Image matching method and system

Abstract

The invention provides an image matching method and system, and the method comprises the steps: obtaining a to-be-matched optical image and a to-be-matched SAR image, respectively extracting the edge information of linear ground objects in the optical image and the SAR image, carrying out the matching of the edge information corresponding to the optical image and the edge information corresponding to the SAR image, obtaining a first matching result, carrying out the matching of the optical image and the SAR image, and obtaining a second matching result; and obtaining a second matching result, and determining an image matching result between the optical image and the SAR image according to the first matching result and the second matching result. Through matching from two dimensions, the data volume is greatly increased, so that the richness and comprehensiveness of information matching can be improved. And therefore, the image matching comprehensiveness, reliability and accuracy can be improved.

Description

Technical Field

[0001] This specification relates to the field of image processing technologies, and in particular, to an image matching method and system. Background Art

[0002] Image matching, which can also be referred to as image registration, refers to the process of aligning two or more images corresponding to different or the same sensors, times, or field of view angles taken of the same scene. That is, image matching is the process of finding the pixel mapping relationship between multiple images of the same scene taken at different times, perspectives, and acquisition devices.

[0003] The matching process between SAR images and optical images is one of the most fundamental yet quite challenging operations in remote sensing. Due to the presence of radiometric and geometric differences, SAR images and optical images of the same area differ in spatial resolution, spatial alignment, satellite type, and time dimension. In addition, due to the imaging principle of the SAR sensor, the electromagnetic waves emitted will interfere with the echoes after being reflected by rough surfaces, forming a very strong granular noise on the SAR image, which is called speckle noise. This noise will greatly interfere with the recognition and extraction of the edge details of the SAR image, making it unable to meet the requirements of the matching algorithm, and has become a hot topic in the research on the matching of optical images and SAR images.

[0004] It should be noted that the content of the above related technologies is only the information known to the inventor personally, and does not represent that the above information has entered the public domain before the filing date of this specification, nor does it represent that it can become the prior art of this specification. Summary of the Invention

[0005] This specification provides an image matching method and system to avoid at least one of the above technical problems.

[0006] In a first aspect, this specification provides an image matching method, including:

[0007] Obtain an optical image and an SAR image to be matched;

[0008] Extract the edge information of linear features in the optical image and the SAR image respectively, and match the edge information corresponding to the optical image and the SAR image respectively to obtain a first matching result;

[0009] Match the optical image and the SAR image to obtain a second matching result; and

[0010] Determine the image matching result between the optical image and the SAR image according to the first matching result and the second matching result.

[0011] Second aspect, this specification provides an image matching system, including:

[0012] At least one storage medium storing at least one instruction set for performing image matching;

[0013] At least one processor communicatively connected to the at least one storage medium, wherein when the at least one processor runs, it reads the at least one instruction set and executes the method described in the first aspect according to the instructions of the at least one instruction set.

[0014] Third aspect, this specification provides a computer-readable non-transitory storage medium, wherein the computer-readable non-transitory storage medium stores at least one instruction set, and the at least one instruction set is executed by at least one processor to implement the method described in the first aspect.

[0015] As can be seen from the above technical solutions, the image matching method and system provided in this specification match the optical image and SAR image to be matched from the linear feature dimension and the image itself dimension, greatly increasing the amount of data, thereby improving the richness and comprehensiveness of information matching. Furthermore, the comprehensiveness, reliability, and accuracy of image matching can be improved.

[0016] Other functions of the image matching method and system provided in this specification will be partially listed in the following description. The creative aspects of the image matching method and system provided in this specification can be fully explained by practicing or using the methods, devices, and combinations described in the following detailed examples. Description of the Drawings

[0017] To more clearly illustrate the technical solutions in the embodiments of this specification, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of this specification. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 Schematic diagram of the application scenario of the image matching method provided for the embodiments of this specification;

[0019] Figure 2 Schematic diagram of the structure of the image matching system provided for the embodiments of this specification;

[0020] Figure 3 Schematic diagram of the flow of the image matching method provided for an embodiment of this specification;

[0021] Figure 4 Schematic diagram of the principle of the image matching method provided for an embodiment of this specification;

[0022] Figure 5 Schematic flowchart of the image matching method provided by another embodiment of this specification;

[0023] Figure 6 Schematic diagram of the principle of the image matching method provided by another embodiment of this specification. Detailed implementation manners

[0024] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all the implementation manners consistent with this specification. On the contrary, they are merely examples of devices and methods consistent with some aspects of this specification as detailed in the appended claims.

[0025] It should be understood that the terms "include" and "have" and any variations thereof in the embodiments of this specification are intended to cover but not exclusively include. For example, a product or device including a series of components does not necessarily have to be limited to those components clearly listed, but may include other components not clearly listed or inherent to these products or devices.

[0026] The term "and / or" in the embodiments of this specification describes the association relationship of associated objects and indicates that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after.

[0027] The term "plurality" in the embodiments of this specification means two or more, and other quantifiers are similar thereto.

[0028] The terms "first", "second", "third", etc. in this specification are used to distinguish similar or homogeneous objects or entities, and do not necessarily mean to limit a specific order or sequence, unless otherwise indicated. It should be understood that such terms can be interchanged under appropriate circumstances, for example, they can be implemented in an order other than those given in the illustrations or descriptions of the embodiments of this specification.

[0029] The term "unit / module" used in this specification refers to any known or later developed hardware, software, firmware, artificial intelligence, fuzzy logic, or a combination of hardware or / and software code that can perform functions related to the element.

[0030] To avoid at least one of the technical problems mentioned in the above background art, this specification presents a technical concept achieved through creative efforts: performing image matching from two dimensions, namely the image dimension and the linear feature dimension. For example, matching the edge information of the corresponding linear features in the optical image and the SAR image respectively to obtain the matching result from the linear feature dimension. Matching the optical image and the SAR image to obtain the matching result from the image itself dimension. Finally, combining the matching results from the linear feature dimension and the image itself dimension to jointly determine the final matching result between the optical image and the SAR image.

[0031] Among them, linear features can be understood as various natural and artificial structures existing in the actual geographical environment. Exemplarily, linear features can include rivers, ditches, roads, buildings, etc.

