Extraterrestrial celestial body star catalogue detection image similarity evaluation method

Abstract

The invention provides an extraterrestrial celestial body star catalogue detection image similarity evaluation method, which comprises the following steps of: constructing a semantic-level expected image according to semantic information for describing an extraterrestrial celestial body star catalogue; respectively carrying out feature extraction and analysis on the semantic-level expected image and the acquired shot image of the extraterrestrial celestial body star catalogue to obtain evaluation parameters; wherein the evaluation parameters comprise the number of feature points, feature similarity, pixel error, cosine similarity and Euclidean distance; and determining a similar result of the semantic-level expected image and the shot image according to the evaluation parameters. According to the scheme, the similarity between the shot image and the semantic-level abstract description expected image is autonomously evaluated on the detector, and the credibility and reliability of similarity evaluation are improved.

Description

Technical Field

[0001] The invention relates to the field of aerospace detection technology, and in particular to a method for evaluating the similarity of extraterrestrial body star catalog detection images. Background Art

[0002] Taking photos of the star charts of extraterrestrial objects is one of the main tasks of extraterrestrial star chart exploration. It usually requires taking images of elements including landers, probes, and star chart environments. The photography process is generally implemented by ground-assisted planning and design. Due to the low efficiency of space-ground data transmission, such as the communication delay between the Earth and Mars of about 20 minutes, when multiple images of different scenes need to be taken, the use of ground-assisted methods is not only inefficient, but also consumes a lot of time in the transmission of images and instructions, wasting the precious working life of the probe. However, before shooting, the expected image is unknown, and the designer only has a semantic-level abstract description of the expected image. It is difficult to autonomously evaluate whether the captured image is similar to the expected image described in the semantic-level abstract description on the probe. Therefore, it is urgent to provide an image similarity evaluation method for extraterrestrial star chart exploration. Summary of the invention

[0003] The present invention provides a method for evaluating the similarity of extraterrestrial body star catalog detection images, which realizes the autonomous evaluation of the similarity between a captured image and an expected image described in a semantic level abstract manner on the detector.

[0004] In a first aspect, the present invention provides a method for evaluating the similarity of extraterrestrial object star catalog detection images, the method comprising:

[0005] Constructing semantic level expected images based on semantic information used to describe the catalog of extraterrestrial objects;

[0006] Extracting and analyzing features of the semantic-level expected image and the captured images of the extraterrestrial object catalog, respectively, to obtain evaluation parameters; wherein the evaluation parameters include the number of feature points, feature similarity, pixel error, cosine similarity and Euclidean distance;

[0007] A similarity result between the semantic-level expected image and the captured image is determined according to the evaluation parameter.

[0008] Preferably, constructing a semantic level expected image according to the semantic information used to describe the extraterrestrial object catalog includes:

[0009] According to the semantic information, obtaining the elements included in the extraterrestrial object catalog and the size, position information and geometric relationship of each element;

[0010] Different elements are filled with different colors, and the semantic-level expected image is generated according to the position information of each element; wherein the semantic-level expected image includes different elements filled with different colors.

[0011] Preferably, the elements include a lander, an ascender, a sailboard, the surface of an extraterrestrial body, and space.

[0012] Preferably, the feature extraction and analysis of the semantic level expected image and the captured image of the extraterrestrial object catalog are respectively performed to obtain evaluation parameters, including:

[0013] Performing semantic segmentation on the captured image to extract the same elements as those in the semantic-level expected image;

[0014] Filling different elements with different colors to generate a semantic-level captured image; wherein the semantic-level expected image includes different elements filled with different colors, and the same elements in the semantic-level expected image and the semantic-level captured image are filled with the same color;

[0015] Performing feature extraction on the semantic-level expected image and the semantic-level captured image respectively to obtain a first feature point set of the semantic-level expected image and a second feature point set of the semantic-level captured image;

[0016] Counting a first number of pixel points of each element in the semantic-level expected image; and counting a second number of pixel points of each element in the semantic-level captured image;

[0017] Obtaining a first average value of pixel coordinates of each pixel point of the element in the semantic level expected image; and obtaining a second average value of pixel coordinates of each pixel point of the element in the semantic level captured image;

[0018] The evaluation parameter is obtained according to the first feature point set, the second feature point set, the first number and the second number of each of the elements, and the first average value and the second average value of each of the elements.

