KR102285530B1 - Method for processing image for registration - Google Patents

Abstract

In accordance with an embodiment of the present invention, an image processing method for image registration performed by a computing device is disclosed. The method comprises the steps of: extracting a first content feature and a first style feature from a first image using a pre-trained first model; extracting a second content feature and a second style feature from a second image by using the first model; determining a sensor domain of each of the first image and the second image based on the first style feature and the second style feature by using a pre-trained second model; and performing style translation of the first image based on a determination result of the sensor domain by using a pre-trained third model. In accordance with the present invention, the images which are difficult to register due to a severe style difference can be processed based on the sensor domain.

Description

Image processing method for image registration

The present invention relates to an image processing method, and more particularly, to an image processing technology for solving a performance degradation problem of image registration caused by a change in semantic information between images.

In general, registration is performed through a process of extracting features from each of a plurality of images and matching the extracted features with each other. Such a feature matching method can be smoothly performed when the difference in semantic information between images is not large. However, when the difference in semantic information between images increases due to differences in sensor domains, etc., the accuracy of feature extraction and matching of each of the images is reduced, and thus the matching performance is inevitably deteriorated. .

For example, even if a satellite image including the same object is an image generated by different sensors, a style difference may occur between the images according to characteristics of the image determined by the sensors. In this case, when the style difference between the images is extreme, the accuracy of the process of finding and matching feature points corresponding to each other cannot be guaranteed. Therefore, if the style difference between images caused by the difference in sensor domains is extreme, the overall performance of the registration process is inevitably deteriorated.

Korean Patent Registration No. 10-1982755 (2019.05.21) discloses a method and apparatus for registering an aerial image.

The present disclosure has been devised in response to the above background art, and an object of the present disclosure is to provide a preprocessing algorithm for registration, which is robust to the difference in semantic information of images.

An image processing method for image registration performed by a computing device according to an embodiment of the present disclosure for realizing the above-described problems is disclosed. The method comprises the steps of extracting a first content feature and a first style feature from a first image using a pre-trained first model; extracting a second content feature and a second style feature from a second image using the first model; determining a sensor domain of each of the first image and the second image based on the first style feature and the second style feature using a pre-trained second model; and using the pre-trained third model, performing style translation of the first image based on the determination result of the sensor domain.

In an alternative embodiment, determining the sensor domain of each of the first image and the second image comprises: using the second model, based on the first style characteristic and the second style characteristic. classifying a sensor domain of each of an image and the second image; and determining whether a sensor domain matches between the first image and the second image based on a result of the classification.

In an alternative embodiment, the performing of the style conversion of the first image may include, when the sensor domains between the first image and the second image match, the object included in each of the first image and the second image. determining the third model based on a characteristic related to ; and using the determined third model, performing style transformation of the first image based on the first content feature and the second style feature.

In an alternative embodiment, the determining of the third model based on a characteristic associated with an object included in each of the first image and the second image comprises: using a pre-trained fourth model, the first classifying a characteristic related to an object included in each of the first image and the second image based on the style characteristic and the second style characteristic; and selecting a third model corresponding to the classified characteristic from among a plurality of pre-trained candidate models.

In an alternative embodiment, classifying a characteristic associated with an object included in each of the first image and the second image comprises: using a first sub-model of the fourth model, based on the first style characteristic classifying the use state of the land included in the first image; and classifying a season represented by each of the first image and the second image based on the first style feature and the second style feature using a second sub-model of the fourth model.

In an alternative embodiment, the performing of the style conversion of the first image may include: when a sensor domain between the first image and the second image is different, based on a sensor domain of the first image, the third model determining; and using the determined third model, performing style transformation of the first image based on the first content feature and the second style feature.

In an alternative embodiment, the determining of the third model based on the sensor domain of the first image may include selecting the sensor domain of the first image from among a plurality of pre-trained candidate models as the sensor of the second image. It may include selecting a third model to be transformed into a domain.

In an alternative embodiment, the sensor domain of each of the first image and the second image may comprise an electro-optic sensor or a synthetic aperture radar sensor.

Disclosed is a computer program stored in a computer-readable storage medium according to an embodiment of the present disclosure for realizing the above-described problems. the computer program, when executed on one or more processors, to perform the following operations for processing an image for image registration, the operations comprising: using a pre-trained first model, a first content from a first image extracting a (content) feature and a first style feature; extracting a second content feature and a second style feature from a second image using the first model; determining a sensor domain of each of the first image and the second image based on the first style feature and the second style feature using a pre-trained second model; and performing style translation of the first image based on the determination result of the sensor domain by using the pre-trained third model.

Disclosed is a computing device for processing an image for image registration according to an embodiment of the present disclosure for realizing the above-described problem. The apparatus includes: a processor including at least one core; a memory containing program codes executable by the processor; and a network unit for receiving an image, wherein the processor extracts a first content feature and a first style feature from the first image using the pre-trained first model, and Using a model, extract a second content feature and a second style feature from a second image, and using a pre-trained second model, the first image based on the first style feature and the second style feature and determining a sensor domain of each of the second images, and using a pre-trained third model to perform style translation of the first image based on the determination result of the sensor domain. .

The present disclosure may provide a method of processing images that are difficult to register due to a severe style difference based on a sensor domain.

1 is a block diagram of a computing device that processes an image for image registration according to an embodiment of the present disclosure.
2 is a schematic diagram illustrating a network function according to an embodiment of the present disclosure;
3 is a block diagram illustrating an image processing process of a computing device according to an embodiment of the present disclosure.
4 and 5 are block diagrams illustrating a model selection process for image processing of a computing device according to an embodiment of the present disclosure.
6 is a block diagram illustrating a process of classifying a sensor domain of a satellite image according to an embodiment of the present disclosure.
7 and 8 are block diagrams illustrating a style translation process classified according to a sensor domain of a satellite image according to an embodiment of the present disclosure.
9 and 10 are flowcharts illustrating an image processing method for image registration according to an embodiment of the present disclosure.
11 is a schematic diagram of a computing environment according to an embodiment of the present disclosure.

Various embodiments are now described with reference to the drawings. In this specification, various descriptions are presented to provide an understanding of the present disclosure. However, it is apparent that these embodiments may be practiced without these specific descriptions.

The terms “component,” “module,” “system,” and the like, as used herein, refer to a computer-related entity, hardware, firmware, software, a combination of software and hardware, or execution of software. For example, a component can be, but is not limited to being, a process running on a processor, a processor, an object, a thread of execution, a program, and/or a computer. For example, both an application running on a computing device and the computing device may be a component. One or more components may reside within a processor and/or thread of execution. A component may be localized within one computer. A component may be distributed between two or more computers. In addition, these components can execute from various computer readable media having various data structures stored therein. Components may communicate via a network such as the Internet with another system, for example via a signal having one or more data packets (eg, data and/or signals from one component interacting with another component in a local system, distributed system, etc.) may communicate via local and/or remote processes depending on the data being transmitted).