[0032] The technical solution provided in this specification is implemented based on the above technical concept. Based on the description of the above technical concept, in the technical solution provided in this specification, by matching the optical image and the SAR image to be matched from the linear feature dimension and the image itself dimension, the amount of data is greatly increased, thereby improving the richness and comprehensiveness of information matching. Furthermore, the comprehensiveness, reliability, and accuracy of image matching can be improved.

[0033] For the convenience of readers to understand this specification, the application scenarios of this specification are introduced below.

[0034] The technical solution provided in this specification is applicable to scenarios where optical images and SAR images need to be matched. For example, the technical solution provided in this specification can be applied to: Geographic Information System (GIS) and map update, natural disaster detection and assessment, agricultural detection and crop management, urban planning and infrastructure detection, target recognition, and so on.

[0035] Taking the scenario of the above-mentioned Geographic Information System and map update as an example:

[0036] In the Geographic Information System, it is necessary to update map data regularly to reflect surface changes (such as urban expansion, road construction, etc.). Optical images are usually used to provide high-resolution details of ground features, while SAR images can supplement information that cannot be obtained by optical images under cloud cover or rainy weather.

[0037] Correspondingly, in the map update task of a certain area, the optical image clearly shows the urban building layout and road network during the day, but subsequent updates are difficult due to continuous rainy weather. At this time, data of the same area is obtained through the SAR image, and the image matching method provided in this specification is used to match the optical image and the SAR image, so as to align and fuse the two to generate a complete updated map of surface information. So that the matched multi-modal image can provide more comprehensive surface information to support efficient map update and land use analysis.

[0038] Taking the above-mentioned scenario of agricultural detection and crop management as an example:

[0039] Agricultural detection requires long-term tracking of crop growth, but weather conditions (such as clouds, rainfall) may interfere with the acquisition of optical images. The SAR image can obtain data under any weather conditions, but its analysis ability is limited. By using the image matching method provided in this specification to match the optical image and the SAR image, the advantages of both can be combined to perform accurate crop classification and growth status evaluation.

[0040] For example, in the crop detection of a certain farmland, the optical image clearly shows the color and texture characteristics of different crops, while the SAR image provides the height and structure information of the crops. By using the image matching method provided in this specification to match the optical image and the SAR image, different types of crops (such as wheat, corn, rice) can be distinguished and their growth status can be evaluated. So that the matched image can improve the accuracy of crop classification and support precision agriculture management and yield prediction.

[0041] For the description of the application of this specification to other application scenarios, reference can be made to the above examples, and they will not be listed one by one here.

[0042] Figure 1 FIG. is a schematic diagram of the application scenario of the image matching method according to the embodiment of this specification. Among them, the image matching method of this specification can be applied to, for example Figure 1 Scene 100 shown in FIG. As Figure 1 shown, scene 100 may include target user 101, client 102, server 103, and network 104.

[0043] The target user 101 can be a user who triggers the matching of the optical image to be matched and the SAR image. For example, the target user 101 can perform a target operation on the client 102 to trigger the matching of the optical image and the SAR image. In some embodiments, the target user 101 can upload the optical image and the SAR image on the client 102 to trigger the matching of the two. In some embodiments, the target user 101 can input indication information on the client 102 to indicate the optical image and the SAR image to be matched. The above indication information can be the identifiers of the optical image and the SAR image, etc.

[0044] The client 102 can be an electronic device that provides an interaction function for the target user 101. For example, the client 102 can provide an interaction interface for the target user 101, and the target user 101 can perform interaction operations on the interaction page. In some embodiments, in response to detecting an operation triggered by the target user 101 to match the optical image and the SAR image, the client 102 executes the image matching method described in this specification. At this time, the client 102 can store data or instructions for executing the image matching method described in this specification, and can execute or be used to execute the data or instructions. In some embodiments, the client 102 can include a hardware device with data information processing functions and the necessary programs for driving the hardware device to work, so as to execute the image matching method described in this specification.

[0045] In some embodiments, the client 102 can include a mobile device, a tablet computer, a laptop computer, a built-in device of a motor vehicle, or the like, or any combination thereof. In some embodiments, the mobile device can include a smart home device, a smart mobile device, a virtual reality device, an augmented reality device, or the like, or any combination thereof. In some embodiments, the smart home device can include a smart TV, a desktop computer, etc., or any combination. In some embodiments, the smart mobile device can include a smart phone, a personal digital assistant, a gaming device, a navigation device, etc., or any combination thereof. In some embodiments, the built-in device in a motor vehicle can include an in-vehicle computer, an in-vehicle TV, etc. In some embodiments, the client 102 can include a collection device for collecting optical images and SAR images.

[0046] In some embodiments, the client 102 may be installed with one or more applications (APPs). The APPs can provide the target user 101 with the ability to interact with the outside world through the network 104 and an interface. The APPs include, but are not limited to: web browser APPs, search APPs, chat APPs, shopping APPs, video APPs, financial management APPs, instant messaging tools, email clients, social platform software, and so on. In some embodiments, a target APP may be installed on the client 102. The target APP can collect optical images and SAR images for the client 102.

[0047] As Figure 1 shown, the client 102 may be communicatively connected to the server 103. Among them, the server 103 may be communicatively connected to one client 102 or multiple clients 102. In some embodiments, the client 102 may interact with the server 103 through the network 104 to receive or send messages, etc.

[0048] The server 103 may be a server that provides various services. For example, the server 103 may be a cloud server or a local server. The server 103 may be communicatively connected to one client 102 and receive the data sent by the client 102, or may be communicatively connected to multiple clients 102 and receive the data sent by each client 102 respectively.

[0049] In some embodiments, the image matching method described in this specification may be executed on the server 103. At this time, the server 103 may store the data or instructions for executing the image matching method described in this specification and may execute or be used to execute the data or instructions. The server 103 may include a hardware device with data information processing capabilities and the necessary programs for driving the hardware device to work.

[0050] The network 104 is a medium for providing a communication connection between the client 102 and the server 103. The network 104 can facilitate the exchange of information or data. As Figure 1 shown, the client 102 and the server 103 may be respectively connected to the network 104 and transmit information or data to each other through the network 104.