[0019] Preferably, obtaining the evaluation parameter according to the first feature point set, the second feature point set, the first quantity and the second quantity of each of the elements, and the first average value and the second average value of each of the elements includes:

[0020] Calculate feature similarity according to the number of feature points in the first feature point set, the second feature point set, and the first feature point set;

[0021] Calculate a pixel error according to the first number and the second number of each of the elements;

[0022] The cosine similarity and the Euclidean distance are calculated respectively according to the first average value and the second average value of each of the elements.

[0023] Preferably, the feature similarity is determined by the following formula:

[0024]

[0025] Among them, r s is the feature similarity; m is the number of feature points; S tj is the two-dimensional pixel coordinate of the jth feature point in the first feature point set; S o’j is the two-dimensional pixel coordinates of the j-th feature point in the second feature point set;

[0026] The pixel error is determined by the following formula:

[0027]

[0028] Among them, r l is the pixel error; n is the total number of elements; l ti is the first quantity of the i-th element; l o’i is the second number of the i-th element.

[0029] Preferably, determining the similarity result between the semantic-level expected image and the captured image according to the evaluation parameter comprises:

[0030] Determine whether the number of feature points is greater than a preset number of feature points, and obtain a first determination result;

[0031] Determine whether the feature similarity is less than a preset first similarity threshold, and obtain a second determination result;

[0032] Determine whether the pixel error is less than a preset error threshold, and obtain a third determination result;

[0033] Determine whether the cosine similarity is greater than a preset second similarity threshold, and obtain a fourth determination result;

[0034] Determine whether the Euclidean distance is less than a preset distance threshold, and obtain a fifth determination result;

[0035] When the first judgment result, the second judgment result, the third judgment result, the fourth judgment result and the fifth judgment result are all yes, it is determined that the similarity result is similar.

[0036] In a second aspect, the present invention further provides an extraterrestrial body star catalog detection image similarity evaluation device, comprising:

[0037] A construction module, for constructing a semantic-level expected image according to semantic information for describing an extraterrestrial object catalog;

[0038] An evaluation module is used to extract and analyze features of the semantic-level expected image and the captured image of the extraterrestrial body catalog, respectively, to obtain evaluation parameters; wherein the evaluation parameters include the number of feature points, feature similarity, pixel error, cosine similarity and Euclidean distance; and determine the similarity result between the semantic-level expected image and the captured image based on the evaluation parameters.

[0039] In a third aspect, the present invention further provides a computing device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements any of the above-mentioned methods for evaluating the similarity of images of extraterrestrial body catalog detection.

[0040] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute any of the above-mentioned methods for evaluating the similarity of images of extraterrestrial body catalog detection.

[0041] In a fifth aspect, an embodiment of the present invention further provides a computer program product, comprising computer instructions, which, when executed by a processor, implement the steps of the method described in any first aspect of this specification.

[0042] The present invention provides an extraterrestrial body catalog detection image similarity evaluation method. The method constructs a semantic level expected image through semantic information used to describe the extraterrestrial body catalog, and then obtains evaluation parameters by extracting and analyzing the features of the semantic level expected image and the captured image. In this way, the detector realizes the similarity evaluation of the captured image and the semantic information based on the evaluation parameters of four different dimensions under the condition of only the semantic level expected image, thereby solving the problem that the unknown semantic level expected image is difficult to visualize and the captured image cannot be quantitatively compared with the expected image. At the same time, the similarity of the semantic level expected image and the captured image is evaluated in four different dimensions, thereby improving the credibility and reliability of the similarity evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1It is a flow chart of a method for evaluating the similarity of extraterrestrial body star catalog detection images provided by one embodiment of the present invention;