In addition, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or.” That is, unless otherwise specified or clear from context, "X employs A or B" is intended to mean one of the natural implicit substitutions. That is, X employs A; X employs B; or when X employs both A and B, "X employs A or B" may apply to either of these cases. It should also be understood that the term “and/or” as used herein refers to and includes all possible combinations of one or more of the listed related items.

Also, the terms "comprises" and/or "comprising" should be understood to mean that the feature and/or element in question is present. However, it should be understood that the terms "comprises" and/or "comprising" do not exclude the presence or addition of one or more other features, elements and/or groups thereof. Also, unless otherwise specified or unless the context is clear as to designating a singular form, the singular in the specification and claims should generally be construed to mean "one or more."

In addition, the term "at least one of A or B" should be interpreted as meaning "when including only A", "when including only B", and "when combined with the configuration of A and B".

Those skilled in the art will further appreciate that the various illustrative logical blocks, configurations, modules, circuits, means, logics, and algorithm steps described in connection with the embodiments disclosed herein may be implemented in electronic hardware, computer software, or combinations of both. It should be recognized that they can be implemented with To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, configurations, means, logics, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application. However, such implementation decisions should not be interpreted as causing a departure from the scope of the present disclosure.

Descriptions of the presented embodiments are provided to enable those skilled in the art to use or practice the present invention. Various modifications to these embodiments will be apparent to those skilled in the art of the present disclosure. The generic principles defined herein may be applied to other embodiments without departing from the scope of the present disclosure. Thus, the present invention is not limited to the embodiments presented herein. This invention is to be construed in its widest scope consistent with the principles and novel features presented herein.

In the present disclosure, a network function, an artificial neural network, and a neural network may be used interchangeably.

On the other hand, the term "image", "image" or "image data" used throughout the detailed description and claims of the present invention refers to multidimensional data composed of discrete image elements (eg, pixels in a two-dimensional image). In other words, it is a term that refers to a visible object (eg, displayed on a video screen) or a digital representation of that object (eg, a file corresponding to the pixel output of the back).

1 is a block diagram of a computing device that processes an image for image registration according to an embodiment of the present disclosure.

The configuration of the computing device 100 shown in FIG. 1 is only a simplified example. In an embodiment of the present disclosure, the computing device 100 may include other components for performing the computing environment of the computing device 100 , and only some of the disclosed components may configure the computing device 100 .

The computing device 100 may include a processor 110 , a memory 130 , and a network unit 150 .

The processor 110 may include one or more cores, and a central processing unit (CPU) of a computing device, a general purpose graphics processing unit (GPGPU), and a tensor processing unit (TPU). unit) and the like, and may include a processor for data analysis and deep learning. The processor 110 may read a computer program stored in the memory 130 and perform data processing for machine learning according to an embodiment of the present disclosure. According to an embodiment of the present disclosure, the processor 110 may perform an operation for learning the neural network. The processor 110 for learning of the neural network, such as processing input data for learning in deep learning (DL), extracting features from the input data, calculating an error, updating the weight of the neural network using backpropagation calculations can be performed. At least one of a CPU, a GPGPU, and a TPU of the processor 110 may process learning of a network function. For example, the CPU and the GPGPU can process learning of a network function and data classification using the network function together. In addition, in an embodiment of the present disclosure, learning of a network function and data classification using the network function may be processed by using the processors of a plurality of computing devices together. In addition, the computer program executed in the computing device according to an embodiment of the present disclosure may be a CPU, GPGPU, or TPU executable program.

According to an embodiment of the present disclosure, the processor 110 may perform preprocessing for registration of a plurality of images. The processor 110 may extract features of each of the images input for registration by disentangle. For example, the processor 110 may extract a content feature and a style feature of each of the images input for registration by using a pre-trained deep learning model. In this case, the content feature may be understood as a feature representing semantic information about an object (including a spatial structure, etc.) included in an image that is invariant even when a domain changes. The style feature may be understood as a feature representing specific semantic information of an object included in an image to which domain change is reflected.

The processor 110 may compare and analyze the sensor domain of each image based on the features extracted from the input images. The processor 110 may determine the sensor domain of each image based on the characteristics of each of the input images. The processor 110 may select a model for style translation of a target image by comparing sensor domains corresponding to each image. For example, the processor 110 may estimate a difference in a sensor domain between images based on a style feature of each of the input images. The processor 110 may determine whether the sensor domains match each other based on the style features of each of the input images using the pre-trained deep learning model. The processor 110 may perform an operation for selecting a differentiated model based on whether the sensor domains between the input images match each other. When the sensor domains between the input images match, the processor 110 may select a model suitable for style transformation of the target image from among a plurality of predetermined candidate models through an additional operation on the style characteristics of each image. When the sensor domains between the input images do not match, the processor 110 may select a model suitable for style transformation of the target image from among a plurality of predetermined candidate models based on the sensor domain of the target image.

The processor 110 may perform style conversion of the target image based on the model selected through the above-described process. The processor 110 may convert the style of the target image into the style of the source image using the selection model. For example, the processor 110 may use the selection model to convert the style of the target image to the style of the source image based on the content characteristics of the target image and the style characteristics of the source image.

Through the pre-processing operation of the processor 110 as described above, the style of the image, which is the target of matching, is accurately converted into the style of the source image, thereby solving the problem of degradation of image matching caused by the style difference.

According to an embodiment of the present disclosure, the memory 130 may store any type of information generated or determined by the processor 110 and any type of information received by the network unit 150 .

According to an embodiment of the present disclosure, the memory 130 is a flash memory type, a hard disk type, a multimedia card micro type, and a card type memory (eg, SD or XD memory, etc.), Random Access Memory (RAM), Static Random Access Memory (SRAM), Read-Only Memory (ROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Programmable Read (PROM) -Only Memory), a magnetic memory, a magnetic disk, and an optical disk may include at least one type of storage medium. The computing device 100 may operate in relation to a web storage that performs a storage function of the memory 130 on the Internet. The description of the above-described memory is only an example, and the present disclosure is not limited thereto.

The network unit 150 according to an embodiment of the present disclosure may use any type of known wired/wireless communication system.

The network unit 150 may receive an image in which an object of interest is expressed from an external system. For example, the network unit 150 may receive a ground-captured image including content such as a forest or a city from an artificial satellite system, an aerial system, or the like. The ground-captured image may be data for training or inference of a neural network model. The ground-captured image in which the content is expressed may include both an electro-optic image and a synthetic aperture radar (SAR) image captured through a sensor mounted on a satellite or an aircraft. The terrestrial photographed image in which the object of interest is expressed is not limited to the above-described examples, and may be variously configured within a range that can be understood by those skilled in the art.

