Method for repairing surface radiation image

Abstract

The invention provides an earth surface radiation image restoration method which is applied to the technical field of remote sensing and the technical field of image processing. The method comprises: according to a restoration identifier, determining a target similar pixel matched with a to-be-restored pixel from a plurality of pixels of a to-be-restored earth surface radiation image, the target similar pixel being obtained by updating an offset of the similar pixel relative to the to-be-restored pixel based on a gradient descent algorithm; according to the dynamic adjustment parameters, the spatial distance between the to-be-repaired pixel and the target similar pixel and the spectrum difference degree, a first pixel prediction value is obtained through prediction; determining a time information weight of the auxiliary earth surface radiation image according to a spectrum difference degree between an auxiliary to-be-repaired pixel and an auxiliary target similar pixel in the auxiliary earth surface radiation image; according to the time information weight, a second pixel prediction value of the to-be-restored earth surface radiation image at the time interval is obtained through prediction; and repairing the to-be-repaired earth surface radiation image according to the first pixel prediction value and the second pixel prediction value.

Description

Technical Field

[0001] The present invention relates to the fields of remote sensing technology and image processing technology, and in particular to a method for repairing a surface radiation image. Background Art

[0002] As the main data source of surface observation, surface radiation images (also called surface background radiation products) have been widely used in many fields such as urban construction, disaster assessment, vegetation detection and environmental change monitoring. However, due to the influence of natural conditions or sensor systems, some pixels in surface radiation images are missing or distorted, making it difficult to obtain the true surface radiation characteristics of some pixels. In order to obtain the complete surface radiation characteristics of the image, the surface radiation image needs to be repaired.

[0003] In the process of realizing the concept of the present invention, the inventors found that there are at least the following problems in the related art: the information used to repair the surface radiation image is relatively insufficient, resulting in low image repair accuracy. Summary of the invention

[0004] In view of this, the present invention provides a method for repairing a surface radiation image.

[0005] One aspect of the present invention provides a method for repairing a surface radiation image, comprising: determining, according to a repair identifier, a target similar pixel matching the pixel to be repaired from multiple pixels of the surface radiation image to be repaired, the target pixel being obtained by updating the offset of the similar pixel relative to the pixel to be repaired based on a gradient descent algorithm, and the surface radiation image to be repaired characterizes the surface radiation characteristics of the spectral band; predicting a first pixel prediction value of the pixel to be repaired relative to the target similar pixel in terms of spatial distance according to a dynamic adjustment parameter, the spatial distance and spectral difference between the pixel to be repaired and the target similar pixel, wherein the dynamic adjustment parameter is used to balance the spatial distance and spectral difference between multiple target similar pixels. The weight of the difference degree is determined; according to the spectral difference degree between the auxiliary pixel to be repaired and the auxiliary target similar pixel in the auxiliary surface radiation image, the time information weight of the auxiliary surface radiation image is determined, the auxiliary pixel to be repaired is determined according to the position of the pixel to be repaired in the surface radiation image to be repaired, and the auxiliary target similar pixel is determined according to the position of the target similar pixel in the surface radiation image to be repaired; according to the time information weight, the second pixel prediction value of the surface radiation image to be repaired in the time interval is predicted, and the time interval is the acquisition time difference between the surface radiation image to be repaired and the auxiliary surface radiation image; according to the first pixel prediction value and the second pixel prediction value, the surface radiation image to be repaired is repaired.

[0006] One aspect of the present invention provides a device for repairing a surface radiation image, including: a first determination module is used to determine, according to a repair identifier, a target similar pixel that matches the pixel to be repaired from multiple pixels of the surface radiation image to be repaired, wherein the target similar pixel is obtained by updating the offset of the similar pixel relative to the pixel to be repaired based on a gradient descent algorithm, and the surface radiation image to be repaired characterizes the surface radiation characteristics of the spectral band; a first prediction module is used to predict a first pixel prediction value of the pixel to be repaired relative to the target similar pixel in the spatial distance according to a dynamic adjustment parameter, the spatial distance between the pixel to be repaired and the target similar pixel, and the spectral difference, wherein the dynamic adjustment parameter is used to balance the multiple target similar pixels. The weight of the spatial distance and spectral difference between similar pixels; the second determination module is used to determine the time information weight of the auxiliary surface radiation image according to the spectral difference between the auxiliary pixel to be repaired and the auxiliary target similar pixel in the auxiliary surface radiation image, the auxiliary pixel to be repaired is determined according to the position of the pixel to be repaired in the surface radiation image to be repaired, and the auxiliary target similar pixel is determined according to the position of the target similar pixel in the surface radiation image to be repaired; the second prediction module is used to predict the second pixel prediction value of the surface radiation image to be repaired in the time interval according to the time information weight; the repair module is used to repair the surface radiation image to be repaired according to the first pixel prediction value and the second pixel prediction value.

[0007] Another aspect of the present invention provides an electronic device, comprising:

[0008] one or more processors;

[0009] a memory for storing one or more programs,

[0010] When one or more programs are executed by one or more processors, the one or more processors implement the above method.

[0011] Another aspect of the present invention provides a computer-readable storage medium storing computer-executable instructions, which are used to implement the above method when executed.

[0012] Another aspect of the present invention provides a computer program product, which includes computer executable instructions, and the instructions are used to implement the above method when executed.

[0013] According to an embodiment of the present invention, based on the restoration identifier, a pixel to be restored is determined from multiple pixels of the surface radiation image to be restored, and a target similar pixel matching the pixel to be restored is determined from the multiple pixels, the target similar pixel being obtained by updating the offset of the similar pixel relative to the pixel to be restored based on the gradient descent algorithm; based on the dynamic adjustment parameters, the spatial distance and spectral difference between the pixel to be restored and the target similar pixel, a first pixel prediction value of the pixel to be restored relative to the target similar pixel in spatial distance is predicted; based on the spectral difference between the auxiliary pixel to be restored and the auxiliary target similar pixel in the auxiliary surface radiation image, a time information weight of the auxiliary surface radiation image is determined; based on the time information weight, a second pixel prediction value of the surface radiation image to be restored in a time interval is predicted; based on the first pixel prediction value and the second pixel prediction value, the surface radiation image to be restored is restored, that is, the spatial information of the pixel to be restored relative to the target similar pixel and the time information of the surface radiation image to be restored relative to the auxiliary surface radiation image are combined to restore the surface radiation image to be restored, so that the image restoration accuracy is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The above and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:

[0015] Figure 1 A diagram showing an application scenario of a method for repairing a ground surface radiation image according to an embodiment of the present invention;

[0016] Figure 2 A flow chart of a method for repairing a ground surface radiation image according to an embodiment of the present invention is shown;

[0017] Figure 3 A schematic diagram showing target similar pixels in a surface radiation image to be restored according to an embodiment of the present invention is shown;

[0018] Figure 4 A schematic diagram showing a method for repairing a surface radiation image according to another embodiment of the present invention is shown;

[0019] Figure 5 A structural block diagram of a device for repairing a ground surface radiation image according to an embodiment of the present invention is shown;

[0020] Figure 6 A block diagram of an electronic device suitable for implementing a method for repairing a ground surface radiation image according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0021] Below, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present invention. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of embodiments of the present invention. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of concepts of the present invention.

[0022] The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. The terms "include", "comprises", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0023] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0024] When using expressions such as "at least one of A, B, and C, etc.", they should generally be interpreted according to the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0025] In the technical solution of the present invention, the user information (including but not limited to user personal information, user image information, user device information, such as location information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, invention and application of the relevant data comply with relevant laws, regulations and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0026] In the scenario of using personal information for automated decision-making, the methods, devices, and systems provided by the embodiments of the present invention provide users with corresponding operation portals for users to choose to agree or reject the automated decision-making results; if the user chooses to reject, the expert decision-making process will be entered. The expression "automated decision-making" here refers to the activity of automatically analyzing and evaluating an individual's behavioral habits, interests and hobbies, or economic, health, credit status, etc. through computer programs, and making decisions. The expression "expert decision-making" here refers to the activity of making decisions by people who specialize in a certain field, have specialized experience, knowledge and skills, and have reached a certain level of professionalism.

[0027] Surface radiation images are the main data source for surface observation. However, due to natural conditions or sensor systems, surface radiation images are often unable to obtain the true surface radiation characteristics of some pixels. In order to ensure that the surface radiation characteristics of the image are complete, the surface radiation image needs to be repaired. Due to the lack of information used to repair the surface radiation image, the image repair accuracy is low.

[0028] In view of this, the present invention provides a method for repairing a surface radiation image, comprising: according to a repair identifier, a target similar pixel that matches the pixel to be repaired from multiple pixels of the surface radiation image to be repaired, the target similar pixel being obtained by updating the offset of the similar pixel relative to the pixel to be repaired based on a gradient descent algorithm, and the surface radiation image to be repaired characterizes the surface radiation characteristics of the spectral band; according to a dynamic adjustment parameter, the spatial distance and spectral difference between the pixel to be repaired and the target similar pixel, a first pixel prediction value of the pixel to be repaired relative to the target similar pixel in the spatial distance is predicted, and the dynamic adjustment parameter is used to balance the target similar pixel. The weight of the spatial distance and spectral difference between target similar pixels is determined; the time information weight of the auxiliary surface radiation image is determined according to the spectral difference between the auxiliary pixel to be repaired and the auxiliary target similar pixel in the auxiliary surface radiation image, the auxiliary pixel to be repaired is determined according to the position of the pixel to be repaired in the surface radiation image to be repaired, and the auxiliary target similar pixel is determined according to the position of the target similar pixel in the surface radiation image to be repaired; according to the time information weight, the second pixel prediction value of the surface radiation image to be repaired in the time interval is predicted; according to the first pixel prediction value and the second pixel prediction value, the surface radiation image to be repaired is repaired.

[0029] Figure 1 A diagram showing an application scenario of a method for repairing a surface radiation image according to an embodiment of the present invention.

[0030] like Figure 1As shown, the application scenario according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is used to provide a medium for a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or optical fiber cables, etc.

[0031] The user can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only for example).

[0032] The first terminal device 101, the second terminal device 102, and the third terminal device 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers, etc. For example, the first terminal device 101, the second terminal device 102, and the third terminal device 103 may be used to store surface radiation images.

[0033] The server 105 may be a server that provides various services, such as a request for a user to repair a locally stored surface radiation image by using the first terminal device 101, the second terminal device 102, and the third terminal device 103 to the server 105. The background management server may repair the surface radiation image in response to the received user request, and feed back the repaired surface radiation image to the terminal device.

[0034] It should be noted that the surface radiation image repair method provided in the embodiment of the present invention can generally be executed by the server 105. Accordingly, the surface radiation image repair device provided in the embodiment of the present invention can generally be set in the server 105. The surface radiation image repair method provided in the embodiment of the present invention can also be executed by a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Correspondingly, the surface radiation image repair device provided in the embodiment of the present invention can also be set in a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105.

[0035] It should be understood that Figure 1The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to the implementation requirements.

[0036] Figure 2 A flow chart of a method for repairing a ground surface radiation image according to an embodiment of the present invention is shown.

[0037] like Figure 2 As shown, the surface radiation image repair method of this embodiment includes operations S210 to S250.

[0038] In operation S210, according to the restoration identifier, a target similar pixel matching the pixel to be restored is determined from a plurality of pixels of the surface radiation image to be restored.

[0039] According to an embodiment of the present invention, the repair mark may be a unique mark of an invalid pixel value, for example, a symbol such as "-1", "X", or "589".

[0040] According to an embodiment of the present invention, the surface radiation image to be repaired represents the surface radiation characteristics of the spectral band. For example, the surface radiation image to be repaired can be a remote sensing image of the surface taken by a sensor of an aerospace platform. The surface radiation image to be repaired can also be a reflectivity map of the surface taken by a sensor of an aerospace platform.

[0041] According to an embodiment of the present invention, the target similar pixel is obtained by updating the offset of the similar pixel relative to the pixel to be repaired based on a gradient descent algorithm. The similar pixel has similarity with the pixel to be repaired.

[0042] According to an embodiment of the present invention, the target similar pixel represents the same type of ground object as that displayed by the pixel to be restored. For example, if the pixel to be restored displays plant leaves, the target similar pixel also displays plant leaves.

[0043] In operation S220, a first pixel prediction value of the pixel to be restored relative to the target similar pixel in spatial distance is predicted based on the dynamic adjustment parameter, the spatial distance between the pixel to be restored and the target similar pixel, and the spectral difference.

