Map tile generation method and device of navigation map, electronic equipment and storage medium

Abstract

The invention discloses a map tile generation method of a navigation map, which comprises the following steps of: inputting a remote sensing image to be processed into a generator of a navigation map tile generation model to generate a navigation map tile image; wherein the navigation map tile generation model is obtained by training the CycleGAN by adopting a first training set, and pre-training the trained CycleGAN by adopting a second training set and using a Pix2Pix method. According to the method, the corresponding map tile image can be directly generated from the remote sensing image through the navigation map tile generation model based on the generative adversarial network, vector data does not need to be used, the map tile generation method is simplified, the generation efficiency is improved, and meanwhile, the navigation map tile generation model has the capacity of extracting the topological relation in the remote sensing image. The invention further discloses a device for implementing the method, electronic equipment and a computer readable storage medium.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing image processing, and in particular to a method, device, electronic equipment and storage medium for generating map tiles of a navigation map. Background Art

[0002] At present, map services have been widely used in people's work and life, for example, road navigation, location search, terrain analysis, etc. Map service providers publish the acquired data on the Internet, and users can access the map server through a browser or dedicated application software to obtain map data, and then display the map on a personal computer or handheld device. The map displayed by the electronic device is usually composed of multiple map tiles pre-acquired by the acquisition device. Traditional online map tiles (such as Google Maps, Baidu Maps, etc., which are widely used) are derived from the rendering of vector data, and the method of timely updating these online map tiles from vector data is complicated, and the efficiency of generating and updating map tiles is low. Summary of the invention

[0003] In order to solve the above problems existing in the prior art, the present invention provides a method, device, electronic device and storage medium for generating map tiles of a navigation map. The technical problem to be solved by the present invention is achieved through the following technical solutions:

[0004] A first aspect of an embodiment of the present invention provides a method for generating map tiles of a navigation map, comprising the following steps:

[0005] The remote sensing image to be processed is input into the generator of the navigation map tile generation model to generate a navigation map tile image; wherein the navigation map tile generation model is obtained by training CycleGAN with a first training set, and training the trained CycleGAN with a second training set and using the Pix2Pix method; the first training set includes: a first remote sensing image data set and a corresponding binarized data set; the binarized data set is obtained by binarizing the first remote sensing image data set; the second training set includes: a second remote sensing image data set and a corresponding navigation map data set.

[0006] In one embodiment of the present invention, the training step of the navigation map tile generation model includes:

[0007] Construct the first training set and the second training set;

[0008] Using the first training set to train the CycleGAN model to obtain a trained CycleGAN model;

[0009] The second training set is used to train the trained CycleGAN model using the Pix2Pix method, and a navigation map tile generation model is obtained after the training is completed.

[0010] In one embodiment of the present invention, constructing the first training set includes:

[0011] Acquire a first remote sensing image dataset;

[0012] Binarizing the images in the first remote sensing image data set to obtain a binarized data set; wherein the first remote sensing image data set and the binarized data set serve as a first training set;

[0013] The constructing of the second training set comprises:

[0014] Acquiring a second remote sensing image dataset;

[0015] In the navigation map image, a navigation map image containing an object having the same longitude, latitude and shooting angle as the object in the remote sensing image in the second remote sensing image data set is determined as a paired navigation map image; and the paired navigation map images of all the remote sensing images are used as a navigation map data set.

[0016] In one embodiment of the present invention, the second training set is used to train the trained CycleGAN model using the Pix2Pix method, and the navigation map tile generation model is obtained after the training is completed, including:

[0017] The second training set is used to train the generator and discriminator of the trained CycleGAN model using the Pix2Pix method, and a navigation map tile generation model is obtained after the training is completed.

[0018] A second aspect of an embodiment of the present invention provides a map tile generation device for a navigation map, comprising:

[0019] A generation module is used to input the remote sensing image to be processed into the generator of the navigation map tile generation model to generate a navigation map tile image; wherein the navigation map tile generation model is obtained by training CycleGAN using a first training set, and training the trained CycleGAN using a second training set and a Pix2Pix method; the first training set includes: a first remote sensing image data set and a corresponding binarized data set; the binarized data set is obtained by binarizing the first remote sensing image data set; the second training set includes: a second remote sensing image data set and a corresponding navigation map data set.