Meanwhile, the computing device 100 according to an embodiment of the present disclosure may include a server as a computing system that transmits and receives information through communication with a client. In this case, the client may be any type of terminal that can access the server. For example, the computing device 100, which is a server, receives a terrestrial photographed image from a satellite system, performs matching with a source image, and provides the matching result to a server or user terminal for image analysis (eg object detection, etc.). can

In an additional embodiment, the computing device 100 may include any type of terminal that receives a data resource generated by an arbitrary server and performs additional information processing.

2 is a schematic diagram illustrating a network function according to an embodiment of the present disclosure;

The model, deep learning model, or Nth model (N is a natural number) referred to in the present disclosure may include a neural network. Throughout this specification, the terms neural network, network function, and neural network may be used interchangeably. A neural network may be composed of a set of interconnected computational units, which may generally be referred to as nodes. These nodes may also be referred to as neurons. A neural network is configured by including at least one or more nodes. Nodes (or neurons) constituting the neural networks may be interconnected by one or more links.

In the neural network, one or more nodes connected through a link may relatively form a relationship between an input node and an output node. The concepts of an input node and an output node are relative, and any node in an output node relationship with respect to one node may be in an input node relationship in a relationship with another node, and vice versa. As described above, an input node-to-output node relationship may be created around a link. One or more output nodes may be connected to one input node through a link, and vice versa.

In the relationship between the input node and the output node connected through one link, the value of the data of the output node may be determined based on data input to the input node. Here, a link interconnecting the input node and the output node may have a weight. The weight may be variable, and may be changed by the user or algorithm in order for the neural network to perform a desired function. For example, when one or more input nodes are interconnected to one output node by respective links, the output node sets values input to input nodes connected to the output node and links corresponding to the respective input nodes. An output node value may be determined based on the weight.

As described above, in a neural network, one or more nodes are interconnected through one or more links to form an input node and an output node relationship within the neural network. The characteristics of the neural network may be determined according to the number of nodes and links in the neural network, the correlation between the nodes and the links, and the value of a weight assigned to each of the links. For example, when the same number of nodes and links exist and there are two neural networks having different weight values of the links, the two neural networks may be recognized as different from each other.

A neural network may consist of a set of one or more nodes. A subset of nodes constituting the neural network may constitute a layer. Some of the nodes constituting the neural network may configure one layer based on distances from the initial input node. For example, a set of nodes having a distance n from the initial input node may constitute n layers. The distance from the initial input node may be defined by the minimum number of links that must be traversed to reach the corresponding node from the initial input node. However, the definition of such a layer is arbitrary for description, and the order of the layer in the neural network may be defined in a different way from the above. For example, a layer of nodes may be defined by a distance from the final output node.

The initial input node may refer to one or more nodes to which data is directly input without going through a link in a relationship with other nodes among nodes in the neural network. Alternatively, in a relationship between nodes based on a link in a neural network, it may mean nodes that do not have other input nodes connected by a link. Similarly, the final output node may refer to one or more nodes that do not have an output node in relation to other nodes among nodes in the neural network. In addition, the hidden node may mean nodes constituting the neural network other than the first input node and the last output node.

The neural network according to an embodiment of the present disclosure may be a neural network in which the number of nodes in the input layer may be the same as the number of nodes in the output layer, and the number of nodes decreases and then increases again as progresses from the input layer to the hidden layer. can In addition, the neural network according to another embodiment of the present disclosure may be a neural network in which the number of nodes in the input layer may be less than the number of nodes in the output layer, and the number of nodes decreases as the number of nodes progresses from the input layer to the hidden layer. there is. In addition, the neural network according to another embodiment of the present disclosure may be a neural network in which the number of nodes in the input layer may be greater than the number of nodes in the output layer, and the number of nodes increases as the input layer progresses to the hidden layer. can The neural network according to another embodiment of the present disclosure may be a neural network in a combined form of the aforementioned neural networks.

A deep neural network (DNN) may refer to a neural network including a plurality of hidden layers in addition to an input layer and an output layer. Deep neural networks can be used to identify the latent structures of data. In other words, it can identify the potential structure of photos, texts, videos, voices, and music (for example, what objects are in the photos, what the text and emotions are, what the texts and emotions are, etc.) . Deep neural networks include convolutional neural networks (CNNs), recurrent neural networks (RNNs), auto encoders, generative adversarial networks (GANs), and restricted boltzmann machines (RBMs). machine), a deep belief network (DBN), a Q network, a U network, a Siamese network, a Generative Adversarial Network (GAN), and the like. The description of the deep neural network described above is only an example, and the present disclosure is not limited thereto.

In an embodiment of the present disclosure, the network function may include an autoencoder. The auto-encoder may be a kind of artificial neural network for outputting output data similar to input data. The auto encoder may include at least one hidden layer, and an odd number of hidden layers may be disposed between the input/output layers. The number of nodes in each layer may be reduced from the number of nodes of the input layer to an intermediate layer called the bottleneck layer (encoding), and then expanded symmetrically with reduction from the bottleneck layer to the output layer (symmetrically with the input layer). The auto-encoder can perform non-linear dimensionality reduction. The number of input layers and output layers may correspond to a dimension after preprocessing the input data. In the auto-encoder structure, the number of nodes of the hidden layer included in the encoder may have a structure that decreases as the distance from the input layer increases. If the number of nodes in the bottleneck layer (the layer with the fewest nodes between the encoder and the decoder) is too small, a sufficient amount of information may not be conveyed, so a certain number or more (e.g., more than half of the input layer, etc.) ) may be maintained.

The neural network may be trained by at least one of teacher learning, unsupervised learning, semi-supervised learning, and reinforcement learning. Learning of the neural network may be a process of applying knowledge for the neural network to perform a specific operation to the neural network.

A neural network can be trained in a way that minimizes output errors. In the training of a neural network, iteratively inputs the training data to the neural network, calculates the output of the neural network and the target error for the training data, and calculates the error of the neural network from the output layer of the neural network to the input layer in the direction to reduce the error. It is a process of updating the weight of each node in the neural network by backpropagation in the direction. In the case of teacher learning, learning data in which the correct answer is labeled in each learning data is used (ie, labeled learning data), and in the case of comparative learning, the correct answer may not be labeled in each learning data. That is, for example, the learning data in the case of teacher learning regarding data classification may be data in which categories are labeled in each of the learning data. Labeled training data is input to the neural network, and an error can be calculated by comparing the output (category) of the neural network with the label of the training data. As another example, in the case of comparison learning about data classification, an error may be calculated by comparing the input training data with the neural network output. The calculated error is back propagated in the reverse direction (ie, from the output layer to the input layer) in the neural network, and the connection weight of each node of each layer of the neural network may be updated according to the back propagation. A change amount of the connection weight of each node to be updated may be determined according to a learning rate. The computation of the neural network on the input data and the backpropagation of errors can constitute a learning cycle (epoch). The learning rate may be applied differently according to the number of repetitions of the learning cycle of the neural network. For example, in the early stage of training of a neural network, a high learning rate can be used to enable the neural network to quickly acquire a certain level of performance, thereby increasing efficiency, and using a low learning rate at the end of learning can increase accuracy.

