Remote sensing image change detection method, device and system

Abstract

The present invention discloses a remote sensing image change detection method, device and system, wherein the method comprises: acquiring dual-time remote sensing images of a monitoring area; performing feature extraction on each remote sensing image to obtain spatial domain features; converting the spatial domain features from the spatial domain to the frequency domain to obtain frequency domain features; performing filtering processing on the frequency domain features to obtain filtered frequency domain features; converting the filtered frequency domain features from the frequency domain to the spatial domain to obtain filtered spatial domain features; and determining a change area in the monitoring area according to the filtered spatial domain features respectively corresponding to the dual-time remote sensing images.

Description

Technical Field

[0001] The present invention relates to the technical field of image change detection, and more specifically, to a remote sensing image change detection method, device and system. Background Art

[0002] Drilling construction site management requires real-time monitoring of construction progress, ensuring construction safety, controlling construction quality, and complying with environmental protection standards. In related technologies, drilling construction site management includes two methods: one is mainly based on manual inspections and ground monitoring methods, and the other is a digital management method that uses image data comparison.

[0003] However, manual inspections may be difficult and have great safety risks, and ground monitoring methods may not be able to provide sufficient monitoring accuracy, especially in areas with large terrain changes. The image data collected by the image data comparison method may be greatly affected by pseudo-changes, have errors in geometric alignment, and scale transformations between different data, resulting in low management accuracy for drilling construction sites. Summary of the invention

[0004] An object of an embodiment of the present invention is to provide a new technical solution for remote sensing image change detection to improve the efficiency, accuracy and safety of construction management.

[0005] According to a first aspect of the present invention, a remote sensing image change detection method is provided, which comprises:

[0006] Obtain dual-time remote sensing images of the monitoring area;

[0007] Perform feature extraction on each remote sensing image to obtain spatial domain features;

[0008] Converting the spatial domain features from the spatial domain to the frequency domain to obtain frequency domain features;

[0009] Performing filtering on the frequency domain features to obtain filtered frequency domain features;

[0010] Converting the filtered frequency domain features from the frequency domain to the spatial domain to obtain filtered spatial domain features;

[0011] The changed area in the monitoring area is determined according to the filtered spatial domain features respectively corresponding to the dual-time remote sensing images.

[0012] Optionally, extracting features from each remote sensing image to obtain spatial domain features includes:

[0013] obtaining a difference image according to the dual-time remote sensing image;

[0014] Performing enhancement processing on the changes in the difference image to obtain a change intensity map; wherein the change intensity map reflects the position and intensity of the changes;

[0015] Extracting a change feature map of the dual-time remote sensing image;

[0016] Feature fusion processing is performed on the change intensity map and the change feature map to obtain the spatial domain feature.

[0017] Optionally, obtaining a difference image according to the dual-time remote sensing image includes:

[0018] The difference image is obtained according to the pixel values ​​of corresponding pixel points in the dual-time remote sensing image; wherein the pixel value of each pixel point in the difference image is the difference of the pixel values ​​of the corresponding pixel points in the dual-time remote sensing image.

[0019] Optionally, the enhancing the changes in the difference image includes:

[0020] The changed pixel values ​​in the difference image are enhanced to obtain the change intensity map.

[0021] Optionally, extracting the change features of the dual-time remote sensing image includes:

[0022] Extracting local features of the dual-time remote sensing image to obtain a multi-scale feature map;

[0023] The multi-scale feature map is subjected to multi-scale feature enhancement processing to obtain the change characteristics of the dual-time remote sensing image.

[0024] Optionally, performing multi-scale feature enhancement processing on the multi-scale feature map to obtain the change feature map of the dual-time remote sensing image includes:

[0025] Perform resolution enhancement processing on feature maps of any scale;

[0026] In the process of performing resolution enhancement processing on a feature map of any scale, performing feature recalibration processing on the feature map to obtain a processed feature map;

[0027] The processed feature maps corresponding to the multiple scales are subjected to multi-scale feature fusion processing to obtain the change feature map of the dual-time remote sensing image.

[0028] Optionally, performing feature recalibration processing on the feature map to obtain a processed feature map includes:

[0029] According to the weight corresponding to each spatial position in the feature map, the feature map is subjected to weighted feature enhancement processing to obtain a first feature map;

[0030] According to the weight corresponding to each channel, channel weighted feature processing is performed on the first feature map corresponding to the channel to obtain a second feature map as the processed feature map.

[0031] Optionally, before extracting features from each remote sensing image to obtain spatial domain features, the method further includes:

[0032] Performing image enhancement processing on the dual-time remote sensing image.

[0033] According to a second aspect of the present invention, a remote sensing image change detection device is also provided, which includes a memory and a processor, wherein the memory is used to store executable instructions; the processor is used to operate according to the control of the instructions to execute the method as described in the first aspect of the present invention.