[0044] The dynamic adjustment parameter is used to balance the weight of the spatial distance and spectral difference between the target similar pixels, which can improve the accuracy of the first pixel prediction value.

[0045] According to an embodiment of the present invention, the spatial distance between the pixel to be restored and the target similar pixel is mainly determined according to the coordinates and geometric positions of the pixel to be restored and the target similar pixel.

[0046] For example, the surface radiation image to be restored may also be a reflectance map of the surface taken by a sensor of an aerospace platform, and the spectral difference may be obtained by calculating the Euclidean distance between the reflectances of the pixel to be restored and the target similar pixel.

[0047] For example, the surface radiation image to be restored can be a remote sensing image of the surface taken by a sensor of an aerospace platform. According to the mapping relationship between the reflectance value and the pixel value, the reflectance of the pixel to be restored and the target similar pixel can be obtained. Then, the Euclidean distance calculation is performed on the reflectance of the pixel to be restored and the target similar pixel to obtain the spectral difference.

[0048] For example, the surface radiation image to be restored can be a remote sensing image of the surface taken by a sensor of an aerospace platform. The spectral vectors of the pixel to be restored and the target similar pixel are obtained from the surface radiation image to be restored; the angle between the spectral vector of the pixel to be restored and the spectral vector of the target similar pixel is calculated to obtain the spectral difference.

[0049] In operation S230, the time information weight of the auxiliary surface radiation image is determined according to the spectral difference between the auxiliary pixel to be restored and the auxiliary target similar pixel in the auxiliary surface radiation image.

[0050] According to an embodiment of the present invention, the auxiliary surface radiation image is obtained based on the image identifier of the surface radiation image to be restored. The image identifier may be a geographical location. For example, the image identifier may be (longitude, latitude).

[0051] For example, the auxiliary surface radiation image and the surface radiation image to be restored have the same shooting content, but are shot at different times.

[0052] According to an embodiment of the present invention, the auxiliary pixel to be restored is determined from a plurality of auxiliary pixels in the auxiliary surface radiation image according to the position of the pixel to be restored in the surface radiation image to be restored.

[0053] For example, the position of the auxiliary pixel to be restored in the auxiliary surface radiation image is the same as the position of the pixel to be restored in the surface radiation image to be restored.

[0054] According to an embodiment of the present invention, the auxiliary target similar pixel is determined from a plurality of auxiliary pixels according to the position of the target similar pixel in the surface radiation image to be restored.

[0055] For example, the position of the auxiliary target similar pixel in the auxiliary surface radiation image is the same as the position of the target similar pixel in the surface radiation image to be restored.

[0056] According to an embodiment of the present invention, the time information weight represents the degree of change of the surface spectral characteristics in the time interval between taking the auxiliary surface radiation image and taking the surface radiation image to be restored. For example, the greater the time information weight, the smaller the change of the surface spectral characteristics.

[0057] In operation S240, a second pixel prediction value of the surface radiation image to be restored at a time interval is predicted based on the time information weight.

[0058] According to an embodiment of the present invention, the time interval is the acquisition time difference between the surface radiation image to be restored and the auxiliary surface radiation image.

[0059] In operation S250, the surface radiation image to be repaired is repaired according to the first pixel prediction value and the second pixel prediction value.

[0060] According to an embodiment of the present invention, the influence of the first pixel prediction value and the second pixel prediction value on the change of the spectral feature can be combined to calculate the target pixel prediction value of the pixel to be restored. The pixel to be restored in the surface radiation image to be restored is filled with a color corresponding to the target pixel prediction value.

[0061] According to an embodiment of the present invention, based on the restoration identifier, a pixel to be restored is determined from multiple pixels of the surface radiation image to be restored, and a target similar pixel matching the pixel to be restored is determined from the multiple pixels; based on the dynamic adjustment parameters, the spatial distance and spectral difference between the pixel to be restored and the target similar pixel, a first pixel prediction value of the pixel to be restored relative to the target similar pixel in spatial distance is predicted; based on the spectral difference between the auxiliary pixel to be restored and the auxiliary target similar pixel in the auxiliary surface radiation image, a time information weight of the auxiliary surface radiation image is determined; based on the time information weight, a second pixel prediction value of the surface radiation image to be restored in a time interval is predicted; based on the first pixel prediction value and the second pixel prediction value, the surface radiation image to be restored is restored, that is, the spatial information of the pixel to be restored relative to the target similar pixel and the time information of the surface radiation image to be restored relative to the auxiliary surface radiation image are combined to restore the surface radiation image to be restored, so that the image restoration accuracy is improved.

[0062] According to an embodiment of the present invention, the above method further includes: performing cluster analysis on multiple pixels in the surface radiation image to be restored to obtain target similar pixels.

[0063] For example, according to the restoration identifier, the surface radiation image to be restored is searched for the pixels to be restored, and a cluster analysis is performed on multiple pixels in the surface radiation image to be restored to obtain target similar pixels to the pixels to be restored.

[0064] Figure 3A schematic diagram of target similar pixels in a surface radiation image to be restored according to an embodiment of the present invention is shown.

[0065] like Figure 3 As shown, the surface radiation image to be repaired has an area to be repaired and two centers of the area to be repaired with different image identifiers. Image identifier " " indicates the center of the area to be repaired 1, image identification " ” indicates the center of the area to be restored 2. Similar pixels are distributed in multiple locations of the surface radiation image to be restored.

[0066] For example, the spatial distance between the pixel to be repaired and the target similar pixel is calculated as shown in formula (1), and the spectral difference between the pixel to be repaired and the target similar pixel is calculated as shown in formula (3).

[0067] For the surface radiation image to be restored, which includes multiple target similar pixels, since the spatial distance between the pixel to be restored and the target similar pixel may be very different, while the spectral distance is small, it may make the spatial distance range difficult to compare with the spectral distance. Therefore, the spatial distance and spectral distance are normalized according to formula (3) and formula (4).

[0068] (1);

[0069] i represents the i-th target similar pixel, i is an integer greater than 1, the pixel to be repaired is located at (x, y) of the image to be repaired, and the target similar pixel is located at (x i ,y i ), Di is the spatial distance between the pixel to be repaired and the target similar pixel. p is the parameter of distance measurement. When p=1, it is equivalent to Manhattan distance, which emphasizes the sum of absolute differences in each dimension. When p=2, it is equivalent to Euclidean distance, which tends to consider the sum of squares of differences in each dimension. You can choose a suitable method for calculation according to your needs.