In the training of neural networks, in general, the training data may be a subset of real data (that is, data to be processed using the trained neural network), and thus the error on the training data is reduced, but the error on the real data is reduced. There may be increasing learning cycles. Overfitting is a phenomenon in which errors on actual data increase by over-learning on training data as described above. For example, a phenomenon in which a neural network that has learned a cat by showing a yellow cat does not recognize that it is a cat when it sees a cat other than yellow may be a type of overfitting. Overfitting can act as a cause of increasing errors in machine learning algorithms. In order to prevent such overfitting, various optimization methods can be used. In order to prevent overfitting, methods such as increasing the training data, regularization, dropout that deactivate some nodes of the network in the process of learning, and the use of a batch normalization layer can be applied.

3 is a block diagram illustrating an image processing process of a computing device according to an embodiment of the present disclosure.

Referring to FIG. 3 , the processor 110 included in the computing device 100 according to an embodiment of the present disclosure uses a pre-trained first model 210 to obtain a first image 11 and a second image ( 12) Each feature can be extracted. In this case, the first image 11 may be understood as an input image that is a target of style transformation. The second image 12 may be understood as an input image serving as a source of style transformation. The first image 11 and the second image 12 may be images generated to include objects corresponding to each other. Here, the object may be understood to include not only a specific object included in the image, but also a spatial structure, a background, and the like.

For example, the processor 110 may extract the first content feature 13 and the first style feature 14 from the first image 11 using the first model 210 . The processor 110 may extract the second content feature 15 and the second style feature 16 from the second image 12 using the first model 210 . The first model 210 may be pre-trained and generated based on a style transformation model such as multimodal unsupervised image-to-image translation (MUNIT), but this is only an example. Accordingly, various models for extracting content characteristics and style characteristics from an image, which are changeable within a range that those skilled in the art can understand, may be applied to the first model 210 . 3 , it is exemplified that features of the two images 11 and 12 are extracted using the same first model 210 , but a case in which features are extracted using different models for each image may be considered depending on circumstances. can

The processor 110 uses the pre-trained second model 220 , based on the first style feature 14 and the second style feature 16 , respectively, the first image 11 and the second image 12 , respectively. can determine the sensor domain of The processor 110 inputs the first style feature 14 and the second style feature 16 into the second model 220 to determine in which sensor domain each of the two images 11 and 12 was generated, It can be determined whether the sensor domains match each other. For example, the processor 110 may use the second model 220 to map the sensor domain corresponding to each of the first style feature 14 and the second style feature 16 to an electro-optical sensor or synthetic aperture radar (SAR). ) can be classified as sensors. The processor 110 may determine whether the sensor domain of the first style feature 14 and the sensor domain of the second style feature 16 match each other based on the classification result. In other words, if both sensor domains are classified as either electro-optical sensors or synthetic aperture radar (SAR) sensors, then the processor 110 may determine the sensor domain of the first style feature 14 and the sensor domain of the second style feature 16 . It can be determined as a single (mono) domain in which the sensor domains coincide with each other. If one of the two sensor domains is an electro-optical sensor and the other is classified as a synthetic aperture radar (SAR) sensor, the processor 110 configures the sensor domain of the first style feature 14 and the second style feature 16 . It can be determined that the sensor domain of the different heterogeneous (heterogeneous) domain. The processor 110 may perform style conversion by performing different calculation processes according to whether the sensor domain matches.

The processor 110 may use the pre-trained third model 230 to generate the style conversion image 17 based on the determination and comparison results of both sensor domains. The processor 110 inputs the first content feature 13 and the second style feature 16 into the third model 230 to transform the style of the first image 11 into the style of the second image 12 . can In this case, the third model 230 may be selected from among a plurality of candidate models through different calculation processes performed according to whether the sensor domains match and used to transform the style of the first image 11 . For example, when both sensor domains correspond to a single domain, the processor 110 converts the style of the first image 11 among a plurality of candidate models through an additional operation using a characteristic related to an object included in the image. It is possible to select a third model 230 for When both sensor domains are different heterogeneous domains, the processor 110 generates a third model 230 for style conversion of the first image 11 among a plurality of candidate models based on the sensor domain of the first image 11 . can be selected. The processor 110 may generate the style conversion image 17 of the first image 11 by inputting the first content feature 13 and the second style feature 16 into the selected third model 230 . .

The style transformation of the first image 11 reduces the difference in semantic information between the first image 11 and the second image 12 , thereby improving image matching performance for matching features between images.

Hereinafter, a selection operation of a model for style transformation, which is divided according to whether the sensor domains match, and a style transformation operation using the selection model, will be described in more detail.

4 illustrates a process for style conversion of a first image performed when a single domain is determined according to an embodiment of the present disclosure.

Referring to FIGS. 3 and 4 , when the sensor domains between the first image 11 and the second image 12 match, the processor 110 applies the first image 11 and the second image 12 respectively. The third model 230 may be determined based on characteristics related to the included object. In this case, the object-related properties refer to properties related to objects, backgrounds, spatial structures, and the like included in the image. For example, the processor 110 may use the pre-trained fourth model 240 to generate a first image 11 and a second image ( 12) It is possible to classify the properties related to the objects included in each. The processor 110 inputs the first style feature 14 and the second style feature 16 into the fourth model 240, so that the object, background, and It is possible to estimate states, properties, etc. of spatial structures. The processor 110 may select any one of several candidate models 231a, 232a, and 233a as the third model 230a corresponding to the classification characteristic. At this time, each of the plurality of candidate models such as the first candidate model 231a, the second candidate model 232a, ... the Nth candidate model 233a, etc. is pre-trained for each characteristic related to the object and stored in the memory 130. can be stored and managed. When the second candidate model 232a corresponding to the characteristic related to the object classified through the above process is selected as the third model 230a, the processor 110 performs the first content characteristic 13 and the second style characteristic By inputting (16) as the second candidate model 232a selected as the third model 230a, the style conversion image 17a of the first image 11 may be generated. The processor 110 converts the style of the first image 11 into the style of the second image 12 using the third model 230a, so that even in the same sensor domain, due to differences in object-related characteristics, in the registration process, It is possible to improve the problem that the features do not match.

5 illustrates a process for style conversion of a first image performed when it is determined as a heterogeneous domain according to an embodiment of the present disclosure.