[0045] Figure 2 is a semantic level expected image provided by an embodiment of the present invention;

[0046] Figure 3 is a photographed image provided by an embodiment of the present invention;

[0047] Figure 4 is a semantic-level captured image provided by an embodiment of the present invention;

[0048] Figure 5 is a hardware architecture diagram of a computing device provided by an embodiment of the present invention;

[0049] Figure 6 It is a structural diagram of an extraterrestrial body star catalog detection image similarity evaluation device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0051] Please refer to Figure 1 The embodiment of the present invention provides a method for evaluating the similarity of extraterrestrial object star catalog detection images, comprising:

[0052] Step 100, constructing a semantic level expected image based on semantic information used to describe the extraterrestrial object catalog;

[0053] Step 102, extracting and analyzing features of the semantic level expected image and the captured images of the extraterrestrial object catalog, respectively, to obtain evaluation parameters; wherein the evaluation parameters include the number of feature points, feature similarity, pixel error, cosine similarity and Euclidean distance;

[0054] Step 104 : determining similarity results between the semantic-level expected image and the captured image according to the evaluation parameters.

[0055] In the present invention, a semantic-level expected image is constructed by using semantic information for describing a star catalog of an extraterrestrial body, and then evaluation parameters are obtained by extracting and analyzing features of the semantic-level expected image and the captured image. In this way, the detector realizes similarity evaluation between the captured image and the semantic information based on evaluation parameters of four different dimensions when only the semantic-level expected image is available, thereby solving the problems that an unknown semantic-level expected image is difficult to visualize and the captured image cannot be quantitatively compared with the expected image. At the same time, the similarity between the semantic-level expected image and the captured image is evaluated in four different dimensions, thereby improving the credibility and reliability of the similarity evaluation.

[0056] Described below Figure 1 How the various steps are performed.

[0057] In step 100, a semantic level expected image is constructed according to the semantic information used to describe the extraterrestrial object catalog, including:

[0058] According to the semantic information, the elements included in the extraterrestrial object catalog and the size, position information and geometric relationship of each element are obtained;

[0059] Different elements are filled with different colors, and a semantic-level expected image is generated according to the position information of each element; wherein the semantic-level expected image includes different elements filled with different colors.

[0060] In the present invention, by processing the semantic information of the known extraterrestrial body catalog, the semantic-level abstract description is constructed into a specific image, so that the known semantic information can be directly compared with the captured image through the semantic-level expected image, thereby solving the problem that the unknown semantic-level expected image is difficult to visualize and the captured image cannot be quantitatively compared with the expected image.

[0061] In the present invention, different elements are filled with different colors to facilitate the visual comparison of two images later.

[0062] In a preferred embodiment, the elements include a lander, an ascender, a sailboard, the surface of an extraterrestrial body, and space.

[0063] In step 102, feature extraction and analysis are performed on the semantic level expected image and the captured image of the extraterrestrial object catalog to obtain evaluation parameters, including:

[0064] Perform semantic segmentation on the captured image to extract the same elements as those in the expected image at the semantic level;

[0065] Filling different elements with different colors to generate a semantic-level captured image; wherein the semantic-level expected image includes different elements filled with different colors, and the same elements in the semantic-level expected image and the semantic-level captured image are filled with the same color;

[0066] Extracting features from the semantic-level expected image and the semantic-level captured image respectively to obtain a first feature point set of the semantic-level expected image and a second feature point set of the semantic-level captured image;

[0067] Counting a first number of pixel points of each element in the semantic-level expected image; and counting a second number of pixel points of each element in the semantic-level captured image;

[0068] Obtaining a first average value of pixel coordinates of each element in the semantic level expected image; and obtaining a second average value of pixel coordinates of each element in the semantic level captured image;

[0069] An evaluation parameter is obtained according to the first feature point set, the second feature point set, the first number and the second number of each element, and the first average value and the second average value of each element.