[0070] (2);

[0071] D min and D max are the minimum and maximum spatial distances among all target similar pixels, respectively. ε is the correction number. is the normalized spatial distance.

[0072] (3);

[0073] t n The pixel value of the auxiliary pixel to be repaired at (x, y) in the middle b band of the auxiliary surface radiation image at the time, t n The middle b band at the historical moment is located at the auxiliary surface radiation image (x i ,y i ), K is the number of bands, is the spectral difference between the pixel to be repaired and the target similar pixel.

[0074] (4);

[0075] is the normalized spectral difference, is the minimum value of the spectral difference of all target similar pixels, is the maximum value of the spectral difference of all target similar pixels. ε is the correction factor.

[0076] The above method also includes: calculating a dynamic adjustment parameter according to the normalized spatial distance, the normalized spectral difference and the adjustment coefficient.

[0077] According to an embodiment of the present invention, the formula of spatial information weight is as follows:

[0078] (5);

[0079] (6);

[0080] i represents the i-th target similar pixel, and is the normalized spatial distance and the normalized spectral difference, W i is the spatial information weight of the i-th target similar pixel. i It is a dynamic adjustment parameter used to balance the weight of the spatial distance and spectral difference between multiple similar pixels, so as to increase the weight of spectral similarity when the spectral similarity is high; α 0 Is the basic weight value, used to initialize α i , for example α 0 is 1; k is the adjustment coefficient, used to control α i The range of change, k can be 0.5.

[0081] For example, if the number of target similar pixels is 30, the spatial information weights of the 30 pixels can be compared to obtain the maximum spatial information weight, and the maximum spatial information weight is used to predict the first pixel prediction value and the second pixel prediction value.

[0082] The offset of each similar pixel relative to the position of the pixel to be repaired is counted to obtain the initial offset set, as shown in formula (7):

[0083] (7);

[0084] in, It represents the offset of the i-th similar pixel corresponding to the pixel to be repaired, that is, . Similar pixels correspond one-to-one to target similar pixels.

[0085] Considering the different influences of different similar pixels on the pixels to be restored, weighted evaluation is performed from two dimensions: spatial spectral characteristics and temporal spectral characteristics. Assuming that the land cover type does not change in a short period of time, the auxiliary surface radiation image is used to construct the MRF (Markov random field) energy function according to formula (8) for calculation.

[0086] (8);

[0087] in, represents the missing area, (x, y) represents the position of the pixel to be repaired in the image to be repaired; represents the auxiliary pixel to be repaired at the position (x, y) in the auxiliary surface radiation image; N represents the number of pixels in the spatially connected area of ​​the auxiliary surface radiation image; Represents the adjacent pixels in the spatially connected region of the pixel to be repaired in the image to be repaired; Indicates the offset between the position of the pixel to be repaired and the position of the similar pixel; Represents the offset between the position of the auxiliary pixel to be repaired at the position (x, y) in the auxiliary surface radiation image and the position of the auxiliary similar pixel; Represents the data item, ensuring that the pixels used are in the valid area of ​​the image. If the similar pixel is in the valid area of ​​the image to be repaired, it is 0, otherwise +1; represents the time smoothing term, and the specific calculation formula is as shown in formula (8); represents the spatial smoothing term, and the specific calculation formula is as shown in formula (9); α and β are used to balance the temporal smoothing term respectively. and spatial smoothing , set them to 1 and 0.5.

[0088] (9);

[0089] (10);

[0090] in, Represents the offset of the i-th similar pixel relative to the pixel to be repaired; The auxiliary surface radiation image The offset of the auxiliary similar pixel relative to the auxiliary pixel to be repaired at the position (x, y) in the auxiliary surface radiation image; Represents the adjacent pixels in the spatially connected area of ​​the ith similar pixel in the auxiliary surface radiation image; represents the pixel value of the pixel to be repaired at (x, y); m represents the total number of similar pixels, which can be the same as the number of target similar pixels. and They all represent the pixel values ​​of adjacent pixels in the spatially connected area of ​​the i-th similar pixel in the auxiliary surface radiation image.

[0091] Use the gradient descent algorithm to update the offset of each similar pixel relative to the pixel to be repaired :

[0092] (11);

[0093] in, is the total energy function in The gradient at is the learning rate, which can be set to 0.001.

[0094] According to an embodiment of the present invention, based on the dynamic adjustment parameters, the spatial distance and spectral difference between the pixel to be repaired and the target similar pixel, the first pixel prediction value of the pixel to be repaired relative to the target similar pixel in the spatial distance is predicted, including: determining the spatial information weight of the target similar pixel based on the dynamic adjustment parameters, the spatial distance and spectral difference between the pixel to be repaired and the target similar pixel; predicting the first pixel prediction value of the pixel to be repaired relative to the target similar pixel in the spatial distance based on the pixel value of the auxiliary target similar pixel and the spatial information weight of the target similar pixel.

[0095] According to an embodiment of the present invention, the weighted average pixel value of all auxiliary target similar pixels can be calculated based on the pixel values ​​of the auxiliary target similar pixels and the spatial information weights of the target similar pixels; and then the first pixel prediction value is predicted based on the weighted average pixel value.

[0096] According to an embodiment of the present invention, the surface radiation image to be repaired includes multiple target similar pixels; based on the pixel values ​​of the auxiliary target similar pixels and the spatial information weights of the target similar pixels, predicting the first pixel prediction value of the pixel to be repaired relative to the target similar pixels in spatial distance includes: obtaining the first pixel prediction value to be counted of the auxiliary surface radiation image based on the pixel values ​​of each of the multiple auxiliary target similar pixels and the spatial information weights of each of the multiple target similar pixels; for the auxiliary surface radiation images at multiple moments, predicting the first pixel prediction value based on the corresponding first pixel prediction values ​​to be counted of each of the auxiliary surface radiation images at multiple moments.

[0097] According to an embodiment of the present invention, the surface radiation image to be restored includes a plurality of target-similar pixels, that is, the auxiliary restoration surface radiation image also includes a plurality of auxiliary target-similar pixels.