3 and 5 , when the sensor domains between the first image 11 and the second image 12 are different, the processor 110 controls the third model 230 based on the sensor domain of the first image 11 . ) can be determined. Specifically, the processor 110 may select the third model 230 that converts the sensor domain of the first image 11 into the sensor domain of the second image 12 from among the plurality of candidate models. For example, when it is determined that the sensor domain of the first image 11 is electro-optical and the sensor domain of the second image 12 is a synthetic aperture radar (SAR), the processor 110 selects a plurality of candidate models ( Among 231b, 232b, and 233b, a second candidate model 232b that converts electro-optics into a synthetic aperture radar may be selected as a third model 230b for style conversion of the first image 11 . The processor 110 inputs the first content feature 13 and the second style feature 16 as the second candidate model 232b selected as the third model 230b, and is a style conversion image of the first image 11 . (17b) can be generated. The processor 110 converts the sensor domain of the first image 11 into the sensor domain of the second image 12 using the third model 230b, so that features are not matched in the registration process due to the difference in sensor domains. Problems that cannot be addressed can be improved.

Hereinafter, as an embodiment of the present disclosure, a process in which a satellite image is input to a computing device and processed will be described in more detail.

6 illustrates a process of classifying a sensor domain of a satellite image according to an embodiment of the present disclosure. 7 illustrates a style conversion process of a satellite image according to a result determined as a single domain according to an embodiment of the present disclosure. 8 is a diagram illustrating a style conversion process of a satellite image according to a result determined as a heterogeneous domain according to an embodiment of the present disclosure.

Referring to FIG. 6 , the processor 110 included in the computing device 100 according to an embodiment of the present disclosure pre-learned each of a target satellite image 21 and a source satellite image 22 photographed in a forest area. Content features 23 and 25 and style features 24 and 26 of each of the satellite images 21 and 22 may be extracted by input to the first model. In this case, the first sub-models 311 and 313 of the first model for extracting the content features 23 and 25 may include a down-sampling neural network and a residual neural network. The second sub-model 312 , 314 of the first model for extracting the style features 24 , 26 may include a down-sampling neural network, a global pooling neural network, and a fully-connected neural network. . The structure of the above-described first model is only an example, and various changes can be made within a range that can be understood by those skilled in the art.

The processor 110 may classify the sensor domains of the target satellite image 21 and the source satellite image 22 based on the first style feature 24 and the second style feature 26 . In this case, the processor 110 may use the pre-trained second model 320 to classify the sensor domains based on whether the sensor domains between the plurality of images match. Specifically, the processor 110 uses the second model 320 to determine whether the sensor domains of the two satellite images 21 , 22 match based on the first style feature 24 and the second style feature 26 . It can be classified as a single domain or a heterogeneous domain according to the

Referring to FIG. 7 , when the sensor domains of the two satellite images 21 and 22 are classified into a matching single domain, the processor 110 uses the fourth model to generate the first style feature 24 and the second style feature. Based on (26), the land use state of the target satellite image 21 and the natural domain indicated by each of the target satellite image 21 and the source satellite image 22 may be classified. Specifically, the processor 110 may input the first style feature 24 as the first sub-model 331 of the fourth model to classify the land use state of the target satellite image 21 as “forest”. . The processor 110 inputs the first style feature 24 and the second style feature 26 into the first sub-model 332 of the fourth model so that the target satellite image 21 and the source satellite image 22 are each The seasons can be classified.

When the target satellite image 21 is classified as "forest", the target satellite image 21 is classified as "summer", and the source satellite image 22 is classified as "winter", the processor 110 is Among the candidate models 341 , 342 , 3343 , and 344 , the candidate model 341 for style-converting “forest-summer” into “forest-winter” is used as a third model for style transformation of the target satellite image 21 . can decide The processor 110 inputs the first content feature 23 and the second style feature 26 to the model 341 selected as the third model, so that the source satellite image 22 is selected in the style of the target satellite image 21 . A satellite image 29 converted to a style can be created.

Referring to FIG. 8 , when the sensor domains of the two satellite images 21 and 22 are classified into different heterogeneous domains, the processor 110 determines the target satellite image 21 based on the sensor domain of the target satellite image 21 . A third model for style transformation may be determined. Specifically, when it is determined that the sensor domain of the target satellite image 21 is the "electro-optic" 31 among the electro-optics 31 or the synthetic aperture radar (SAR) 33, the processor 110 converts the style. A third candidate model 351 for style conversion of "electro-optic" to "synthetic aperture radar (SAR)" among a plurality of candidate models 351 and 352 for can be determined by the model. The processor 110 inputs the first content feature 23 and the second style feature 26 to the model 341 selected as the third model, so that the source satellite image 22 in the sensor domain of the target satellite image 21 is It is possible to generate a satellite image 29 converted to the sensor domain of .

Meanwhile, the processor 110 may generate a plurality of candidate models for style conversion of the satellite image in advance based on the land use state, season, sensor domain, etc. indicated by the satellite image and store it in the memory 13 . For example, the processor 110 first classifies the models based on the land use state of the area in which the satellite image is taken, such as forests, residential districts, highways, agricultural land, etc. . , summer, autumn, winter, by secondarily classifying the models based on the season, it is possible to generate a plurality of candidate models in advance. Also, the processor 110 may classify models based on a sensor domain in which an image is generated, such as electro-optical, synthetic aperture radar (SAR), and the like, and generate a plurality of candidate models in advance. The plurality of candidate models are not limited to the above-described examples, and may be variously classified and generated according to various categories indicating characteristics of an image, such as a country.

9 and 10 are flowcharts illustrating an image processing method for image registration according to an embodiment of the present disclosure.

Referring to FIG. 9 , in step S110 , the computing device 100 according to an embodiment of the present disclosure may receive a first image, which is a target of style change, from a satellite system or the like. Also, the computing device 100 may receive a second image corresponding to a source for style conversion of the first image from a database server or the like. The computing device 100 may extract content features and style features of the input first image and second image, respectively. For example, as shown in FIG. 10 , the computing device 100 uses a deep learning model that can disentangle an image into features with style and content information, regardless of the change in the domain from the first image 41 . The first content feature 43 having certain information and the first style feature 44 having specific information in which domain changes are reflected may be extracted (S211). Also, the computing device 100 may extract the second content feature 45 and the second style feature 46 from the second image 42 using the deep learning model described above ( S212 ).

In operation S120 , the computing device 100 may determine a sensor domain of each of the first image and the second image based on the style feature of the first image and the style feature of the second image. For example, as shown in FIG. 10 , the computing device 100 uses a deep learning model that classifies sensor domains based on whether the sensor domains of both images 41 and 42 match the first style feature 44 and the second It may be determined whether the sensor domains match each other based on the two style features 46 ( S220 ).