[0020] In one embodiment of the present invention, it also includes:

[0021] A construction module, used for constructing a first training set and a second training set;

[0022] A first training module, used to train the CycleGAN model using the first training set to obtain a trained CycleGAN model;

[0023] The second training module is used to adopt the second training set and use the Pix2Pix method to train the trained CycleGAN model, and obtain the navigation map tile generation model after the training is completed.

[0024] In one embodiment of the present invention, the building block is further used for:

[0025] Acquire a first remote sensing image dataset;

[0026] Binarizing the images in the first remote sensing image data set to obtain a binarized data set; wherein the first remote sensing image data set and the binarized data set serve as a first training set;

[0027] Acquiring a second remote sensing image dataset;

[0028] In the navigation map image, a navigation map image containing an object having the same longitude, latitude and shooting angle as the object in the remote sensing image in the second remote sensing image data set is determined as a paired navigation map image; and the paired navigation map images of all the remote sensing images are used as a navigation map data set.

[0029] In one embodiment of the present invention, the second training module is further used for:

[0030] The second training set is used to train the generator and discriminator of the trained CycleGAN model using the Pix2Pix method, and a navigation map tile generation model is obtained after the training is completed.

[0031] A third aspect of an embodiment of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a method for generating map tiles for a navigation map provided in the first aspect of an embodiment of the present invention is implemented.

[0032] A fourth aspect of an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for generating map tiles for a navigation map provided by the first aspect of an embodiment of the present invention is implemented.

[0033] Beneficial effects of the present invention:

[0034] The present invention can directly generate corresponding map tile images from remote sensing images through a navigation map tile generation model based on a generative adversarial network, without using vector data, thereby simplifying the map tile generation method and improving the generation efficiency. At the same time, the navigation map tile generation model has the ability to extract topological relationships in remote sensing images.

[0035] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0036] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0038] Figure 1 A schematic flow chart of a method for training a navigation map tile generation model provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The present invention is further described in detail below with reference to specific embodiments, but the embodiments of the present invention are not limited thereto.

[0040] A first aspect of an embodiment of the present invention provides a method for generating map tiles of a navigation map, comprising the following steps:

[0041] Step 11: input the remote sensing image to be processed into the generator of the navigation map tile generation model to generate a navigation map tile image.

[0042] The navigation map tile generation model is obtained by training CycleGAN using the first training set, and pre-training the trained CycleGAN using the second training set and the Pix2Pix method.

[0043] The first training set includes: a first remote sensing image data set and a corresponding binarization data set; the binarization data set is obtained by binarizing the first remote sensing image data set.

[0044] The second training set includes: a second remote sensing image data set and a corresponding navigation map data set.

[0045] In this embodiment, the navigation map tile generation model based on the generative adversarial network can directly generate corresponding map tile images from remote sensing images without using vector data, thereby simplifying the map tile generation method and improving the generation efficiency. At the same time, the navigation map tile generation model has the ability to extract topological relationships in remote sensing images.

[0046] like Figure 1 As shown, a second aspect of an embodiment of the present invention provides a method for training a navigation map tile generation model of a navigation map, comprising the following steps:

[0047] Step 21, constructing a first training set and a second training set.

[0048] Specifically, the specific steps of constructing the first training set include step A1-step A2:

[0049] Step A1, obtaining a first remote sensing image data set.

[0050] Step A2: binarize the images in the first remote sensing image data set to obtain a binarized data set.

[0051] Among them, the first remote sensing image data set and the binarization data set are used as the first training set.

[0052] The specific steps of constructing the second training set include step B1-step B2:

[0053] Step B1, obtaining a second remote sensing image data set.

[0054] Step B2, determining in the navigation map image the navigation map image where the object having the same longitude, latitude and shooting angle as the object in the remote sensing image in the second remote sensing image data set is located as the paired navigation map image. The paired navigation map images of all remote sensing images are used as the navigation map data set. The navigation map data set and the corresponding remote sensing images constitute the second training set.