[0051] In some embodiments, network 104 can be any type of wired or wireless network, or a combination thereof. For example, network 104 can include a cable network, a wired network, an optical fiber network, a telecommunication network, an intranet, the Internet, a Local Area Network (LAN), a Wide Area Network (WAN), a Wireless Local Area Network (WLAN), a Metropolitan Area Network (MAN), a Public Switched Telephone Network (PSTN), a Bluetooth networkTM, a short-range wireless network (ZigBeeTM), a Near Field Communication (NFC) network, or a similar network.

[0052] In some embodiments, network 104 can include one or more network access points. For example, network 104 can include a wired or wireless network access point, such as a base station or an Internet exchange point, through which one or more components of client 102 and server 103 can be connected to network 104 to exchange data or information.

[0053] It is worth noting that Figure 1 the numbers of client 102, server 103, and network 104 in

[0054] are merely illustrative. According to the implementation requirements, there can be any number of client 102, server 103, and network 104. And the image matching method provided in this specification can be executed entirely on client 102, entirely on server 103, or partially on client 102 and partially on server 103. Figure 1 That is to say, Figure 1 and the above description for

[0055] Figure 2The hardware structure diagram of an image matching system 200 provided according to an embodiment of this specification is shown. The image matching system 200 can execute the image matching method described in this specification. The image matching method is introduced in other parts of this specification. When the image matching method is executed on the client 102, the image matching system 200 can be the client 102. When the image matching method is executed on the server 103, the image matching system 200 can be the server 103. When the image matching method is partially executed on the client 102 and partially executed on the server 103, the image matching system 200 can be a system including the client 102 and the server 103.

[0056] As Figure 2 shown, the image matching system 200 can include at least one storage medium 203 and at least one processor 202. In some embodiments, the image matching system 200 can further include a communication port 204 and an internal communication bus 201. The image matching system 200 can further include I / O components 205.

[0057] The internal communication bus 201 can connect different system components. For example, the internal communication bus 201 can connect the storage medium 203, the processor 202, the communication port 204, and the I / O components 205.

[0058] The I / O components 205 support input / output between the image matching system 200 and other components.

[0059] The communication port 204 is used for data communication between the image matching system 200 and the outside world. For example, the communication port 204 can be used for data communication between the image matching system 200 and the network 104. The communication port 204 can be a wired communication port or a wireless communication port.

[0060] The storage medium 203 can include a data storage device. The data storage device can be a non-temporary storage medium or a temporary storage medium. For example, the data storage device can include one or more of a magnetic disk 2031, a read-only storage medium (ROM) 2032, or a random access storage medium (RAM) 2033. The storage medium 203 further includes at least one instruction set stored in the data storage device. The instruction set includes computer program code, and the computer program code can include programs, routines, objects, components, data structures, processes, modules, etc. for executing the image matching method provided in this specification.

[0061] At least one processor 202 may be communicatively connected to at least one storage medium 203. The at least one processor 202 is configured to execute the at least one instruction set described above. When the image matching system 200 runs, the at least one processor 202 reads the at least one instruction set and, according to the instructions of the at least one instruction set, executes the image matching method provided in this specification. The processor 202 may execute all steps included in the image matching method. The processor 202 may be in the form of one or more processors. In some embodiments, the processor 202 may include one or more hardware processors, such as a microcontroller, a microprocessor, a reduced instruction set computer (RISC), an application specific integrated circuit (ASIC), an application specific instruction set processor (ASIP), a central processing unit (CPU), a graphics processing unit (GPU), a physics processing unit (PPU), a microcontroller unit, a digital signal processor (DSP), a field programmable gate array (FPGA), an advanced RISC machine (ARM), a programmable logic device (PLD), any circuit or processor capable of executing one or more functions, etc., or any combination thereof.

[0062] For illustrative purposes only, only one processor 202 is shown in the image matching system 200 in the drawings. However, it should be noted that the image matching system 200 in this specification may also include multiple processors. Therefore, the operations and / or method steps disclosed in this specification may be executed by one processor or jointly executed by multiple processors. For example, if it is described in this specification that the processor 202 of the image matching system 200 executes step A and step B, it should be understood that step A and step B may also be jointly or separately executed by two different processors 202 (for example, the first processor executes step A, the second processor executes step B, or the first and second processors jointly execute steps A and B).

[0063] Please refer to Figure 3 , Figure 3 which is a schematic flowchart of the image matching method provided in an embodiment of this specification. Among them, Figure 3 the execution subject of the shown image matching method may be the image matching system. For the description of the image matching system, reference may be made to the above examples, which will not be elaborated here.

[0064] As Figure 3 shown, the method includes the following S301 to S304:

[0065] S301: Obtain an optical image and a SAR image to be matched.

[0066] The optical image can be understood as a surface image obtained based on the reflection signal in the visible light or near-infrared band, usually providing high-resolution color and texture information.

[0067] The SAR image can be understood as a surface image obtained based on microwave radiation signals, which has the ability to image all-weather and all-time, but relatively lacks color information and mainly reflects the geometric structure and electromagnetic characteristics of ground objects.

[0068] In this embodiment, the manner in which the image matching system obtains the optical image and the SAR image is not limited. Taking the example of the image matching system obtaining the optical image, it can be implemented by the following examples:

[0069] In one example, the image matching system can be connected to an image acquisition device and receive the optical image collected and transmitted by the image acquisition device.

[0070] In another example, the image matching system can provide a tool for loading images, and the user can transmit the optical image to the image matching system through this tool for loading images.

[0071] Among them, the tool for loading images can be an interface for connecting to an external device, such as an interface for connecting to other storage devices, and the optical image transmitted by the external device is obtained through this interface; the tool for loading images can also be a display device. For example, the image matching system can output an interface for the function of loading images on the display device, and the user can import the optical image into the image matching system through this interface.

[0072] S302: Extract the edge information of linear features in the optical image and the SAR image respectively, and match the edge information corresponding to the optical image and the SAR image respectively to obtain a first matching result.

[0073] The edge information can be understood as the contour information of an object (such as a linear feature) in the image, which can be used to identify the object boundary.

[0074] Correspondingly, the edge information of the optical image can be understood as the contour information of linear features in the optical image for identifying the boundary of linear features, etc. The edge information of the SAR image can be understood as the contour information of linear features in the SAR image for identifying the boundary of linear features, etc.