[0070] It should be noted that in the present invention, the semantic-level expected image has the same size as the captured image and the semantic-level captured image.

[0071] In the present invention, semantic segmentation is performed on the captured image to obtain the same elements as the semantic level expected image, and then the semantic level captured image is constructed by the same method as step 100, so that the colors of the same elements in the semantic level expected image and the semantic level captured image are the same, so that the two images are aligned. Then, feature extraction is performed on the semantic level expected image and the semantic level captured image to obtain feature point sets of the two images respectively; then, pixel-based evaluation parameters are obtained by the pixel points and pixel point coordinates of the elements in the two images, and evaluation parameters based on multiple dimensions are obtained.

[0072] In a preferred embodiment, obtaining the evaluation parameters according to the first feature point set, the second feature point set, the first number and the second number of each element, and the first average value and the second average value of each element includes:

[0073] Calculate the feature similarity according to the number of feature points in the first feature point set, the second feature point set, and the first feature point set;

[0074] Calculate a pixel error based on the first number and the second number of each element;

[0075] According to the first average value and the second average value of each element, the cosine similarity and the Euclidean distance are calculated respectively.

[0076] In the present invention, the feature similarity used to describe the scale space similarity between the semantic level expected image and the semantic level captured image is calculated based on the feature point set; the pixel error used to describe the element size similarity between the semantic level expected image and the semantic level captured image is calculated based on the number of each element in the two images; the cosine similarity used to describe the angle similarity between the semantic level expected image and the semantic level captured image, and the Euclidean distance used to describe the similarity of the element distribution between the two images are calculated based on the average value of the two-dimensional pixel coordinates of all the pixels included in each element in the two images. In this way, the evaluation parameters of four dimensions are calculated through the feature point set, the number of pixels included in each element, and the average value of the coordinates of all the pixels included in each element, and the similarity evaluation is performed based on these four evaluation parameters, thereby improving the credibility and reliability of the similarity evaluation.

[0077] In a preferred embodiment, the feature similarity is determined by the following formula:

[0078]

[0079] Among them, r s is the feature similarity; m is the number of feature points; S tj is the two-dimensional pixel coordinate of the jth feature point in the first feature point set; S o’j is the two-dimensional pixel coordinate of the j-th feature point in the second feature point set;

[0080] The pixel error is determined by the following formula:

[0081]

[0082] Among them, r l is the pixel error; n is the total number of elements; l ti is the first number of the i-th element; l o’i is the second number of the i-th element.

[0083] In a preferred embodiment, the cosine similarity is determined by the following formula:

[0084]

[0085] Among them, r c is the cosine similarity; n is the total number of elements; c ti is the first average value of the pixel coordinates of all the pixels included in the i-th element; c o’i is the second average value of the pixel coordinates of all the pixels included in the i-th element.

[0086] In a preferred embodiment, the Euclidean distance is determined by the following formula:

[0087]

[0088] Among them, r e is the Euclidean distance; n is the total number of elements; c ti is the first average value of the pixel coordinates of all the pixels included in the i-th element; c o’i is the second average value of the pixel coordinates of all the pixels included in the i-th element.

[0089] In step 104, determining similarity results between the semantic level expected image and the captured image according to the evaluation parameters includes:

[0090] Determine whether the number of feature points is greater than a preset number of feature points, and obtain a first determination result;

[0091] Determine whether the feature similarity is less than a preset first similarity threshold, and obtain a second determination result;

[0092] Determine whether the pixel error is less than a preset error threshold, and obtain a third determination result;

[0093] Determine whether the cosine similarity is greater than a preset second similarity threshold, and obtain a fourth determination result;

[0094] Determine whether the Euclidean distance is less than a preset distance threshold, and obtain a fifth determination result;

[0095] When the first judgment result, the second judgment result, the third judgment result, the fourth judgment result and the fifth judgment result are all yes, the similar results are determined to be similar.