[0034] According to the third aspect of the present invention, there is also provided a remote sensing image change detection system, comprising a remote sensing image acquisition device and the remote sensing image change detection device as described in the second aspect, wherein the remote sensing image acquisition device is used to acquire dual-time remote sensing images of the monitoring area and send them to the remote sensing image change detection device.

[0035] One beneficial effect of the present invention is that by acquiring dual-time remote sensing images of the monitoring area, extracting features from each remote sensing image to obtain spatial domain features, converting the spatial domain features from the spatial domain to the frequency domain to obtain frequency domain features, filtering the frequency domain features to obtain filtered frequency domain features, converting the filtered frequency domain features from the frequency domain to the spatial domain to obtain filtered spatial domain features, and determining the change area in the monitoring area according to the filtered spatial domain features respectively corresponding to the dual-time remote sensing images. It is possible to realize monitoring of the construction site based on high-resolution remote sensing images, thereby improving the efficiency, accuracy and safety of construction management. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention.

[0037] Figure 1 is a schematic diagram of the hardware structure of a remote sensing image change detection system according to an embodiment of the present invention;

[0038] Figure 2 is a schematic diagram of a flow chart of a remote sensing image change detection method according to an embodiment of the present invention;

[0039] Figure 3 is a schematic structural diagram of a remote sensing image change detection device according to an embodiment of the present invention;

[0040] Figure 4Schematic diagram of the structure of a remote sensing image change detection system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0041] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention unless otherwise specifically stated.

[0042] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.

[0043] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0044] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0045] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0046] <Hardware Configuration>

[0047] Figure 1 Schematic diagram of the structure of a remote sensing image change detection system according to an embodiment of the present invention.

[0048] like Figure 1 As shown, the remote sensing image change detection system 100 includes a remote sensing image acquisition device 1000 and a remote sensing image change detection device 2000. The remote sensing image acquisition device 1000 is used to acquire remote sensing images of the monitoring area and provide the acquired remote sensing images to the remote sensing image change detection device 2000. The remote sensing image acquisition device 1000 includes high-resolution satellites such as domestic 2-meter series satellites and Changguang Satellite sub-meter wide-band satellites.

[0049] Those skilled in the art should understand that the high-resolution satellites included in the remote sensing image acquisition device 1000 may also be the satellites shown in Table 1, which is not limited here.

[0050] Table 1 List of common high-resolution satellites

[0051]

[0052] The remote sensing image change detection device 2000 can be any electronic device, such as a PC, a notebook computer, a server, etc.

[0053] In this embodiment, refer to Figure 1 As shown, the remote sensing image change detection device 2000 may include a processor 2100, a memory 2200, an interface device 2300, a communication device 2400, a display device 2500, an input device 2600, a speaker 2700, a microphone 2800, and the like.

[0054] The processor 2100 may be a mobile version processor. The memory 2200 may include, for example, a ROM (read-only memory), a RAM (random access memory), a non-volatile memory such as a hard disk, etc. The interface device 2300 may include, for example, a USB interface, a headphone interface, etc. The communication device 2400 may, for example, be capable of wired or wireless communication. The communication device 2400 may include a short-range communication device, for example, any device for short-range wireless communication based on short-range wireless communication protocols such as Hilink protocol, WiFi (IEEE 802.11 protocol), Mesh, Bluetooth, ZigBee, Thread, Z-Wave, NFC, UWB, LiFi, etc. The communication device 2400 may also include a remote communication device, for example, any device for WLAN, GPRS, 2G / 3G / 4G / 5G remote communication. The display device 2500 may be, for example, a liquid crystal display, a touch display, etc., and the display device 2500 is used to display the collected remote sensing images. The input device 2600 may include, for example, a touch screen, a keyboard, etc. The user may input / output voice information through the speaker 2700 and the microphone 2800.

[0055] In this embodiment, the memory 2200 of the remote sensing image change detection device 2000 is used to store instructions, and the instructions are used to control the processor 2100 to operate to at least perform the remote sensing image change detection method according to any embodiment of the present invention. A technician can design instructions according to the scheme disclosed in the present invention. How instructions control the processor to operate is well known in the art, so it will not be described in detail here.

[0056] Despite Figure 1 , multiple devices of the remote sensing image change detection device 2000 are shown, but the present invention may only involve some of the devices, for example, the remote sensing image change detection device 2000 only involves the memory 2200, the processor 2100 and the display device 2500.

[0057] In this embodiment, the remote sensing image acquisition device is used to acquire remote sensing images of the monitoring area and provide them to the remote sensing image change detection device 2000. The remote sensing image change detection device 2000 implements the method according to any embodiment of the present invention based on the image to obtain the change detection result.