[0075] Exemplarily, in combination with Figure 4 It can be known that this step can be understood as: the image matching system can extract the first edge information (such as contour information, etc.) of linear features in the optical image, and can also extract the second edge information of linear features in the SAR image. Then, the image matching system can match the first edge information and the second edge information to obtain a first matching result. That is, the first matching result is a matching result obtained by matching from the dimension of linear features based on the extracted edge information.

[0076] In some embodiments, the linear features include at least one of roads, rivers, ditches, and buildings.

[0077] Taking a linear feature including a road as an example, the image matching system can extract the first edge information corresponding to the road from an optical image, and can also extract the second edge information of the road from a SAR image, and match the first edge information and the second edge information to obtain a matching result (i.e., the first matching result) of two different modal images (i.e., the optical image and the SAR image) in terms of the road dimension.

[0078] Relatively speaking, by combining the first matching results of linear features such as roads and rivers, the image matching system can assist in determining the final matching result of the optical image and the SAR image, which can improve the accuracy and reliability of the matching. Especially in the case where the linear feature includes multiple types of objects, the diversity and comprehensiveness of the matching can be further improved.

[0079] S303: Match the optical image and the SAR image to obtain a second matching result.

[0080] Although there may be significant differences between the two image modalities (the optical image and the SAR image), in some cases, directly comparing pixel values or local region features may also reveal valuable matching relationships.

[0081] Therefore, continuing to refer to the above example and Figure 4 it can be known that in this step, the image matching system can also match the optical image and the SAR image from the dimension of the image itself. For example, the image matching system can use techniques such as a method based on gray values (such as normalized cross-correlation), a transformation model (such as affine transformation or homography transformation), etc. to match the optical image and the SAR image to obtain the corresponding second matching result.

[0082] S304: Determine the image matching result between the optical image and the SAR image according to the first matching result and the second matching result.

[0083] Continuing to combine the above example and Figure 4 , after obtaining the first matching result and the second matching result, the image matching system can combine the first matching result and the second matching result to jointly determine the final image matching result of the optical image and the SAR image.

[0084] Combined with the above analysis, it can be known that the first matching result is the matching result between the optical image and the SAR image determined from the dimension of linear features such as roads and rivers; the second matching result is the matching result between the optical image and the SAR image determined from the dimension of the image itself.

[0085] In this embodiment, the image matching system determines the final matching result (i.e., the image matching result) between the optical image and the SAR image by combining the matching results in two dimensions (including the first matching result and the second matching result), so as to enhance the robustness and accuracy of the matching by integrating information from different perspectives.

[0086] That is to say, the image matching system can overcome the limitations of a single matching strategy and improve the accuracy of matching by performing matching by combining edge information and original image information respectively. Thus, more accurate and reliable cross-modal image matching can be achieved to meet the requirements of various practical applications.

[0087] For the convenience of readers to further understand the technical principle of the image matching method provided in this specification, now in combination with Figure 5 and Figure 6 and taking linear features as roads as an example, the image matching method provided in this specification will be described in more detail.

[0088] Among them, Figure 5 is a schematic flowchart of the image matching method provided in another embodiment of this specification. As Figure 5 shown, this method includes the following S501 to S507:

[0089] S501: Obtain the optical image and the SAR image to be matched.

[0090] It can be understood that, in order to avoid cumbersome statements, regarding the technical features and principles that are the same or similar in this embodiment and the above examples, this embodiment will not be elaborated again. For example, regarding the implementation principle of S501, reference can be made to the description of S301 in the above example.

[0091] S502: Extract the first edge information of the road from the optical image, and extract the second edge information of the road from the SAR image.

[0092] Similarly, regarding the implementation principle of S502, reference can be made to the description of the technical principle of edge information extraction in S302 of the above example, which will not be elaborated here.

[0093] In some embodiments, taking the image matching system extracting the first edge information as an example:

[0094] The image matching system can first determine the binary image of the optical image, and then extract the edge information corresponding to the road in the optical image (i.e., the first edge information) from the binary image of the optical image.

[0095] Exemplarily, the image matching system can use the OSTU binary method and the dilation and erosion algorithm to smooth the road edges to obtain the first edge information.

[0096] For example, the image matching system can use a binarization method to segment the road from the optical image, such as using contrast normalization to enhance the edge information of the road. i It can be realized based on formula 1:

[0097]

[0098] Among them, M is the original gray level, m i is the number of pixels with gray level i.

[0099] If the threshold t divides the grayscale into two categories, the occurrence probabilities of the two categories are w0 and w1, which can be expressed by formula 2:

[0100]

[0101] Correspondingly, the average gray value of the occurrence probability w0 and w1 can be expressed by Formula 3:

[0102]

[0103] in,

[0104] In addition, the variance between the two categories can be expressed by Formula 4:

[0105] σ 2 =w0*(μ0-μ T ) 2 +w1*(μ0-μ T ) 2

[0106] Correspondingly, the gray level 1 to M is cyclically increased so that formula 4 reaches the maximum, which is the optimal segmentation threshold. However, considering that there may be more interference in the optical image, the edge information of the road in the optical image may be missing, such as discontinuity in some places. Therefore, the image matching system can use the expansion and corrosion algorithm to process the binary image after the OSTU binarization method to achieve the purpose of reducing interference and smoothing edges, thereby obtaining relatively more accurate and reliable first edge information.

[0107] Exemplarily, the image matching system may first perform an erosion operation on the binary image to remove small noise points or unnecessary details at the edge of the road, so that the road boundary is clearer but may become thinner.

[0108] Subsequently, the image matching system can perform a dilation operation on the eroded image obtained after the erosion operation to restore information such as the actual width of the road. Since noise has been removed previously, the current dilation operation will not introduce new noise. Instead, it helps to connect road segments that might have been disconnected due to noise, making the road edges appear more continuous and smooth, thereby obtaining the final first edge information, such as a road edge image including a clearly defined and continuous road edge.

[0109] Combining the above analysis, it can be seen that in this embodiment, the image matching system can improve the accuracy and reliability of the determined edge information by first determining the binary image and then extracting the corresponding edge information from the binary image.