[0096] In the present invention, when the above judgment results are all yes, the semantic level expected image and the captured image are considered similar, and if any of the judgment results is no, the semantic level expected image and the captured image are considered dissimilar. In this way, the present invention realizes autonomous judgment on the detector whether the semantic level expected image and the captured image are similar, without the need for inefficient ground-based auxiliary methods for evaluation, reducing the time spent on image and command transmission, while improving evaluation efficiency and reliability.

[0097] In a specific embodiment, step 1, constructing Figure 2 The semantic level expected image P shown t :

[0098] According to the semantic information of the target extraterrestrial object catalog, determine the elements (lander, sailboard, ascender, ground of the extraterrestrial object, space) included in the semantic information and the size, position information and geometric relationship of each element;

[0099] Assign a unique color to each element: yellow for the lander, cyan for the sailboard, purple for the ascender, red for the ground of an extraterrestrial body, and black for the space. Fill them with the corresponding colors to get the following: Figure 2 The semantic level expected image P shown t ;

[0100] Step 2: extract and analyze the features of the semantic level expected image and the captured images of the extraterrestrial object catalog to obtain evaluation parameters, including:

[0101] (1) For Figure 3 The captured image P shown o Perform semantic segmentation and extract the expected image P at the semantic level t the same elements in P o With P t The colors of the corresponding elements in the image are consistent, and the segmented semantic-level captured image P is obtained. o' , the semantic level captured image after semantic segmentation is as follows Figure 4 As shown;

[0102] (2) Using SIFT feature extraction algorithm, we can extract the semantic level expected image P t and semantic level captured image P o' The corresponding feature point set S t {s t1 ,s t2 ,…,s tm} and S o' {s o'1 ,s o'2 ,…,s o'm};

[0103] (3) Statistical semantic level expected image P t The first number of pixels L for each element in t {l t1 ,l t2 ,…,l tn}, l tn is the number of pixels of the nth element; and counts the semantic level captured image P o' The second number L of pixels of each element in o' {l o'1 ,l o'2 ,…,l o'n}, where l o'n is the number of pixels of the nth element;

[0104] (4) Statistical semantic level expected image P t The average value C of the two-dimensional pixel coordinates of all pixels of each element in t {c t1 ,ct2 ,…,c tn}, where c tn is the mean value of the two-dimensional pixel coordinates of all pixels of the nth element (c tnx ,c tny ); Statistical semantic level captured image P o' The average value C of the two-dimensional pixel coordinates of all pixels of each element in o' {c o'1 ,c o'2 ,…,c o'n}, where c o'n is the mean value of the two-dimensional pixel coordinates of all pixels of the nth element (c o'nx ,c o'ny );

[0105] (5) Statistics S t and S o' The number of SIFT feature points r ns =7, and the feature similarity r is calculated by formula (1) s =15.2, used to describe the scale space similarity of two images;

[0106] (6) Calculate L using formula (2) t and L o' The pixel error r l =0.11, used to describe the similarity of element sizes between two images;

[0107] (7) Calculate C using formula (3) t and C o' The cosine similarity r c =0.89, used to describe the angular similarity of two images;

[0108] (8) Calculate C using formula (4) t and C o' The Euclidean distance r e =31.8, used to describe the similarity of element distribution between two images;

[0109] Step 3: Set the similarity evaluation index judgment threshold and preset the number of feature points ε ns =5, preset the first similarity threshold ε s =10, preset error threshold ε l =0.15, preset the second similarity threshold ε c =0.97, preset distance threshold ε e =10, then r ns >ε ns 、r s >ε s (Not meeting the conditions), r l <εl 、r c <ε c (Not meeting the conditions), r e >ε e (Not satisfying the condition), 3 of the 5 judgment results do not satisfy the condition, so the captured image P o and semantic level expected image P t Not similar.