In operation S130 , the computing device 100 may convert the style of the first image into the style of the second image based on the determination result of the sensor domain. The computing device 100 may route the deep learning model for style transformation based on the determination result of the sensor domain and transform the style of the first image using the model determined according to the routing. For example, when it is determined that the sensor domains of the first image 41 and the second image 42 match as shown in FIG. 10 , the computing device 100 displays the first image 41 and the second image 44 ) can classify characteristics related to each object (S230). The computing device 100 may select a final model for style transformation through a routing operation to find a model corresponding to an object-related characteristic among a plurality of candidate models pre-trained for style transformation ( S240 ). When it is determined that the sensor domains of the first image 41 and the second image 42 do not match, the computing device 100 selects the sensor domain of the first image 41 from among a plurality of pre-trained candidate models. 2 A final model for style conversion may be selected through a routing operation to find a model converted into the sensor domain of the image 42 (S250). The computing device 100 may perform style conversion of the first image using the final model selected through step S240 or step S250 ( S260 ). The converted image 47 generated through steps S230, S240, and S260 may be an image generated by a style conversion that converts the first image into characteristics related to the object expressed in the second image. The converted image 47 generated through steps S250 and S260 may be an image generated by a style conversion that converts the sensor domain of the first image into the sensor domain of the second image.

11 is a simplified, general schematic diagram of an exemplary computing environment in which embodiments of the present disclosure may be implemented.

Although the present disclosure has been described above generally as being capable of being implemented by a computing device, those skilled in the art will appreciate that the present disclosure is a combination of hardware and software and/or in combination with computer-executable instructions and/or other program modules that may be executed on one or more computers. It will be appreciated that it can be implemented as a combination.

Generally, program modules include routines, programs, components, data structures, etc. that perform particular tasks or implement particular abstract data types. In addition, those skilled in the art will appreciate that the methods of the present disclosure can be applied to single-processor or multiprocessor computer systems, minicomputers, mainframe computers as well as personal computers, handheld computing devices, microprocessor-based or programmable consumer electronics, and the like. It will be appreciated that each of these may be implemented in other computer system configurations, including operable in conjunction with one or more associated devices.

The described embodiments of the present disclosure may also be practiced in distributed computing environments where certain tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.

Computers typically include a variety of computer-readable media. Any medium accessible by a computer can be a computer-readable medium, and such computer-readable media includes volatile and nonvolatile media, transitory and non-transitory media, removable and non-transitory media. including removable media. By way of example, and not limitation, computer-readable media may include computer-readable storage media and computer-readable transmission media. Computer-readable storage media includes volatile and non-volatile media, transitory and non-transitory media, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. includes media. A computer-readable storage medium may be RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital video disk (DVD) or other optical disk storage device, magnetic cassette, magnetic tape, magnetic disk storage device, or other magnetic storage device. device, or any other medium that can be accessed by a computer and used to store the desired information.

Computer readable transmission media typically embodies computer readable instructions, data structures, program modules or other data, etc. in a modulated data signal such as a carrier wave or other transport mechanism, and Includes any information delivery medium. The term modulated data signal means a signal in which one or more of the characteristics of the signal is set or changed so as to encode information in the signal. By way of example, and not limitation, computer-readable transmission media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, RF, infrared, and other wireless media. Combinations of any of the above are also intended to be included within the scope of computer-readable transmission media.

An example environment 1100 implementing various aspects of the disclosure is shown including a computer 1102 , the computer 1102 including a processing unit 1104 , a system memory 1106 , and a system bus 1108 . do. The system bus 1108 couples system components, including but not limited to system memory 1106 , to the processing device 1104 . The processing device 1104 may be any of a variety of commercially available processors. Dual processor and other multiprocessor architectures may also be used as processing unit 1104 .

The system bus 1108 may be any of several types of bus structures that may further be interconnected to a memory bus, a peripheral bus, and a local bus using any of a variety of commercial bus architectures. System memory 1106 includes read only memory (ROM) 1110 and random access memory (RAM) 1112 . A basic input/output system (BIOS) is stored in non-volatile memory 1110, such as ROM, EPROM, EEPROM, etc., which is the basic input/output system (BIOS) that helps transfer information between components within computer 1102, such as during startup. contains routines. RAM 1112 may also include high-speed RAM, such as static RAM, for caching data.

The computer 1102 may also be configured with an internal hard disk drive (HDD) 1114 (eg, EIDE, SATA) - this internal hard disk drive 1114 may also be configured for external use within a suitable chassis (not shown). Yes—a magnetic floppy disk drive (FDD) 1116 (eg, for reading from or writing to removable diskette 1118), and an optical disk drive 1120 (eg, a CD-ROM) for reading from, or writing to, disk 1122, or other high capacity optical media such as DVD. The hard disk drive 1114 , the magnetic disk drive 1116 , and the optical disk drive 1120 are connected to the system bus 1108 by the hard disk drive interface 1124 , the magnetic disk drive interface 1126 , and the optical drive interface 1128 , respectively. ) can be connected to The interface 1124 for external drive implementation includes at least one or both of Universal Serial Bus (USB) and IEEE 1394 interface technologies.

These drives and their associated computer-readable media provide non-volatile storage of data, data structures, computer-executable instructions, and the like. In the case of computer 1102, drives and media correspond to storing any data in a suitable digital format. Although the description of computer readable media above refers to HDDs, removable magnetic disks, and removable optical media such as CDs or DVDs, those skilled in the art will use zip drives, magnetic cassettes, flash memory cards, cartridges, etc. It will be appreciated that other tangible computer-readable media such as etc. may also be used in the exemplary operating environment and any such media may include computer-executable instructions for performing the methods of the present disclosure.

A number of program modules may be stored in the drive and RAM 1112 , including an operating system 1130 , one or more application programs 1132 , other program modules 1134 , and program data 1136 . All or portions of the operating system, applications, modules, and/or data may also be cached in RAM 1112 . It will be appreciated that the present disclosure may be implemented in various commercially available operating systems or combinations of operating systems.

A user may enter commands and information into the computer 1102 via one or more wired/wireless input devices, for example, a pointing device such as a keyboard 1138 and a mouse 1140 . Other input devices (not shown) may include a microphone, IR remote control, joystick, game pad, stylus pen, touch screen, and the like. Although these and other input devices are often connected to the processing unit 1104 through an input device interface 1142 that is connected to the system bus 1108, parallel ports, IEEE 1394 serial ports, game ports, USB ports, IR interfaces, It may be connected by other interfaces, etc.

A monitor 1144 or other type of display device is also coupled to the system bus 1108 via an interface, such as a video adapter 1146 . In addition to the monitor 1144, the computer typically includes other peripheral output devices (not shown), such as speakers, printers, and the like.

Computer 1102 may operate in a networked environment using logical connections to one or more remote computers, such as remote computer(s) 1148 via wired and/or wireless communications. Remote computer(s) 1148 may be workstations, computing device computers, routers, personal computers, portable computers, microprocessor-based entertainment devices, peer devices, or other common network nodes, and are typically connected to computer 1102 . Although it includes many or all of the components described for it, only memory storage device 1150 is shown for simplicity. The logical connections shown include wired/wireless connections to a local area network (LAN) 1152 and/or a larger network, eg, a wide area network (WAN) 1154 . Such LAN and WAN networking environments are common in offices and companies, and facilitate enterprise-wide computer networks, such as intranets, all of which can be connected to a worldwide computer network, for example, the Internet.