[0110] In some embodiments, the above-mentioned "extracting the edge information corresponding to the road in the optical image from the binary image of the optical image (i.e., the first edge information)" may include the following steps 11 to 13:

[0111] Step 11: Extract the gray value corresponding to each pixel point from the binary image of the optical image.

[0112] Step 12: Determine the first edge point of the linear feature in the optical image according to the extracted gray values.

[0113] Exemplarily, combining the above example and Figure 6 it can be known that after processing the optical image based on the OSTU binarization method, a binary image can be obtained. The image matching system can extract edge points based on the binary image.

[0114] For example, the image matching system can use the multi-condition weighting method to detect edge points. For example, the image matching system can detect edge points based on Equation 5. Equation 5:

[0115] M k = λa k + ηb k

[0116] where M k is the comprehensive weight of pixel point X k and is used to determine whether this pixel is a road edge point. λ and η are preset weight coefficients. a k is the sum of the differences between the gray value of pixel point X k and the gray values of its 8 adjacent pixel points X ki and can be expressed as b k is the absolute difference between pixel point X k and the average road gray value R, and can be expressed as b k = abs(X k - R).

[0117] Correspondingly, if M k reaches the preset threshold P, it can be determined that the pixel point X k is a road edge point. For example, the image matching system can use the pixel point X k to determine the first edge point.

[0118] Step 13: Search in the binary image of the optical image based on the first edge point to obtain the edge information corresponding to the linear feature in the optical image.

[0119] Exemplarily, in combination with the above example and Figure 6 it can be known that after the edge point is extracted, the matching system can determine the edge information corresponding to the optical image (i.e., the first edge information) based on the edge point.

[0120] For example, after determining an edge point P0, the image matching system can use a five-neighbor edge detector to search for subsequent edge points. Finally, the image matching system can utilize the characteristics that the road has a constant width and the curvature of the road segment has a maximum value. Based on the extraction of one edge of the road, it can extract the other edge of the road in a complete manner using the bridge connection mode without being affected by internal obstacles of the road to obtain the final first edge information.

[0121] Combined with the above analysis of Steps 11 to 13, it can be known that in this embodiment, first, the image matching system can determine the first edge point based on the gray value in the binary image, which can make the determined first edge point have strong accuracy and reliability. Then, the image matching system can search for each edge point based on the determined first edge point, so as to obtain the edge information corresponding to the optical image. Compared with determining all edge points, the efficiency of obtaining edge information can be improved.

[0122] Similarly, regarding the implementation principle of "extracting the second edge information of the road from the SAR image", reference can be made to the description of "extracting the first edge information of the road from the optical image" in the above example, which will not be elaborated here. As Figure 6 shown, the image matching system can adopt the principle of extracting the first edge information to extract the corresponding edge information from the SAR image.

[0123] S503: Match the first road edge information and the second road edge information to obtain the first matching result.

[0124] In some embodiments, S503 may include the following Steps 21 to 23:

[0125] Step 21: Determine the gradient corresponding to each edge information, where the gradient is used to characterize the change rate of the image pixel value in space.

[0126] Exemplarily, the image matching system may determine a first gradient corresponding to the first road edge information, and may also determine a second gradient corresponding to the second road edge information.

[0127] In some embodiments, the image matching system may determine the first gradient and the second gradient in different ways. For example, Figure 6 as shown, the image matching system may determine the first gradient based on the Sobel operator and may determine the second gradient based on the ROEWA operator.

[0128] For example, for an optical image, the image matching system may use the Sobel operator to calculate the first gradient, which can be represented by Equation 6: Equation 6:

[0129] G o = arctan(G h / G v )

[0130] where G m is the gradient magnitude of the first gradient, and G o is the gradient direction of the first gradient. is a Gaussian kernel with a standard deviation of β j , and are horizontal and vertical rectangular windows, respectively.

[0131] Again, for an SAR image, the image matching system may use the ROEWA operator to calculate the second gradient, which can be represented by Equation 7: Equation 7:

[0132]

[0133] where is the horizontal gradient in the second gradient, and is the vertical gradient in the second gradient. It can be represented by Equation 8: Equation 8:

[0134]

[0135] It can be represented by Equation 9: Equation 9:

[0136]

[0137] where is the horizontal operator, is the vertical operator, and M and N are related to the scale parameter a iRelated, represents the size of the processing window, (x, y) represents the center point position, and I represents the pixel intensity.

[0138] Step 22: Determine the feature points corresponding to the optical image and the SAR image respectively according to their respective corresponding gradients.

[0139] Continuing to combine the above example and Figure 6 , after determining the first gradient, the image matching system can determine the feature points corresponding to the optical image based on the first gradient (for the convenience of distinction and description, called the first feature points). After determining the second gradient, the image matching system can determine the second feature points corresponding to the SAR image based on the second gradient (similarly, called the second feature points).

[0140] Feature points can be understood as key points extracted from the image that are unique and stable. For example, feature points can be uniquely identified from different perspectives, lighting conditions, or scales. Additionally, even if the image undergoes slight deformations (such as rotation, scaling, etc.), the characteristics such as the position of the feature points remain almost the same. Moreover, in different images, the feature points of the same object should be reliably detected.

[0141] It can be understood that the number of feature points can be one or multiple, and usually multiple. For example, the first feature points can be understood as a set of feature points corresponding to the optical image. For example, the second feature points can be understood as a set of feature points corresponding to the SAR image.

[0142] In some embodiments, combining the above example and Figure 6 it can be known that, taking the optical image as an example, after obtaining the first gradient through the Sobel operator processing, the image matching system can determine the first feature points based on the Harris scale - space feature point detection method.

[0143] For example, first, for each pixel a i and β j , the image matching system can respectively construct two Harris spatial matrices M s (a i ) and M O (β j ). Then, the image matching system can obtain the response functions R S (a i ) and R O (β j ) by calculating the determinant det and the trace tr of the matrix. Finally, by adjusting the scale parameters a i and β j , the image matching system can construct the Harris scale spaces R O and R S for detecting feature points.

[0144] M S (a i ) can be represented by Equation 10, Equation 10:

[0145]

[0146] where, is the Gaussian kernel function for smoothing the image; is the gradient in the horizontal direction; is the gradient in the vertical direction.