[0098] According to an embodiment of the present invention, the average value of the first pixel prediction values ​​to be counted of the auxiliary surface radiation image at multiple moments can be calculated, that is, the average value of the first pixel prediction values ​​to be counted of the auxiliary surface radiation image at multiple moments is the first pixel prediction value to be counted, and then the first pixel prediction value is predicted based on the average value of the first pixel prediction values ​​to be counted.

[0099] Since the coordinate offset is updated within the spatial connectivity range of the preliminarily determined similar pixels, the spatial distance and spectral difference do not change much. Therefore, the spectral spatial information weight corresponding to each similar pixel preliminarily obtained is continued to be calculated to obtain the predicted value based on the spectral spatial information of each auxiliary phase.

[0100] For example, considering the different influences of similar pixels of different targets on the pixels to be restored, weighted evaluation is performed from two dimensions: spatial spectral characteristics and temporal spectral characteristics. When the land cover type does not change in a short period of time, the four phase images of adjacent phases and the same period images of the previous and next two years are respectively used to calculate the first pixel prediction value to be counted according to formula (12). Calculation. Based on the average of the first pixel prediction values ​​to be counted of the auxiliary surface radiation images at four different times (such as formula 13), a more robust and comprehensive first pixel prediction value of the spectral spatial distance is obtained. .

[0101] (12);

[0102] (13);

[0103] T is the number of auxiliary surface radiation images. For example, if T is 4, there are four auxiliary surface radiation images at different times.

[0104] When calculating pixel prediction values ​​based on spectral time intervals, the time information weight W is added tn To improve the temporal and spatial consistency of the algorithm restoration results, the weight of each auxiliary surface radiation image for calculating the second pixel prediction value at the time interval is determined according to the minimum spectral difference calculated by each auxiliary surface radiation image. Then, the spatial information weight W is comprehensively used i and time information weight W tn , and obtain more accurate pixel prediction values.

[0105] According to an embodiment of the present invention, determining the time information weight of the auxiliary surface radiation image based on the spatial distance between the auxiliary pixel to be repaired and the auxiliary target similar pixel in the auxiliary surface radiation image includes: for the auxiliary surface radiation image including multiple auxiliary target similar pixels, multiple spectral differences corresponding to the multiple auxiliary target similar pixels are compared to obtain the target spectral difference; determining the time information weight of the auxiliary surface radiation image based on the target spectral difference.

[0106] According to an embodiment of the present invention, the target spectral difference may be the minimum spectral difference among a plurality of spectral differences corresponding to a plurality of auxiliary target similar pixels.

[0107] According to an embodiment of the present invention, determining the time information weight of the auxiliary surface radiation image according to the target spectral difference includes:

[0108] Determine the comprehensive spectral difference reference value of the auxiliary surface radiation images at multiple moments according to the target spectral difference corresponding to each of the auxiliary surface radiation images at multiple moments;

[0109] Determine the spectral difference reference value according to the target spectral difference of the auxiliary surface radiation image;

[0110] The time information weight is determined according to the spectral difference reference value and the comprehensive spectral difference reference value.

[0111] According to an embodiment of the present invention, the formula of the time information weight is as follows:

[0112] (14);

[0113] n represents the nth auxiliary surface radiation image, T is the number of auxiliary surface radiation images, For n The target spectral difference corresponding to the auxiliary surface radiation image at the time is the minimum value of the spectral difference of all auxiliary target similar pixels; W tn t n The time information weight of the auxiliary surface radiation image at time. is the comprehensive spectral difference reference value, is the spectral difference reference value.

[0114] According to an embodiment of the present invention, predicting a second pixel prediction value of a surface radiation image to be repaired at a time interval based on the time information weight includes: for a surface radiation image to be repaired including multiple target similar pixels, calculating the time restoration pixel value of the surface radiation image to be repaired based on the respective time information weights of the auxiliary surface radiation images at multiple moments and the respective spatial information weights of the multiple target similar pixels; predicting the pixel value of the pixel to be repaired at a time interval based on the time restoration pixel value to obtain a second pixel prediction value.

[0115] According to an embodiment of the present invention, the temporal restoration pixel value The calculation formula is as follows:

[0116] = (15);

[0117] The method predicts the pixel value of the pixel to be repaired at a time interval based on the time repair pixel value to obtain a second pixel prediction value, including: determining the pixel value of the auxiliary pixel to be repaired; and determining the second pixel prediction value according to the pixel value of the auxiliary pixel to be repaired and the time repair pixel value.

[0118] According to an embodiment of the present invention, the calculation formula of the second pixel prediction value is as follows:

[0119] (16);

[0120] N is the number of target similar pixels, T is the number of auxiliary surface radiation images, is the b band at time t in the image to be repaired (x+sx i ,y+sy i ) is the pixel value of the pixel to be repaired at, t n The middle b band at the time is located at (x+sx i ,y+sy i ), t n The pixel value of the auxiliary pixel to be repaired at (x, y) in the middle b band of the auxiliary surface radiation image at the time, is the predicted value of the second pixel.

[0121] According to an embodiment of the present invention, repairing the surface radiation image to be repaired based on the first pixel prediction value and the second pixel prediction value includes: repairing the surface radiation image to be repaired based on the statistical spatial distance between multiple target similar pixels and the pixel to be repaired, the spatial distance between the pixel to be repaired and the center of the area to be repaired, the first pixel prediction value and the second pixel prediction value, and the statistical spatial distance is determined based on the spatial distances corresponding to the multiple target similar pixels.

[0122] According to an embodiment of the present invention, the statistical spatial distance may be an average spatial distance of spatial distances respectively corresponding to a plurality of target similar pixels.

[0123] According to an embodiment of the present invention, a target pixel prediction value is obtained based on the statistical spatial distance between multiple target similar pixels and the pixel to be repaired, the spatial distance between the pixel to be repaired and the center of the area to be repaired, the first pixel prediction value and the second pixel prediction value. The target pixel prediction value is used to repair the pixel to be repaired in the surface radiation image to be repaired.

[0124] The calculation formula of the target pixel prediction value is as follows:

[0125] (17);

[0126] r 1 is the average spatial distance between the pixel to be repaired and the target similar pixel, r 2 is the spatial distance between the pixel to be repaired and the center of the area to be repaired. When the pixel to be repaired is close to the boundary of the area to be repaired, due to the first pixel prediction value The prediction value based on the spectral spatial distance is more reliable, so the weight is larger, which can better ensure the spatial continuity; if the pixel to be repaired is near the center of the area to be repaired, the second pixel prediction value based on the spectral time interval is More reliable. Dynamically adjust the prediction value weight according to the distance of the pixel to be repaired relative to the center or boundary of the area to be repaired.