[0058] It should be understood that although Figure 1 Only one remote sensing image acquisition device 1000 and one remote sensing image change detection device 2000 are shown, but this does not mean to limit their respective numbers. The remote sensing image change detection system 100 may include multiple remote sensing image acquisition devices 1000 and / or remote sensing image change detection devices 2000.

[0059] <Method Example>

[0060] Figure 2 2 is a flow chart of a remote sensing image change detection method according to an embodiment of the present invention. The method can be implemented by a remote sensing image change detection device 2000.

[0061] according to Figure 2 As shown, the remote sensing image change detection method of this embodiment may include the following steps S2100-S2600:

[0062] Step S2100, obtaining a dual-time remote sensing image of the monitoring area.

[0063] In this embodiment, the monitoring area may be a geographical area to be monitored. For a drilling site management scenario, the monitoring area is a drilling construction site area. A dual-time remote sensing image may refer to remote sensing images of the same monitoring area at two different time points.

[0064] Remote sensing images can include different types of data, such as image data, elevation data, multispectral data, etc. Each data type provides different information and is suitable for different application scenarios. Among them, image data can be optical remote sensing images, which can be panchromatic or multispectral images, used to observe the visible features of the surface. Elevation data can be obtained through technologies such as radar or lidar to provide elevation information of the surface. Multispectral data captures the characteristics of the earth's surface by using a variety of different spectral bands. It contains images of multiple bands and can reveal specific spectral characteristics of materials such as vegetation and water.

[0065] In one example, the remote sensing image acquisition device can acquire remote sensing images of the monitoring area at a set period for the remote sensing image change detection device to perform change detection. At this time, the two different time points corresponding to the dual-time remote sensing image are the two time points corresponding to the set period.

[0066] In one embodiment, after the dual-time remote sensing image of the monitoring area is acquired, image enhancement processing is performed on the dual-time remote sensing image.

[0067] In this embodiment, the image enhancement processing may be, for example, adjustment, geometric correction, multispectral and panchromatic image registration, panchromatic and multispectral image fusion processing, image color homogenization, mosaicking and framing, etc.

[0068] Adjustment is the process of converting the raw digital counts of image data into surface reflectance or radiance using the sensor's radiometric calibration coefficients and atmospheric correction models. Adjustment of image data in dual-temporal remote sensing images helps correct sensor response differences and make image data comparable at different times or between different sensors. Geometric correction is to adjust image data to align with the ground truth coordinate system using control points and appropriate geometric models to eliminate geometric distortions in the image, such as scale inconsistency, tilt, and distortion caused by terrain, to ensure the accuracy of image data in geospatial space. Multispectral and panchromatic image registration can be achieved by using algorithms such as IHS, Brovey, PCA to fuse the clarity of panchromatic images with the color information of multispectral images to obtain high-resolution color images. Panchromatic and multispectral image fusion processing can be achieved by using fusion techniques such as Laplacian pyramids and HVS (human visual system) models to optimize the visual effects and information content of images to retain details and color information. Image color uniformity is achieved by adjusting the hue, saturation, and brightness of the image, or using color correction algorithms to achieve color uniformity. Mosaic is the process of stitching multiple images into a continuous, seamless image coverage area through image registration, color correction and seamless stitching technology to create a larger surface coverage map. Framing is the process of extracting specific areas or areas of interest from larger images to facilitate analysis or meet specific application requirements.

[0069] Image enhancement processing of dual-time remote sensing images can improve the visual effect of images, increase the usability of remote sensing images, and provide a high-quality data basis for subsequent change detection.

[0070] In one example, in order to facilitate image enhancement processing of dual-time remote sensing images, control point matching can also be performed first. Control point matching can be the process of identifying and corresponding known control points on two or more remote sensing images. These control points have precise geographic coordinates, which are usually obtained through ground measurement and used for subsequent geometric correction and image registration. Through control point matching, a benchmark can be provided for image data for accurate geographic positioning.

[0071] In some examples, before performing image enhancement processing on the dual-time remote sensing image, it is also necessary to perform image preprocessing on the dual-time remote sensing image. Image preprocessing can be, for example, image raw data coverage check, quality check, image registration, etc. Among them, image raw data coverage check can be to check whether the remote sensing image covers the required geographical area. Quality inspection may include checking the clarity, resolution, whether there is cloud cover, image contrast and color balance, etc. of the image. Image registration can be achieved by identifying and matching control points, calculating and applying appropriate geometric transformations, and coordinating interpolation techniques to adjust multi-temporal remote sensing images to achieve precise spatial alignment.

[0072] By preprocessing the dual-time remote sensing images, data support can be provided for subsequent image enhancement processing, facilitating subsequent change detection.

[0073] Step S2200, extracting features from each remote sensing image to obtain spatial domain features.

