obtaining the resolution difference value according to the average value of the distance ratios between multiple pairs of matching points.

4. The registration method of the remote sensing image according to claim 1, characterized in that, Also including: Generate multiple sampling rates between the maximum sampling rate and the minimum sampling rate of the reference image according to a preset sampling rate interval to obtain multiple reference image sampling rates.

5. The registration method of the remote sensing image according to claim 1, characterized in that, The downsampling the feature-unclear image and the reference image respectively to obtain a downsampled remote sensing image and a downsampled reference image specifically includes: Obtain the image width and image height of the feature-unclear image; Obtain the sampling rate of the downsampling process according to the image width, the image height, a preset downsampled image width, and a preset downsampled image height; Downsample the feature-unclear image and the reference image respectively according to the sampling rate of the downsampling process to obtain a downsampled remote sensing image and a downsampled reference image.

6. The registration method of the remote sensing image according to claim 1, wherein The performing feature point matching between the resampled reference image and the downsampled remote sensing image to obtain multiple pairs of matching points includes: Obtain multiple feature points of the downsampled remote sensing image; Obtain a preset range area centered on each feature point to obtain a template window corresponding to each feature point; and set the resampled reference image as the search window; Perform gray-scale template matching between the resampled reference image and the downsampled remote sensing image according to the template window and the search window to obtain multiple pairs of matching points.

7. The registration method of the remote sensing image according to claim 6, characterized in that, Also including: If the feature point matching fails, reselect a target sampling rate from multiple reference image sampling rates and resample the downsampled reference image.

8. A registration device for remote sensing images, characterized in that Including: A downsampling processing module for downsampling the feature-unclear image and the reference image respectively to obtain a downsampled remote sensing image and a downsampled reference image; A first matching module, configured to select a target sampling rate from multiple reference image sampling rates, perform resampling processing on the downsampled reference image to obtain a resampled reference image; and perform feature point matching between the resampled reference image and the downsampled remote sensing image to obtain multiple pairs of matching points; A coordinate adjustment module, configured to adjust the coordinates of multiple pairs of matching points according to the sampling rate of the downsampling process and the sampling rate of the resampling process to obtain multiple pairs of adjusted matching points; An image adjustment module, configured to adjust the resolution of the resampled reference image to be consistent with that of the featureless image according to the average value of the distance ratios between multiple pairs of adjusted matching points to obtain a standard reference image; A second matching module, configured to perform feature point matching between the featureless image and the standard reference image; and perform image geometric correction on the featureless image according to the feature point matching result.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, the registration method of the remote sensing image according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the registration method of the remote sensing image according to any one of claims 1 to 7 is implemented.

Images