[0147] R S (a i ) can be represented by Equation 11, Equation 11:

[0148] R S (a i ) = det(M S (a i )) - d·tr(M S (a i )) 2

[0149] where d is an arbitrary parameter for adjusting the threshold of the response function R S (a i ).

[0150] M O (β j ) can be represented by Equation 12, Equation 12:

[0151]

[0152] where, is the Gaussian kernel function for smoothing the image; is the gradient in the horizontal direction; is the gradient in the vertical direction.

[0153] R O (β j ) can be represented by Equation 13, Equation 13:

[0154] R O (β j ) = det(M O (β j )) - d·tr(M o (β j )) 2

[0155] where d is an arbitrary parameter for adjusting the threshold of the response function R O (β j ).

[0156] In some embodiments, the image matching system may perform line fitting on the edge information corresponding to the optical image and the SAR image respectively to obtain the edge lines corresponding to the optical image and the SAR image respectively.

[0157] Correspondingly, step 22 above may include: determining the feature points corresponding to the optical image and the SAR image respectively from the edge lines corresponding to them according to their respective gradients.

[0158] Exemplarily, after obtaining the first edge information and the second edge information, the image matching system may perform line fitting on the first edge information to obtain the first edge line corresponding to the optical image; or perform line fitting on the second edge information to obtain the edge line corresponding to the SAR image.

[0159] The image matching system may determine the first feature points from the first edge line according to the first gradient; or determine the second feature points from the second edge line according to the second gradient.

[0160] For example, after fitting the corresponding edge lines, the image matching system may use the intersection points, corner angles, turning points (corner points), and relative position relationships between the line segments of the edge lines, etc., to use the intersection points, inflection points, etc. between roads as feature points. So as to perform matching from the dimensions of intersection points and inflection points, thereby assisting the matching of two images (optical image and SAR image).

[0161] It is possible to utilize the significant continuity, smoothness, and tortuosity of the road structure, etc., as constraints to extract the road structure, and use the intersection points, turning points, etc. in the road structure as feature points to assist in the matching of the optical image and the SAR image. It can improve the accuracy and reliability of the matching.

[0162] Especially when facing extremely difficult situations to find matching point pairs, such as in high-difficulty areas where problems such as excessive noise, layover, and foreshortening are serious due to terrain and SAR imaging effects, by using the extracted linear features such as roads and river channels as a reference basis to find effective feature points, the matching success rate of heterogeneous images (optical image and SAR image) can be improved.

[0163] Step 23: Match the feature points corresponding to the optical image and the SAR image respectively to obtain the first matching result.

[0164] Continuing with the above example and Figure 6 , the image matching system may perform feature point matching between the first feature points and the second feature points to obtain the first matching result.

[0165] Combined with the above analysis of steps 21 to 23, it can be seen that in this embodiment, the image matching system determines feature points by combining gradients, and determines the first matching result by means of feature point matching. This can make the determined first matching result have high accuracy and reliability.

[0166] In some embodiments, step 23 may include the following steps 231 and 232:

[0167] Step 231: Determine the main direction and feature descriptor (which can also be simply referred to as descriptor) corresponding to each feature point in the optical image and the SAR image.

[0168] Among them, the main direction can be understood as the main direction of the pixel gradient in the local area around the feature point. The main direction defines a local coordinate system for the feature point, so that the feature descriptor is invariant to rotational changes. For example, even if the image (such as an optical image, SAR image) rotates, as long as the main direction of the feature point can be correctly calculated, its corresponding feature descriptor will not change, so reliable matching can be achieved under different perspectives.

[0169] The feature descriptor can be understood as a mathematical representation used to quantify the local information around the feature point in an image (such as an optical image, SAR image). It extracts information such as pixel distribution, gradient direction, and color near the feature point, and generates a vector or matrix to describe the characteristics of the feature point. That is to say, the feature descriptor can provide a compact and discriminative representation form for the feature point, so as to perform matching between different images (such as optical images, SAR images).

[0170] Continuing to combine the above examples and Figure 6 , after feature point detection, the image matching system can determine the main direction and feature descriptor corresponding to each feature point.

[0171] For example, for the optical image, the matching image system can determine the first main direction and the first feature descriptor corresponding to the first feature point. For the SAR image, the matching image system can determine the second main direction and the second feature descriptor corresponding to the second feature point.

[0172] In some embodiments, after detecting the feature points, the image matching system can first preliminarily modify the positions of the feature points using the position information of each pixel point. Then, the image matching system can describe the feature points using the Gradient Location and Orientation Histogram (GLOH) descriptor to obtain the feature descriptor.

[0173] Among them, GLOH can divide the circular neighborhood with the feature point as the center. For example, its radius is 12α, and logarithmic polar coordinate sectors (17 position intervals) are used to create the feature descriptor. Considering the gradient inversion, the gradient direction is quantized into eight intervals, and the length of the feature descriptor is 136.

[0174] To make the feature descriptor more discriminative, the image matching system can use multiple image patches to construct the feature descriptor. The larger the neighborhood used to construct the feature descriptor, the more structural information the feature descriptor will contain, and the higher the possibility of generating a stable registration result.

[0175] For example, the image matching system can use three circular neighborhoods of GLOH with sizes {8α, 12α, 16α} respectively to construct the feature descriptor.

[0176] Step 232: Match the feature points corresponding to the optical image and the SAR image respectively according to their respective corresponding principal directions and feature descriptors to obtain the first matching result.

[0177] Continuing to combine the above example and Figure 6 , after obtaining the feature descriptors and principal directions corresponding to the optical image and the SAR image respectively, the feature descriptors and principal directions of the two modalities can be combined for feature point matching to obtain the corresponding first matching result.

[0178] Based on the above analysis of Step 231 and Step 232, in this embodiment, by combining the feature descriptors and principal directions of the two modalities for feature point matching, reliable matching can be achieved not only between different images (such as optical images and SAR images), but also under different perspectives.

[0179] In some embodiments, Step 232 may include the following Sub-step 1 and Sub-step 2:

[0180] Sub-step 1: For any feature point in the optical image, based on the principal direction and feature descriptor of the any feature point, determine the respective neighbor distances corresponding to the two neighbor feature points closest to the any feature point from the feature points of the SAR image. The neighbor distance includes the first neighbor distance and the second neighbor distance, the first neighbor distance is less than the second neighbor distance, and the first neighbor feature point is the feature point corresponding to the first neighbor distance.