[0055] Here, the navigation map image and the remote sensing image in the second remote sensing image data set are paired one by one, and the training data of the second training set are paired images. The remote sensing image and the navigation map image with the same position and angle of the object are a pair of paired images. Among them, the image data can also be classified, including season, terrain (urban, rural, mountainous, etc.), and the classification information is used as part of the second training set and participates in the training in the form of text annotations.

[0056] Step 22: Use the first training set to train the CycleGAN model to obtain a trained CycleGAN model.

[0057] In this step, the binary data set in the first training set is used as label data. The training process of the final navigation map tile generation model is divided into two steps. The amount of data in the first training set is large enough, so the first training set is used to train the CycleGAN model as the first step of training, which can solve the problem of insufficient training data during model training.

[0058] Here, the binarized images in the binarized dataset are used to train the model’s ability to extract topological relationships in remote sensing images. There are topological relationships between elements in the images, such as roads, tracks, rivers, etc.

[0059] Step 23: Use the second training set and the Pix2Pix method to train the trained CycleGAN model, and obtain a navigation map tile generation model after the training is completed.

[0060] Specifically, the second training set is used to train the generator and discriminator of the trained CycleGAN model using the Pix2Pix method, and the navigation map tile generation model is obtained after the training. The model obtained after training in this step has global mapping capabilities. The migration model obtained by the training optimization here can perform fast map imaging without extracting individual categories.

[0061] After the navigation map tile generation model is obtained through the training of the above steps, when using the model, the remote sensing image to be processed is input into the model, and a navigation map tile image corresponding to the remote sensing image can be directly output, so that the corresponding navigation map can be updated and processed using the navigation map tile image.

[0062] It should be noted that if the Pix2Pix training method is used directly to obtain the model, the remote sensing image and network map Figure 1 The lack of one-to-one paired data leads to low model accuracy. Moreover, the img2img method cannot preserve the topological structure relationship between pixels in the generated network map.

[0063] A third aspect of an embodiment of the present invention provides a map tile generation device for a navigation map, comprising:

[0064] A generation module is used to input the remote sensing image to be processed into the generator of the navigation map tile generation model to generate a navigation map tile image; wherein the navigation map tile generation model is obtained by training CycleGAN using a first training set, and training the trained CycleGAN using a second training set and a Pix2Pix method; the first training set includes: a first remote sensing image data set and a corresponding binarized data set; the binarized data set is obtained by binarizing the first remote sensing image data set; the second training set includes: a second remote sensing image data set and a corresponding navigation map data set.

[0065] In one embodiment of the present invention, a map tile generation device for a navigation map further includes:

[0066] A construction module, used for constructing a first training set and a second training set;

[0067] A first training module, used to train the CycleGAN model using the first training set to obtain a trained CycleGAN model;

[0068] The second training module is used to train the trained CycleGAN model using the second training set and the Pix2Pix method, and obtain the navigation map tile generation model after the training is completed.

[0069] In one embodiment of the present invention, the building block is further used to:

[0070] Acquire a first remote sensing image dataset;

[0071] Binarizing the images in the first remote sensing image data set to obtain a binarized data set; wherein the first remote sensing image data set and the binarized data set are used as a first training set;

[0072] Acquiring a second remote sensing image dataset;

[0073] In the navigation map image, a navigation map image containing an object having the same longitude, latitude and shooting angle as the object in the remote sensing image in the second remote sensing image data set is determined as a paired navigation map image; and the paired navigation map images of all remote sensing images are used as the navigation map data set.

[0074] In one embodiment of the present invention, the second training module is further used for:

[0075] The second training set is used to train the generator and discriminator of the trained CycleGAN model using the Pix2Pix method, and the navigation map tile generation model is obtained after the training is completed.

[0076] A fourth aspect of an embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a method for generating map tiles for a navigation map provided in the above-mentioned embodiment of the present invention is implemented.