[0127] For example, by taking advantage of the fact that the spectrum of the ground object does not change in a short period of time, the adjacent time phases of the surface radiation image to be repaired and the images of the same period of the previous and next two years are combined as auxiliary surface radiation images (auxiliary surface radiation images of four moments in total). Usually, the abnormal pixels in the impact are uniformly assigned values ​​according to the user manual of the data product, and the assigned values ​​can be used as the repair mark. According to the repair mark, by searching for target similar pixels in the surface radiation image to be repaired and performing weight analysis on the target similar pixels, the spatial information weight of each target similar pixel is obtained. According to the geographical location of the target area, the image mark of the surface radiation image to be repaired can be formed. According to the image mark, the auxiliary surface radiation images at different times corresponding to the target area can be obtained. According to the pixel values ​​of the auxiliary target similar pixels in the auxiliary surface radiation image and the spatial information weights of the target similar pixels, the weighted average pixel value of all auxiliary target similar pixels is calculated; then, a more stable first pixel prediction value in the spectral space distance is predicted based on the weighted average pixel value.

[0128] On the basis of spatial information weight, time information weight is further introduced. For the surface radiation image to be repaired, which includes multiple target similar pixels, the time repair pixel value of the surface radiation image to be repaired is calculated according to the time information weights of the auxiliary surface radiation images at multiple times and the spatial information weights of the multiple target similar pixels; the pixel value of the pixel to be repaired is predicted in the time interval based on the time repair pixel value to obtain the second pixel prediction value. Finally, the pixel to be repaired is repaired by combining the first pixel prediction value in the spectral space distance and the second pixel prediction value in the spectral time interval. The image data obtained by the same sensor is fully utilized to realize the repair of the surface background radiation data with high precision, effectively reducing the influence of noise and outliers, making the subsequent data applications more accurate and reliable, and having important practical application value.

[0129] Figure 4 A schematic diagram of a method for repairing a ground surface radiation image according to another embodiment of the present invention is shown.

[0130] like Figure 4As shown, cluster analysis is performed on multiple pixels 401 in the surface radiation image to be repaired to obtain similar pixels of the pixel to be repaired. The target similar pixel 402 is obtained by updating the offset of the similar pixel relative to the pixel to be repaired based on the gradient descent algorithm. According to the spatial distance and spectral difference of the target similar pixel 402, the spatial information weight 403 of the pixel to be repaired relative to the target similar pixel in the spatial distance is calculated by formula (5). According to the spatial information weight 403 and the minimum spectral difference in the auxiliary surface radiation images at multiple different times, the time information weight 404 of the surface radiation image to be repaired in the time interval is calculated by formula (14). Combining the time information weight 404 and the spatial information weight 403, the second pixel prediction value 405 of the surface radiation image to be repaired in the time interval is calculated by formula (16).

[0131] According to the position of the target similar pixel 402 in the surface radiation image to be restored, the auxiliary target similar pixel 407 is determined from multiple auxiliary pixels 406 in the auxiliary surface radiation images at multiple different times. The first pixel prediction value 408 to be counted in the spatial distance of the auxiliary surface radiation images at multiple different times is calculated by formula (12). Then, the first pixel prediction value 408 to be counted of the auxiliary surface radiation images at multiple different times is averaged (as shown in formula (13)) to obtain the first pixel prediction value 409 in the spatial distance of the surface radiation image to be restored. According to the statistical spatial distance between the multiple target similar pixels and the pixel to be restored, the spatial distance between the pixel to be restored and the center of the area to be restored, the first pixel prediction value 409 and the second pixel prediction value 405, the target pixel prediction value 410 is obtained by formula (17).

[0132] By comprehensively utilizing image data from adjacent time phases and the same period of the previous and next two years, and based on the fact that the spectral characteristics of objects remain basically unchanged over a short period of time, effective restoration of surface background radiation data is achieved. This method can more accurately fill in missing values, reduce data gaps caused by cloud cover, sensor failures, etc., and thus improve the overall quality of the data.

[0133] The description of the pixels to be repaired in the data product manual is used in the labeling of the pixels to be repaired, and they are uniformly labeled to avoid the uncertainty of the positioning of the pixels to be repaired. During the repair process, the most representative similar pixels are screened, and the weights of spatial, temporal and spectral information are comprehensively considered to give each target similar pixel an appropriate weight, thereby reducing the impact of noise and providing more accurate surface background radiation data.

[0134] It does not rely on remote sensing data from other sensors. Compared with the method of using remote sensing data from other sensors as auxiliary data for restoration, it does not require preprocessing steps such as image registration and image fusion. In remote sensing image restoration, it is not limited to a certain ground object and underlying surface, and can be used for complex surfaces, with a wide range of applications.

[0135] Figure 5 A structural block diagram of a device for repairing a surface radiation image according to an embodiment of the present invention is shown.

[0136] like Figure 5 As shown, the surface radiation image repairing device of this embodiment includes a first determination module 510 , a first prediction module 520 , a second determination module 530 , a second prediction module 540 and a repairing module 550 .

[0137] The first determination module 510 is used to determine, according to the restoration identifier, a target similar pixel that matches the pixel to be restored from multiple pixels of the surface radiation image to be restored, wherein the target similar pixel is obtained by updating the offset of the similar pixel relative to the pixel to be restored based on the gradient descent algorithm. The surface radiation image to be restored represents the surface radiation characteristics of the spectral band. In one embodiment, the first determination module 510 can be used to perform the operation S210 described above, which will not be repeated here.

[0138] The first prediction module 520 is used to predict the first pixel prediction value of the pixel to be repaired relative to the target similar pixel in the spatial distance according to the dynamic adjustment parameter, the spatial distance and the spectral difference between the pixel to be repaired and the target similar pixel, and the dynamic adjustment parameter is used to balance the weights of the spatial distance and the spectral difference between the multiple target similar pixels. In one embodiment, the first prediction module 520 can be used to perform the operation S220 described above, which will not be repeated here.