[0181] For example, for feature point A in an optical image, the image matching system can determine two neighbor feature points closest to feature point A from the feature points of the SAR image: feature point A1 and feature point A2. The distance between feature point A and feature point A1 is L1. The distance between feature point A and feature point A2 is L2. L1 < L2. That is, feature point A1 is the first neighbor feature point, and feature point A2 is the second neighbor feature point. L1 is the first neighbor distance, and L2 is the second neighbor distance.

[0182] Sub-step 2: If the ratio between the first neighbor distance and the second neighbor distance is less than a preset threshold, then any feature point and the first neighbor feature point are determined as matching feature points to obtain a first matching result.

[0183] Continuing with the above example, if L1 / L2 < the preset threshold, then feature point A1 is determined as the feature point matching feature point A. And so on, thus obtaining a first matching result.

[0184] Combining the above analysis of sub-step 1 and sub-step 2, it can be seen that in this embodiment, by combining the distance ratio of the two closest neighbor feature points to determine the matching feature points, it is possible to avoid matching feature points that may be incorrect (such as due to noise or repeated textures), so as to effectively reduce the number of false matches. Thereby improving the accuracy and reliability of the matching.

[0185] S504: Perform filtering processing on the optical image and the SAR image respectively to obtain their respective corresponding filtered images.

[0186] Continuing with Figure 6 , the image matching system can perform Nonlinear Diffusion Filtering processing on the optical image and the SAR image respectively to obtain a filtered image corresponding to the optical image and a filtered image corresponding to the SAR image.

[0187] Nonlinear diffusion filtering is an advanced technique for image processing. Its core idea is to smooth the image by simulating the heat conduction process in physics while retaining important edge and detail information. Different from traditional linear filtering methods (such as Gaussian blur), nonlinear diffusion filtering can dynamically adjust the smoothing degree according to the local characteristics of the image, thereby avoiding blurring the key structure of the image while removing noise.

[0188] The nonlinear diffusion equation can be expressed by Equation 14, Equation 14:

[0189]

[0190] where c(x, y, t) is the diffusion function and div is the divergence operator. ∇ and Δ are the gradient and Laplacian operators respectively.

[0191] When c(x, y, t) is a constant, Equation 14 can be simplified to c(x, y, t) can also be expressed as

[0192] where ||·|| is the modulus operation, is the filtered map, g(·) is the edge function, and it can be represented by Equation 15, Equation 15:

[0193]

[0194] where K is a constant that controls the diffusion rate. Relatively speaking, the larger the value of K, the less edge information is retained. In Equation 15, g(·) is specifically a monotonically decreasing non-negative function, and the function value changes along the image gradient direction.

[0195] S505: Match the respective filtered images to obtain a second matching result.

[0196] Continuing to combine the above example and Figure 6 , regarding the implementation principle of S505, refer to the matching principle of edge information in the above example, which will not be elaborated here.

[0197] S506: Perform a fusion process on the first matching result and the second matching result to obtain a fused matching result.

[0198] Exemplarily, the fused matching result may include the first matching result and the second matching result.

[0199] Therefore, relatively speaking, the amount of data in the fused matching result is much larger than the amount of data obtained by matching from a single dimension of the image. That is, the fused matching result is relatively more abundant and diverse. So it can improve the accuracy and reliability of determining the final image matching result subsequently.

[0200] S507: Perform a false match filtering process on the fused matching result to obtain an image matching result, where the false match filtering process includes sample consensus (FSC) and / or affine transformation.

[0201] Continuing to combine the above example and Figure 6 , after obtaining the fused matching result, the image matching result can be obtained by means of a false match filtering process to delete the incorrect matching results in the fused matching result.

[0202] For example, the image matching system can use the sample consensus algorithm to remove the outliers in the fused matching result. The image matching system can also correct the incorrect matching results through the affine transformation model estimated from the correctly matched corresponding points.

[0203] Based on the above analysis of S506 and S507, in this embodiment, after obtaining the fusion matching result, the image matching system can perform mis - matching filtering on the fusion matching result to eliminate the wrong matching result. Thus, the accuracy and reliability of the matching can be improved.

[0204] It should be noted that the above examples are only used to exemplarily illustrate the possible implementation manners of the image matching method in this specification, and should not be construed as a limitation on the implementation manners of the image matching method in this specification. Exemplarily, based on the above technical concept, some of the above technical features can be combined to obtain a new embodiment; new technical features can also be added based on the above examples to obtain a new embodiment; some technical features can also be reduced based on the above examples to obtain a new embodiment; some of the technical features in the above examples can be replaced with other technical features; the order of some of the technical features in the above examples can also be adjusted to obtain a new embodiment, and so on. These are not listed one by one here.

[0205] According to the above technical concept, this specification also provides a computer - readable non - transitory storage medium, in which at least one instruction set is stored. When the at least one instruction set is executed by a processor, the steps of the image matching method described in this specification are implemented.

[0206] In some possible embodiments, various aspects of this specification can also be implemented in the form of a program product, which includes program code. When the program product runs on the image matching system 200, the program code is used to cause the image matching system 200 to execute the steps of the image matching method described in this specification. The program product for implementing the above method can be a portable compact disc read-only memory (CD-ROM) that includes program code and can run on the image matching system 200. However, the program product of this specification is not limited to this. In this specification, a readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system. The program product can adopt any combination of one or more readable media. The readable media can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the readable storage medium include: an electrical connection with one or more wires, a portable disc, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. The computer-readable storage medium can include a data signal propagated in a baseband or as part of a carrier wave, in which the readable program code is carried. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable storage medium can also be any readable medium other than the readable storage medium, and this readable medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium can be transmitted with any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above. The program code for performing the operations of this specification can be written in any combination of one or more programming languages, including object-oriented programming languages - such as Java, C++, etc., and also including conventional procedural programming languages - such as the "C" language or similar programming languages. The program code can be executed entirely on the image matching system 200, partially on the image matching system 200, executed as an independent software package, partially on the image matching system 200 and partially on a remote image matching system, or entirely on a remote image matching system 200.