When used in a LAN networking environment, the computer 1102 is connected to the local network 1152 through a wired and/or wireless communication network interface or adapter 1156 . Adapter 1156 may facilitate wired or wireless communication to LAN 1152 , which also includes a wireless access point installed therein for communicating with wireless adapter 1156 . When used in a WAN networking environment, the computer 1102 may include a modem 1158, be connected to a communication computing device on the WAN 1154, or establish communications over the WAN 1154, such as over the Internet. have other means. A modem 1158 , which may be internal or external and a wired or wireless device, is coupled to the system bus 1108 via a serial port interface 1142 . In a networked environment, program modules described for computer 1102 , or portions thereof, may be stored in remote memory/storage device 1150 . It will be appreciated that the network connections shown are exemplary and other means of establishing a communication link between the computers may be used.

Computer 1102 may be associated with any wireless device or object that is deployed and operates in wireless communication, for example, printers, scanners, desktop and/or portable computers, portable data assistants (PDAs), communications satellites, wireless detectable tags. It operates to communicate with any device or place, and phone. This includes at least Wi-Fi and Bluetooth wireless technologies. Accordingly, the communication may be a predefined structure as in a conventional network or may simply be an ad hoc communication between at least two devices.

Wi-Fi (Wireless Fidelity) makes it possible to connect to the Internet, etc. without a wire. Wi-Fi is a wireless technology such as cell phones that allows these devices, eg, computers, to transmit and receive data indoors and outdoors, ie anywhere within range of a base station. Wi-Fi networks use a radio technology called IEEE 802.11 (a, b, g, etc.) to provide secure, reliable, and high-speed wireless connections. Wi-Fi can be used to connect computers to each other, to the Internet, and to wired networks (using IEEE 802.3 or Ethernet). Wi-Fi networks can operate in unlicensed 2.4 and 5 GHz radio bands, for example at 11 Mbps (802.11a) or 54 Mbps (802.11b) data rates, or in products that include both bands (dual band). .

On the other hand, a computer-readable medium storing a data structure according to an embodiment of the present disclosure is disclosed.

The data structure may refer to the organization, management, and storage of data that enables efficient access and modification of data. A data structure may refer to an organization of data to solve a specific problem (eg, data retrieval, data storage, and data modification in the shortest time). A data structure may be defined as a physical or logical relationship between data elements designed to support a particular data processing function. The logical relationship between data elements may include a connection relationship between user-defined data elements. Physical relationships between data elements may include actual relationships between data elements physically stored on a computer-readable storage medium (eg, persistent storage). A data structure may specifically include a set of data, relationships between data, and functions or instructions applicable to data. Through an effectively designed data structure, a computing device can perform an operation while using the resources of the computing device to a minimum. Specifically, the computing device may increase the efficiency of operations, reads, insertions, deletions, comparisons, exchanges, and retrievals through effectively designed data structures.

A data structure may be classified into a linear data structure and a non-linear data structure according to the type of the data structure. The linear data structure may be a structure in which only one piece of data is connected after one piece of data. The linear data structure may include a list, a stack, a queue, and a deck. A list may mean a set of data in which an order exists internally. The list may include a linked list. The linked list may be a data structure in which data is linked in such a way that each data is linked in a line with a pointer. In a linked list, a pointer may contain information about a link with the next or previous data. A linked list may be expressed as a single linked list, a doubly linked list, or a circularly linked list according to a shape. A stack can be a data enumeration structure with limited access to data. A stack can be a linear data structure in which data can be processed (eg, inserted or deleted) at only one end of the data structure. The data stored in the stack may be a data structure LIFO-Last in First Out. A queue is a data listing structure that allows limited access to data. Unlike a stack, the queue may be a data structure that comes out later (FIFO-First in First Out) as data stored later. A deck can be a data structure that can process data at either end of the data structure.

The nonlinear data structure may be a structure in which a plurality of data is connected after one data. The nonlinear data structure may include a graph data structure. A graph data structure may be defined as a vertex and an edge, and the edge may include a line connecting two different vertices. A graph data structure may include a tree data structure. The tree data structure may be a data structure in which one path connects two different vertices among a plurality of vertices included in the tree. That is, it may be a data structure that does not form a loop in the graph data structure.

Throughout this specification, computational model, neural network, network function, and neural network may be used interchangeably. Hereinafter, the neural network is unified and described. The data structure may include a neural network. And the data structure including the neural network may be stored in a computer-readable medium. Data structures, including neural networks, also include preprocessed data for processing by the neural network, data input to the neural network, weights of the neural network, hyperparameters of the neural network, data obtained from the neural network, activation functions associated with each node or layer of the neural network, and the neural network. It may include a loss function for learning of . A data structure comprising a neural network may include any of the components disclosed above. That is, the data structure including the neural network includes preprocessed data for processing by the neural network, data input to the neural network, weights of the neural network, hyperparameters of the neural network, data obtained from the neural network, activation functions associated with each node or layer of the neural network, and the neural network It may be configured to include all or any combination thereof, such as a loss function for learning of. In addition to the above-described configurations, a data structure including a neural network may include any other information that determines a characteristic of a neural network. In addition, the data structure may include all types of data used or generated in the operation process of the neural network, and is not limited to the above. Computer-readable media may include computer-readable recording media and/or computer-readable transmission media. A neural network may be composed of a set of interconnected computational units, which may generally be referred to as nodes. These nodes may also be referred to as neurons. A neural network is configured by including at least one or more nodes.

The data structure may include data input to the neural network. A data structure including data input to the neural network may be stored in a computer-readable medium. The data input to the neural network may include learning data input in a neural network learning process and/or input data input to the neural network in which learning is completed. Data input to the neural network may include pre-processing data and/or pre-processing target data. The preprocessing may include a data processing process for inputting data into the neural network. Accordingly, the data structure may include data to be pre-processed and data generated by pre-processing. The above-described data structure is merely an example, and the present disclosure is not limited thereto.

The data structure may include the weights of the neural network. (In this specification, a weight and a parameter may be used interchangeably.) And a data structure including a weight of a neural network may be stored in a computer-readable medium. The neural network may include a plurality of weights. The weight may be variable, and may be changed by the user or algorithm in order for the neural network to perform a desired function. For example, when one or more input nodes are interconnected to one output node by respective links, the output node sets values input to input nodes connected to the output node and links corresponding to the respective input nodes. A data value output from the output node may be determined based on the weight. The above-described data structure is merely an example, and the present disclosure is not limited thereto.