[0110] like Figure 5 , Figure 6 As shown, an embodiment of the present invention provides an apparatus for evaluating the similarity of extraterrestrial star catalog detection images. The apparatus embodiment can be implemented by software, or by hardware or a combination of software and hardware. From the hardware level, Figure 5 As shown, it is a hardware architecture diagram of a computing device where an extraterrestrial body star catalog detection image similarity evaluation device provided by an embodiment of the present invention is located, except Figure 5 In addition to the processor, memory, network interface, and non-volatile memory shown in the figure, the computing device in which the device is located in the embodiment may also generally include other hardware, such as a forwarding chip responsible for processing messages, etc. Taking software implementation as an example, Figure 6 As shown, as a device in a logical sense, the CPU of the computing device in which it is located reads the corresponding computer program in the non-volatile memory into the memory and runs it. This embodiment provides an extraterrestrial body star catalog detection image similarity evaluation device, including:

[0111] A construction module 600 is used to construct a semantic level expected image according to the semantic information used to describe the extraterrestrial object catalog;

[0112] Evaluation module 602 is used to extract and analyze the features of the semantic level expected image and the captured images of the extraterrestrial star catalog, respectively, to obtain evaluation parameters; wherein the evaluation parameters include the number of feature points, feature similarity, pixel error, cosine similarity and Euclidean distance; and determine the similarity results of the semantic level expected image and the captured image according to the evaluation parameters.

[0113] In some specific implementations, the construction module 600 may be used to execute the above step 100 , and the evaluation module 602 may be used to execute the above steps 102 and 104 .

[0114] In some specific implementations, the construction module 600 is further configured to perform the following operations:

[0115] According to the semantic information, the elements included in the extraterrestrial body catalog and the size, position information and geometric relationship of each element are obtained; wherein the elements include the lander, the ascender, the sailboard, the ground of the extraterrestrial body and the space;

[0116] Different elements are filled with different colors, and a semantic-level expected image is generated according to the position information of each element; wherein the semantic-level expected image includes different elements filled with different colors.

[0117] In some specific implementations, the evaluation module 602 is further configured to perform the following operations:

[0118] Perform semantic segmentation on the captured image to extract the same elements as those in the expected image at the semantic level;

[0119] Filling different elements with different colors to generate a semantic-level captured image; wherein the semantic-level expected image includes different elements filled with different colors, and the same elements in the semantic-level expected image and the semantic-level captured image are filled with the same color;

[0120] Extracting features from the semantic-level expected image and the semantic-level captured image respectively to obtain a first feature point set of the semantic-level expected image and a second feature point set of the semantic-level captured image;

[0121] Counting a first number of pixel points of each element in the semantic-level expected image; and counting a second number of pixel points of each element in the semantic-level captured image;

[0122] Obtaining a first average value of pixel coordinates of each element in the semantic level expected image; and obtaining a second average value of pixel coordinates of each element in the semantic level captured image;

[0123] An evaluation parameter is obtained according to the first feature point set, the second feature point set, the first number and the second number of each element, and the first average value and the second average value of each element.

[0124] In some specific implementations, the evaluation module 602 is further configured to perform the following operations:

[0125] The feature similarity is calculated based on the number of feature points in the first feature point set, the second feature point set, and the first feature point set; the feature similarity is determined by the following formula:

[0126]

[0127] Among them, r s is the feature similarity; m is the number of feature points; S tj is the two-dimensional pixel coordinate of the jth feature point in the first feature point set; S o’j is the two-dimensional pixel coordinate of the j-th feature point in the second feature point set;

[0128] A pixel error is calculated based on the first number and the second number of each element; the pixel error is determined by the following formula:

[0129]

[0130] Among them, r l is the pixel error; n is the total number of elements; l ti is the first number of the i-th element; l o’i is the second number of the i-th element;

[0131] According to the first average value and the second average value of each element, the cosine similarity and the Euclidean distance are calculated respectively.