[0139] The second determination module 530 is used to determine the time information weight of the auxiliary surface radiation image according to the spectral difference between the auxiliary pixel to be repaired and the auxiliary target similar pixel in the auxiliary surface radiation image, the auxiliary pixel to be repaired is determined according to the position of the pixel to be repaired in the surface radiation image to be repaired, and the auxiliary target similar pixel is determined according to the position of the target similar pixel in the surface radiation image to be repaired. In one embodiment, the second determination module 530 can be used to perform the operation S230 described above, which will not be repeated here.

[0140] The second prediction module 540 is used to predict the second pixel prediction value of the surface radiation image to be restored at the time interval according to the time information weight. In one embodiment, the second prediction module 540 can be used to perform the operation S240 described above, which will not be repeated here.

[0141] The repair module 550 is used to repair the surface radiation image to be repaired according to the first pixel prediction value and the second pixel prediction value. In one embodiment, the repair module 550 can be used to perform the operation S250 described above, which will not be described in detail here.

[0142] According to an embodiment of the present invention, the first prediction module 520 includes a first determination submodule and a first prediction submodule. The first determination submodule is used to determine the spatial information weight of the target similar pixel according to the spatial distance and spectral difference between the pixel to be repaired and the target similar pixel; the first prediction submodule is used to predict the first pixel prediction value of the pixel to be repaired relative to the target similar pixel in the spatial distance according to the pixel value of the auxiliary target similar pixel and the spatial information weight of the target similar pixel.

[0143] According to an embodiment of the present invention, the first prediction submodule includes a first acquisition unit and a first prediction unit. The first acquisition unit is used to obtain the first pixel prediction value to be counted of the surface radiation image to be restored according to the pixel values ​​of each of the multiple auxiliary target similar pixels and the spatial information weights of each of the multiple target similar pixels, for the surface radiation image to be restored, including multiple target similar pixels; the first prediction unit is used to predict the first pixel prediction value according to the corresponding first pixel prediction values ​​to be counted of each of the auxiliary surface radiation images at multiple moments.

[0144] According to an embodiment of the present invention, the second prediction module 540 includes a calculation submodule and a second prediction submodule. The calculation submodule is used to calculate the time restoration pixel value of the surface radiation image to be restored according to the time information weights of the auxiliary surface radiation images at multiple moments and the spatial information weights of the multiple target similar pixels in the surface radiation image to be restored; the second prediction submodule is used to predict the pixel value of the pixel to be restored at a time interval based on the time restoration pixel value to obtain a second pixel prediction value.

[0145] According to an embodiment of the present invention, the restoration module 550 includes a restoration submodule. The restoration submodule is used to restore the surface radiation image to be restored according to the statistical spatial distances between multiple target similar pixels and the pixel to be restored, the spatial distance between the pixel to be restored and the center of the area to be restored, the first pixel prediction value and the second pixel prediction value, wherein the statistical spatial distance is determined according to the spatial distances respectively corresponding to the multiple target similar pixels.

[0146] According to an embodiment of the present invention, the second determination module 530 includes a comparison submodule and a second determination submodule. The comparison submodule is used for comparing a plurality of spectral differences corresponding to a plurality of auxiliary target similar pixels in the auxiliary surface radiation image to obtain a target spectral difference; and the second determination submodule is used for determining the time information weight of the auxiliary surface radiation image according to the target spectral difference.

[0147] According to an embodiment of the present invention, the above device further comprises a cluster analysis module. The cluster analysis module is used to perform cluster analysis on multiple pixels in the surface radiation image to be restored to obtain target similar pixels.

[0148] According to an embodiment of the present invention, the formula of the time information weight is shown in formula (14).

[0149] According to an embodiment of the present invention, the formula of the spatial information weight is shown in formula (5).

[0150] According to an embodiment of the present invention, the calculation formula for the second pixel prediction value is shown in formula (16).

[0151] According to an embodiment of the present invention, any multiple modules of the first determination module 510, the first prediction module 520, the second determination module 530, the second prediction module 540 and the repair module 550 can be combined into one module for implementation, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present invention, at least one of the first determination module 510, the first prediction module 520, the second determination module 530, the second prediction module 540 and the repair module 550 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware and firmware or in a proper combination of any of them. Alternatively, at least one of the first determination module 510, the first prediction module 520, the second determination module 530, the second prediction module 540 and the repair module 550 may be at least partially implemented as a computer program module, which may perform a corresponding function when executed.

[0152] Figure 6 A block diagram of an electronic device suitable for implementing a method for repairing a ground surface radiation image according to an embodiment of the present invention is shown.

[0153] like Figure 6As shown, the electronic device according to an embodiment of the present invention includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 to a random access memory (RAM) 603. The processor 601 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 601 may also include an onboard memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0154] In RAM 603, various programs and data required for the operation of the electronic device are stored. The processor 601, ROM 602 and RAM 603 are connected to each other via a bus 604. The processor 601 performs various operations of the method flow according to the embodiment of the present invention by executing the programs in ROM 602 and / or RAM 603. It should be noted that the program can also be stored in one or more memories other than ROM 602 and RAM 603. The processor 601 can also perform various operations of the method flow according to the embodiment of the present invention by executing the programs stored in the one or more memories.

[0155] According to an embodiment of the present invention, the electronic device may further include an input / output (I / O) interface 605, which is also connected to the bus 604. The electronic device may further include one or more of the following components connected to the input / output (I / O) interface 605: an input portion 606 including a keyboard, a mouse, etc.; an output portion 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 608 including a hard disk, etc.; and a communication portion 609 including a network interface card such as a LAN card, a modem, etc. The communication portion 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the input / output (I / O) interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage portion 608 as needed.

[0156] The present invention also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiment; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present invention is implemented.

[0157] According to an embodiment of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, it may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus or a device. For example, according to an embodiment of the present invention, the computer-readable storage medium may include the ROM 602 and / or RAM 603 described above and / or one or more memories other than ROM 602 and RAM 603.

[0158] The embodiment of the present invention also includes a computer program product, which includes a computer program, and the computer program contains program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to enable the computer system to implement the method for repairing the surface radiation image provided by the embodiment of the present invention.