By way of example and not limitation, the weight may include a weight variable in a neural network learning process and/or a weight in which neural network learning is completed. The variable weight in the neural network learning process may include a weight at the start of the learning cycle and/or a variable weight during the learning cycle. The weight for which neural network learning is completed may include a weight for which a learning cycle is completed. Accordingly, the data structure including the weight of the neural network may include a data structure including the weight variable in the neural network learning process and/or the weight in which the neural network learning is completed. Therefore, it is assumed that the above-described weights and/or combinations of weights are included in the data structure including the weights of the neural network. The above-described data structure is merely an example, and the present disclosure is not limited thereto.

The data structure including the weights of the neural network may be stored in a computer-readable storage medium (eg, memory, hard disk) after being serialized. Serialization can be the process of converting a data structure into a form that can be reconstructed and used later by storing it on the same or a different computing device. The computing device may serialize the data structure to send and receive data over the network. A data structure including weights of the serialized neural network may be reconstructed in the same computing device or in another computing device through deserialization. The data structure including the weights of the neural network is not limited to serialization. Furthermore, the data structure including the weights of the neural network is a data structure to increase the efficiency of computation while using the resources of the computing device to a minimum (e.g., B-Tree, Trie, m-way search tree, AVL tree, Red-Black Tree). The foregoing is merely an example, and the present disclosure is not limited thereto.

The data structure may include hyper-parameters of the neural network. In addition, the data structure including the hyperparameters of the neural network may be stored in a computer-readable medium. The hyperparameter may be a variable variable by a user. Hyperparameters are, for example, learning rate, cost function, number of iterations of the learning cycle, weight initialization (e.g., setting the range of weight values subject to weight initialization), Hidden Unit The number (eg, the number of hidden layers, the number of nodes of the hidden layer) may be included. The above-described data structure is merely an example, and the present disclosure is not limited thereto.

One of ordinary skill in the art of this disclosure will understand that information and signals may be represented using any of a variety of different technologies and techniques. For example, data, instructions, instructions, information, signals, bits, symbols and chips that may be referenced in the above description are voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields particles or particles, or any combination thereof.

Those of ordinary skill in the art of the present disclosure will recognize that the various illustrative logical blocks, modules, processors, means, circuits, and algorithm steps described in connection with the embodiments disclosed herein include electronic hardware, (convenience For this purpose, it will be understood that it may be implemented by various forms of program or design code (referred to herein as software) or a combination of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. A person skilled in the art of the present disclosure may implement the described functionality in various ways for each specific application, but such implementation decisions should not be interpreted as a departure from the scope of the present disclosure.

The various embodiments presented herein may be implemented as methods, apparatus, or articles of manufacture using standard programming and/or engineering techniques. The term article of manufacture includes a computer program, carrier, or media accessible from any computer-readable storage device. For example, computer-readable storage media include magnetic storage devices (eg, hard disks, floppy disks, magnetic strips, etc.), optical disks (eg, CDs, DVDs, etc.), smart cards, and flash drives. memory devices (eg, EEPROMs, cards, sticks, key drives, etc.). Also, various storage media presented herein include one or more devices and/or other machine-readable media for storing information.

It is understood that the specific order or hierarchy of steps in the presented processes is an example of exemplary approaches. Based on design priorities, it is understood that the specific order or hierarchy of steps in the processes may be rearranged within the scope of the present disclosure. The appended method claims present elements of the various steps in a sample order, but are not meant to be limited to the specific order or hierarchy presented.

The description of the presented embodiments is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the scope of the present disclosure. Thus, the present disclosure is not intended to be limited to the embodiments presented herein, but is to be construed in the widest scope consistent with the principles and novel features presented herein.

Claims (10)

An image processing method for image registration performed by a computing device comprising at least one processor, comprising:
extracting a first content feature and a first style feature from the first image using the pre-trained first model;
extracting a second content feature and a second style feature from a second image using the first model;
determining a sensor domain of each of the first image and the second image based on the first style feature and the second style feature using a pre-trained second model; and
performing style translation of the first image using a third model determined based on a determination result of the sensor domain;
containing,
method.
The method of claim 1,
The step of determining the sensor domain of each of the first image and the second image comprises:
classifying the sensor domain of each of the first image and the second image based on the first style feature and the second style feature using the second model; and
determining whether a sensor domain matches between the first image and the second image based on a result of the classification;
containing,
method.
3. The method of claim 2,
The step of performing the style conversion of the first image,
determining the third model based on a characteristic related to an object included in each of the first image and the second image when the sensor domains between the first image and the second image match; and
performing style transformation of the first image based on the first content feature and the second style feature using the determined third model;
containing,
method.
4. The method of claim 3,
Determining the third model based on the characteristics related to the object included in each of the first image and the second image,
classifying an object-related feature included in each of the first image and the second image based on the first style feature and the second style feature using a fourth pre-trained model; and
selecting a third model corresponding to the classified characteristic from among a plurality of pre-trained candidate models;
containing,
method.
5. The method of claim 4,
The step of classifying a characteristic related to an object included in each of the first image and the second image includes:
classifying a use state of land included in the first image based on the first style feature, using a first sub-model of the fourth model; and
classifying a season represented by each of the first image and the second image based on the first style feature and the second style feature using a second sub-model of the fourth model;
containing,
method.
3. The method of claim 2,
The step of performing the style conversion of the first image,
determining the third model based on the sensor domain of the first image when the sensor domain between the first image and the second image is different; and
performing style transformation of the first image based on the first content feature and the second style feature using the determined third model;
containing,
method.
7. The method of claim 6,
Determining the third model based on the sensor domain of the first image comprises:
selecting a third model that converts the sensor domain of the first image into the sensor domain of the second image from among a plurality of pre-trained candidate models;
containing,
method.
The method of claim 1,
The sensor domain of each of the first image and the second image,
comprising an electro-optic sensor or a synthetic aperture radar sensor;
method.
A computer program stored in a computer readable storage medium, wherein, when the computer program is executed on one or more processors, it performs the following operations of processing an image for image registration, the operations comprising:
extracting a first content feature and a first style feature from the first image using the pre-trained first model;
extracting a second content feature and a second style feature from a second image using the first model;
determining a sensor domain of each of the first image and the second image based on the first style feature and the second style feature using a pre-trained second model; and
performing style translation of the first image using a third model determined based on a determination result of the sensor domain;
comprising,
A computer program stored in a computer-readable storage medium.
A computing device for processing an image for image registration, comprising:
a processor including at least one core;
a memory containing program codes executable by the processor; and
a network unit for receiving an image;
including,
The processor is
extracting a first content feature and a first style feature from the first image using the pre-trained first model;
extracting a second content feature and a second style feature from a second image using the first model;
using a pre-trained second model to determine a sensor domain of each of the first image and the second image based on the first style feature and the second style feature,
performing style translation of the first image using a third model determined based on the determination result of the sensor domain,
Device.

Images