[0132] In some specific implementations, the evaluation module 602 is further configured to perform the following operations:

[0133] Determine whether the number of feature points is greater than a preset number of feature points, and obtain a first determination result;

[0134] Determine whether the feature similarity is less than a preset first similarity threshold, and obtain a second determination result;

[0135] Determine whether the pixel error is less than a preset error threshold, and obtain a third determination result;

[0136] Determine whether the cosine similarity is greater than a preset second similarity threshold, and obtain a fourth determination result;

[0137] Determine whether the Euclidean distance is less than a preset distance threshold, and obtain a fifth determination result;

[0138] When the first judgment result, the second judgment result, the third judgment result, the fourth judgment result and the fifth judgment result are all yes, the similar results are determined to be similar.

[0139] It is to be understood that the structure illustrated in the embodiment of the present invention does not constitute a specific limitation on a device for evaluating the similarity of images detected by a star catalog of an extraterrestrial object. In other embodiments of the present invention, a device for evaluating the similarity of images detected by a star catalog of an extraterrestrial object may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

[0140] The information interaction, execution process and other contents between the modules in the above-mentioned device are based on the same concept as the embodiment of the method of the present invention. For the specific contents, please refer to the description in the embodiment of the method of the present invention, and no further description is given here.

[0141] An embodiment of the present invention further provides a computing device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, a method for evaluating the similarity of extraterrestrial body star catalog detection images in any embodiment of the present invention is implemented.

[0142] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the processor executes a method for evaluating the similarity of extraterrestrial body star catalog detection images in any embodiment of the present invention.

[0143] An embodiment of the present application also provides a computer program product, which includes a computer program. A processor of a computer device reads the computer program from a computer-readable storage medium, and the processor executes the computer program, so that the computer device executes a method for evaluating the similarity of extraterrestrial body star catalog detection images described in any of the above embodiments.

[0144] Specifically, a system or device equipped with a storage medium can be provided, on which software program code that implements the functions of any of the above-mentioned embodiments is stored, and a computer (or CPU or MPU) of the system or device can be enabled to read and execute the program code stored in the storage medium.

[0145] In this case, the program code itself read from the storage medium can realize the function of any one of the above-mentioned embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.

[0146] The storage medium embodiments for providing the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), a magnetic tape, a non-volatile memory card, and a ROM. Alternatively, the program code can be downloaded from a server computer by a communication network.

[0147] In addition, it should be clear that the functions of any of the above embodiments can be implemented not only by executing the program code read by the computer, but also by enabling an operating system operating on the computer to complete part or all of the actual operations based on instructions from the program code.

[0148] In addition, it can be understood that the program code read from the storage medium is written to a memory provided in an expansion board inserted into the computer or to a memory provided in an expansion module connected to the computer, and then based on the instructions of the program code, a CPU installed on the expansion board or expansion module is enabled to perform part or all of the actual operations, thereby realizing the functions of any of the above-mentioned embodiments.

[0149] It should be noted that, in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "comprise a ..." do not exclude the presence of other identical factors in the process, method, article or device including the elements.

[0150] A person of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk, etc., various media that can store program codes.

[0151] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for evaluating the similarity of extraterrestrial object star catalog detection images, characterized in that: include: Constructing semantic level expected images based on semantic information used to describe the catalog of extraterrestrial objects; Extracting and analyzing features of the semantic-level expected image and the captured images of the extraterrestrial object catalog, respectively, to obtain evaluation parameters; wherein the evaluation parameters include the number of feature points, feature similarity, pixel error, cosine similarity and Euclidean distance; A similarity result between the semantic-level expected image and the captured image is determined according to the evaluation parameter.