[0207] The above description has been made of specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the figures do not necessarily require a particular order or a sequential order to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0208] In summary, after reading this detailed disclosure, those skilled in the art will understand that the foregoing detailed disclosure may be presented only by way of example and may not be limiting. Although not explicitly stated herein, those skilled in the art can understand that this specification is intended to encompass various reasonable changes, improvements, and modifications to the embodiments. These changes, improvements, and modifications are intended to be proposed by this specification and are within the spirit and scope of the exemplary embodiments of this specification.

[0209] Furthermore, certain terms in this specification have been used to describe embodiments of this specification. For example, "one embodiment", "an embodiment", and / or "some embodiments" mean that the specific features, structures, or characteristics described in connection with that embodiment may be included in at least one embodiment of this specification. Thus, it should be emphasized and understood that two or more references to "an embodiment" or "one embodiment" or "alternative embodiments" in various parts of this specification do not necessarily all refer to the same embodiment. Additionally, the specific features, structures, or characteristics may be appropriately combined in one or more embodiments of this specification.

[0210] It should be understood that in the foregoing description of the embodiments of this specification, for the purpose of helping to understand a feature and for the purpose of simplifying this specification, this specification combines various features in a single embodiment, drawing, or its description. However, this does not mean that the combination of these features is necessary. Those skilled in the art are quite likely to mark out some of the devices as separate embodiments when reading this specification. That is to say, the embodiments in this specification can also be understood as the integration of multiple sub - embodiments. And the content of each sub - embodiment is also valid when it has fewer features than all the features of a single foregoing disclosed embodiment.

[0211] Each patent, patent application, published patent application, and other materials cited herein, such as articles, books, specifications, publications, documents, references, etc. (excluding any historical prosecution documents associated therewith), are hereby incorporated by reference for all purposes relevant hereto, e.g., in the specification and claims of this application. However, in the event of any inconsistency or conflict between the descriptions, definitions, and / or terms of the above materials and those used in this application, the descriptions, definitions, and / or terms used in this application shall control.

[0212] Finally, it should be understood that the embodiments of the application disclosed herein are illustrative of the principles of the embodiments of this specification. Other modified embodiments are also within the scope of this specification. Therefore, the embodiments disclosed in this specification are provided by way of example only and not by way of limitation. Those skilled in the art may implement the application in this specification by taking alternative configurations based on the embodiments in this specification. Accordingly, the embodiments of this specification are not limited to the embodiments precisely described in the application.

Claims

1. An image matching method, comprising: Obtaining an optical image and a SAR image to be matched; Respectively extracting edge information of linear features in the optical image and the SAR image, and matching the edge information corresponding to the optical image and the SAR image respectively to obtain a first matching result; Matching the optical image and the SAR image to obtain a second matching result; And Determining an image matching result between the optical image and the SAR image according to the first matching result and the second matching result.

2. The method according to claim 1, wherein The step of matching the edge information corresponding to the optical image and the SAR image respectively to obtain a first matching result includes: Determining gradients corresponding to respective edge information, where the gradient is used to characterize the change rate of image pixel values in space; Determining feature points corresponding to the optical image and the SAR image respectively according to the respective corresponding gradients; Matching the feature points corresponding to the optical image and the SAR image respectively to obtain the first matching result.

3. The method according to claim 2, wherein The step of matching the feature points corresponding to the optical image and the SAR image respectively to obtain the first matching result includes: Determining main directions and feature descriptors corresponding to respective feature points in the optical image and the SAR image; Matching the feature points corresponding to the optical image and the SAR image respectively according to the respective corresponding main directions and feature descriptors to obtain the first matching result.

4. The method according to claim 3, wherein The step of matching the feature points corresponding to the optical image and the SAR image respectively according to the respective corresponding main directions and feature descriptors to obtain the first matching result includes: For any feature point in the optical image, based on the main direction and feature descriptor of the any feature point, determining neighbor distances corresponding to two neighbor feature points closest to the any feature point from the feature points of the SAR image; and If the ratio between the first neighbor distance and the second neighbor distance is less than a preset threshold, determining the any feature point and the first neighbor feature point as matching feature points to obtain the first matching result; Wherein, the neighbor distance includes the first neighbor distance and the second neighbor distance, the first neighbor distance is less than the second neighbor distance, and the first neighbor feature point is the feature point corresponding to the first neighbor distance.

5. The method according to claim 2, wherein The method further includes: Performing line fitting processing on the edge information corresponding to the optical image and the SAR image respectively to obtain edge lines corresponding to the optical image and the SAR image respectively; And, the step of determining feature points corresponding to the optical image and the SAR image respectively according to the respective corresponding gradients includes: determining feature points corresponding to the optical image and the SAR image respectively from the respective corresponding edge lines according to the respective corresponding gradients.

6. The method according to any one of claims 1 to 5, wherein Extracting edge information of linear features in the optical image includes: Determining a binary image of the optical image; Extracting edge information corresponding to linear features in the optical image from the binary image of the optical image.

7. The method according to claim 6, wherein, Extracting the edge information corresponding to the linear features in the optical image from the binary image of the optical image includes: Extracting the gray value corresponding to each pixel point from the binary image of the optical image; Determining the first edge point of the linear feature in the optical image according to the extracted gray values; and Searching in the binary image of the optical image according to the first edge point to obtain the edge information corresponding to the linear feature in the optical image.

8. The method according to any one of claims 1 to 5, wherein Determining the image matching result between the optical image and the SAR image according to the first matching result and the second matching result includes: Performing a fusion process on the first matching result and the second matching result to obtain a fused matching result; and Performing a false matching filtering process on the fused matching result to obtain the image matching result, where the false matching filtering process includes sample consensus and / or affine transformation.

9. The method according to any one of claims 1 to 5, wherein The linear features include at least one of roads, rivers, ditches, and buildings.

10. An image matching system, comprising: At least one storage medium storing at least one instruction set for performing image matching; At least one processor communicatively connected to the at least one storage medium, wherein when the at least one processor runs, it reads the at least one instruction set and executes the method according to any one of claims 1 to 9 according to the instructions of the at least one instruction set.

Images