[0159] The computer program executes the above functions defined in the system / device of the embodiment of the present invention when it is executed by the processor 601. According to the embodiment of the present invention, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0160] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 609, and / or installed from a removable medium 611. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0161] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the processor 601, the above functions defined in the system of the embodiment of the present invention are performed. According to the embodiment of the present invention, the system, device, means, module, unit, etc. described above can be implemented by a computer program module.

[0162] According to an embodiment of the present invention, the program code for executing the computer program provided by the embodiment of the present invention can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (e.g., using an Internet service provider to connect through the Internet).

[0163] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0164] It will be appreciated by those skilled in the art that the features described in the various embodiments of the present invention may be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in the present invention. In particular, without departing from the spirit and teachings of the present invention, the features described in the various embodiments of the present invention may be combined and / or combined in various ways. All of these combinations and / or combinations fall within the scope of the present invention.

[0165] The embodiments of the present invention are described above. However, these embodiments are only for the purpose of illustration, and are not intended to limit the scope of the present invention. Although each embodiment is described above, it does not mean that the measures in each embodiment cannot be used in combination advantageously. Without departing from the scope of the present invention, those skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.

Claims

1. A method for repairing a surface radiation image, characterized in that: include: According to the restoration identifier, a target similar pixel matching the pixel to be restored is determined from multiple pixels of the surface radiation image to be restored, wherein the target similar pixel is obtained by updating the offset of the similar pixel relative to the pixel to be restored based on a gradient descent algorithm, and the surface radiation image to be restored represents the surface radiation characteristics of the spectral band; Predicting a first pixel prediction value of the pixel to be restored relative to the target similar pixel at the spatial distance according to a dynamic adjustment parameter, a spatial distance and a spectral difference between the pixel to be restored and the target similar pixel, wherein the dynamic adjustment parameter is used to balance the weights of the spatial distance and the spectral difference between a plurality of target similar pixels; Determine the time information weight of the auxiliary surface radiation image according to the spectral difference between the auxiliary pixel to be restored and the auxiliary target similar pixel in the auxiliary surface radiation image, wherein the auxiliary pixel to be restored is determined according to the position of the pixel to be restored in the surface radiation image to be restored, and the auxiliary target similar pixel is determined according to the position of the target similar pixel in the surface radiation image to be restored; Predicting a second pixel prediction value of the surface radiation image to be restored at a time interval according to the time information weight; The surface radiation image to be repaired is repaired according to the first pixel prediction value and the second pixel prediction value.

2. The method according to claim 1, characterized in that The step of predicting the first pixel prediction value of the pixel to be restored relative to the target similar pixel at the spatial distance according to the dynamic adjustment parameter, the spatial distance between the pixel to be restored and the target similar pixel, and the spectral difference comprises: Determining the spatial information weight of the target similar pixel according to the dynamic adjustment parameter, the spatial distance between the pixel to be repaired and the target similar pixel, and the spectral difference; The first pixel prediction value of the pixel to be repaired relative to the target similar pixel at the spatial distance is predicted based on the pixel value of the auxiliary target similar pixel and the spatial information weight of the target similar pixel.

3. The method according to claim 2, characterized in that The surface radiation image to be restored includes a plurality of target similar pixels; the first pixel prediction value of the pixel to be restored relative to the target similar pixel at the spatial distance is predicted based on the pixel value of the auxiliary target similar pixel and the spatial information weight of the target similar pixel, including: Obtaining a first pixel prediction value to be counted of the auxiliary surface radiation image according to the pixel values ​​of each of the plurality of auxiliary target similar pixels and the spatial information weights of each of the plurality of target similar pixels; For the auxiliary surface radiation images at multiple moments, the first pixel prediction value is predicted based on the first pixel prediction values ​​to be counted corresponding to each of the auxiliary surface radiation images at multiple moments.

4. The method according to claim 3, characterized in that The step of predicting the second pixel prediction value of the surface radiation image to be restored at the time interval according to the time information weight includes: For the surface radiation image to be restored including a plurality of target similar pixels, the time restoration pixel value of the surface radiation image to be restored is calculated according to the respective time information weights of the auxiliary surface radiation images at a plurality of moments and the respective space information weights of the plurality of target similar pixels; The pixel value of the pixel to be repaired is predicted at the time interval based on the time repair pixel value to obtain the second pixel prediction value.

5. The method according to claim 4, characterized in that The step of predicting the pixel value of the pixel to be restored at the time interval based on the time restoration pixel value to obtain the second pixel prediction value includes: Determining the pixel value of the auxiliary pixel to be repaired; The second pixel prediction value is determined according to the pixel value of the auxiliary pixel to be repaired and the temporal repair pixel value.

6. The method according to claim 3, characterized in that The repairing of the surface radiation image to be repaired according to the first pixel prediction value and the second pixel prediction value comprises: The surface radiation image to be repaired is repaired based on the statistical spatial distances between multiple target similar pixels and the pixels to be repaired, the spatial distance between the pixels to be repaired and the center of the area to be repaired, the first pixel prediction value and the second pixel prediction value, and the statistical spatial distance is determined based on the spatial distances corresponding to the multiple target similar pixels respectively.

7. The method according to claim 1, characterized in that The step of determining the time information weight of the auxiliary surface radiation image according to the spectral difference between the auxiliary pixel to be restored and the auxiliary target similar pixel in the auxiliary surface radiation image comprises: The auxiliary surface radiation image includes a plurality of auxiliary target similar pixels, comparing a plurality of spectral differences corresponding to the plurality of auxiliary target similar pixels to obtain a target spectral difference; The time information weight of the auxiliary surface radiation image is determined according to the target spectral difference.

8. The method according to claim 7, characterized in that The step of determining the time information weight of the auxiliary surface radiation image according to the target spectral difference comprises: Determining comprehensive spectral difference reference values ​​of the auxiliary surface radiation images at multiple moments according to the target spectral difference degrees corresponding to each of the auxiliary surface radiation images at multiple moments; Determining a spectral difference reference value according to the target spectral difference of the auxiliary surface radiation image; The time information weight is determined according to the spectrum difference reference value and the comprehensive spectrum difference reference value.

9. The method according to claim 1, characterized in that: The method further comprises: Cluster analysis is performed on the plurality of pixels in the surface radiation image to be restored to obtain the target similar pixels.

10. The method according to claim 2, characterized in that The method further comprises: The dynamic adjustment parameter is calculated according to the normalized spatial distance, the normalized spectral difference and the adjustment coefficient.

Images