2. The method according to claim 1, characterized in that The step of constructing a semantic level expected image according to the semantic information used to describe the extraterrestrial object catalog includes: According to the semantic information, obtaining the elements included in the extraterrestrial object catalog and the size, position information and geometric relationship of each element; Different elements are filled with different colors, and the semantic-level expected image is generated according to the position information of each element; wherein the semantic-level expected image includes different elements filled with different colors.

3. The method according to claim 2, characterized in that The elements include a lander, an ascender, a sailboard, the surface of an extraterrestrial body, and space.

4. The method according to claim 1, characterized in that The feature extraction and analysis of the semantic level expected image and the captured image of the extraterrestrial object catalog are performed respectively to obtain evaluation parameters, including: Performing semantic segmentation on the captured image to extract the same elements as those in the semantic-level expected image; Filling different elements with different colors to generate a semantic-level captured image; wherein the semantic-level expected image includes different elements filled with different colors, and the same elements in the semantic-level expected image and the semantic-level captured image are filled with the same color; Performing feature extraction on the semantic-level expected image and the semantic-level captured image respectively to obtain a first feature point set of the semantic-level expected image and a second feature point set of the semantic-level captured image; Counting a first number of pixel points of each element in the semantic-level expected image; and counting a second number of pixel points of each element in the semantic-level captured image; Obtaining a first average value of pixel coordinates of each pixel point of the element in the semantic level expected image; and obtaining a second average value of pixel coordinates of each pixel point of the element in the semantic level captured image; The evaluation parameter is obtained according to the first feature point set, the second feature point set, the first number and the second number of each of the elements, and the first average value and the second average value of each of the elements.

5. The method according to claim 4, characterized in that The obtaining the evaluation parameter according to the first feature point set, the second feature point set, the first quantity and the second quantity of each of the elements, and the first average value and the second average value of each of the elements comprises: Calculate feature similarity according to the number of feature points in the first feature point set, the second feature point set, and the first feature point set; Calculate a pixel error according to the first number and the second number of each of the elements; The cosine similarity and the Euclidean distance are calculated respectively according to the first average value and the second average value of each of the elements.

6. The method according to claim 5, characterized in that The feature similarity is determined by the following formula: Among them, r s is the feature similarity; m is the number of feature points; S tj is the two-dimensional pixel coordinate of the jth feature point in the first feature point set; S o’j is the two-dimensional pixel coordinates of the j-th feature point in the second feature point set; The pixel error is determined by the following formula: Among them, r l is the pixel error; n is the total number of elements; l ti is the first quantity of the i-th element; l o’i is the second number of the i-th element.

7. The method according to any one of claims 1 to 6, characterized in that: The determining, according to the evaluation parameters, a similarity result between the semantic-level expected image and the captured image comprises: Determine whether the number of feature points is greater than a preset number of feature points, and obtain a first determination result; Determine whether the feature similarity is less than a preset first similarity threshold, and obtain a second determination result; Determine whether the pixel error is less than a preset error threshold, and obtain a third determination result; Determine whether the cosine similarity is greater than a preset second similarity threshold, and obtain a fourth determination result; Determine whether the Euclidean distance is less than a preset distance threshold, and obtain a fifth determination result; When the first judgment result, the second judgment result, the third judgment result, the fourth judgment result and the fifth judgment result are all yes, it is determined that the similarity result is similar.

8. A device for evaluating the similarity of images of extraterrestrial star catalog detection, characterized in that: include: A construction module, for constructing a semantic-level expected image according to semantic information for describing an extraterrestrial object catalog; An evaluation module is used to extract and analyze features of the semantic-level expected image and the captured image of the extraterrestrial body catalog, respectively, to obtain evaluation parameters; wherein the evaluation parameters include the number of feature points, feature similarity, pixel error, cosine similarity and Euclidean distance; and determine the similarity result between the semantic-level expected image and the captured image based on the evaluation parameters.

9. A computing device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute the method according to any one of claims 1 to 7.

Images