[0077] A fifth aspect of an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of a method for generating map tiles for a navigation map provided in the above-mentioned embodiment of the present invention are implemented.

[0078] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0079] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware systems.

[0080] The method provided in the embodiment of the present invention can be applied to electronic devices. Specifically, the electronic device can be: a desktop computer, a portable computer, an intelligent mobile terminal, a server, etc. This is not limited here, and any electronic device that can implement the present invention belongs to the protection scope of the present invention.

[0081] As for the device / electronic device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0082] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0083] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0084] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0085] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for generating map tiles for a navigation map, characterized in that: The following steps are involved: The remote sensing image to be processed is input into the generator of the navigation map tile generation model to generate a navigation map tile image; wherein the navigation map tile generation model is obtained by training CycleGAN with a first training set, and training the trained CycleGAN with a second training set and using the Pix2Pix method; the first training set includes: a first remote sensing image data set and a corresponding binarized data set; the binarized data set is obtained by binarizing the first remote sensing image data set; the second training set includes: a second remote sensing image data set and a corresponding navigation map data set.

2. The method according to claim 1, characterized in that The training steps of the navigation map tile generation model include: Construct the first training set and the second training set; Using the first training set to train the CycleGAN model to obtain a trained CycleGAN model; The second training set is used to train the trained CycleGAN model using the Pix2Pix method, and a navigation map tile generation model is obtained after the training is completed.

3. The method according to claim 2, characterized in that The step of constructing a first training set comprises: Acquire a first remote sensing image dataset; Binarizing the images in the first remote sensing image data set to obtain a binarized data set; wherein the first remote sensing image data set and the binarized data set serve as a first training set; The constructing of the second training set comprises: Acquiring a second remote sensing image dataset; In the navigation map image, a navigation map image containing an object having the same longitude, latitude and shooting angle as the object in the remote sensing image in the second remote sensing image data set is determined as a paired navigation map image; and the paired navigation map images of all the remote sensing images are used as a navigation map data set.

4. The method according to claim 1, characterized in that The step of adopting the second training set and using the Pix2Pix method to train the trained CycleGAN model, and obtaining a navigation map tile generation model after the training, includes: The second training set is used to train the generator and discriminator of the trained CycleGAN model using the Pix2Pix method, and a navigation map tile generation model is obtained after the training is completed.

5. A map tile generation device for a navigation map, characterized in that: include: A generation module is used to input the remote sensing image to be processed into the generator of the navigation map tile generation model to generate a navigation map tile image; wherein the navigation map tile generation model is obtained by training CycleGAN using a first training set, and training the trained CycleGAN using a second training set and a Pix2Pix method; the first training set includes: a first remote sensing image data set and a corresponding binarized data set; the binarized data set is obtained by binarizing the first remote sensing image data set; the second training set includes: a second remote sensing image data set and a corresponding navigation map data set.

6. The device according to claim 5, characterized in that Also includes: A construction module, used for constructing a first training set and a second training set; A first training module, used to train the CycleGAN model using the first training set to obtain a trained CycleGAN model; The second training module is used to adopt the second training set and use the Pix2Pix method to train the trained CycleGAN model, and obtain the navigation map tile generation model after the training is completed.

7. The device according to claim 6, characterized in that The building blocks are also used to: Acquire a first remote sensing image dataset; Binarizing the images in the first remote sensing image data set to obtain a binarized data set; wherein the first remote sensing image data set and the binarized data set serve as a first training set; Acquiring a second remote sensing image dataset; In the navigation map image, a navigation map image containing an object having the same longitude, latitude and shooting angle as the object in the remote sensing image in the second remote sensing image data set is determined as a paired navigation map image; and the paired navigation map images of all the remote sensing images are used as a navigation map data set.

8. The device according to claim 5, characterized in that The second training module is also used for: The second training set is used to train the generator and discriminator of the trained CycleGAN model using the Pix2Pix method, and a navigation map tile generation model is obtained after the training is completed.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the map tile generation method for the navigation map as described in any one of claims 1 to 4 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the map tile generation method for a navigation map according to any one of claims 1 to 4 is implemented.

Images