[0074] In this embodiment, when extracting features from remote sensing images, a variety of artificial intelligence (AI) algorithms can be used, such as Difference-enhancement Dense-attention Convolutional Neural Network (DDCNN), Convolutional Neural Network (CNN), Graph Neural Network (GNN), U-Net network, etc. Those skilled in the art should understand that the manner in which these artificial intelligence algorithms extract features from remote sensing images is well known in the art and will not be elaborated here.

[0075] In one embodiment, the spatial domain features of the remote sensing image are extracted through the DDCNN network, that is, the feature extraction is performed on each remote sensing image in step S2200 to obtain the spatial domain features, including: steps S2201 to S2204.

[0076] Step S2201, obtaining a difference image based on the dual-time remote sensing image.

[0077] In this embodiment, the pixel values ​​of the dual-time remote sensing images can be compared to obtain a difference image, or the pixel value ratio of the remote sensing images at two time points in the same band can be calculated to obtain a difference image, or the texture features of the remote sensing images at two time points, such as roughness, directionality, etc., can be compared to obtain a difference image, etc.

[0078] Those skilled in the art should understand that the method for obtaining the difference image is not limited here.

[0079] In some embodiments, obtaining a difference image according to the dual-time remote sensing image in step S2201 includes: obtaining the difference image according to pixel values ​​of corresponding pixel points in the dual-time remote sensing image.

[0080] In this embodiment, the pixel value of each pixel point in the difference image is the pixel value difference of the corresponding pixel point in the dual-time remote sensing image. In other words, the difference image is obtained according to the pixel value difference of the dual-time remote sensing image.

[0081] Step S2202: performing enhancement processing on the changes in the difference image to obtain a change intensity map.

[0082] In this embodiment, the difference enhancement unit in the DDCNN network can be used to enhance the changes in the difference image, which can improve the visualization and detectability of the changed area.

[0083] In some embodiments, the step S2202 includes performing enhancement processing on the changes in the difference image, including:

[0084] The changed pixel values ​​in the difference image are enhanced to obtain the change intensity map.

[0085] In this embodiment, the enhancement process for the changed pixel value may be contrast enhancement, etc. Contrast enhancement may be to adjust the pixel value of the pixel point corresponding to the changed pixel value in the difference image so as to make the difference between the changed area and the unchanged area more obvious.

[0086] The change intensity map reflects the location and intensity of the changes.

[0087] Step S2203, extracting a change feature map of the dual-time remote sensing image.

[0088] In some embodiments, extracting the change feature map of the dual-time remote sensing image in step S2203 includes: step S2203.1 and step S2203.2.

[0089] Step S2203.1, extracting local features of the dual-time remote sensing image to obtain a multi-scale feature map.

[0090] Step S2203.2, performing multi-scale feature enhancement processing on the multi-scale feature map to obtain a change feature map of the dual-time remote sensing image.

[0091] In this embodiment, when extracting the change feature map of the dual-time remote sensing image through the DDCNN network, the local features of the remote sensing image are first extracted through the convolution unit. The convolution unit is composed of a series of convolution layers, so after the remote sensing image passes through the convolution unit, a multi-scale feature map can be obtained. The multi-scale feature map is subjected to multi-scale feature enhancement processing through the upsampling attention unit to obtain the change feature map of the dual-time remote sensing image. Among them, the multi-scale feature enhancement processing can be a process of analyzing and enhancing the change features of feature maps of different scales.

[0092] According to an embodiment of the present application, by performing multi-scale feature enhancement processing on the multi-scale feature map, a change feature map of the dual-time remote sensing image is obtained, which can highlight the change area and improve the accuracy and robustness of change detection.

[0093] In one embodiment, in step S2203.2, multi-scale feature enhancement processing is performed on the multi-scale feature map to obtain a change feature map of the dual-time remote sensing image, including: steps A1 to A3.

[0094] Step A1: Perform resolution enhancement processing on feature maps of any scale.

[0095] In this embodiment, the resolution of a feature map of any scale is improved by upsampling, wherein upsampling is an operation to improve the spatial resolution of a feature map, which is implemented by an interpolation method, such as nearest neighbor interpolation, bilinear interpolation or transposed convolution.

[0096] Step A2: in the process of performing resolution enhancement processing on a feature map of any scale, performing feature recalibration processing on the feature map to obtain a processed feature map.

[0097] In this embodiment, the feature map may be subjected to feature recalibration processing through the attention mechanism to obtain a processed feature map. The feature recalibration processing may refer to the process of adjusting and optimizing the features in the feature map according to the weights during the resolution enhancement process.

[0098] By performing feature recalibration processing on the feature map through the attention mechanism during the process of performing resolution improvement processing on the feature map, the processed feature map is obtained, which can enhance the changing features, suppress irrelevant or noise features, and improve the expressiveness and discrimination of the features.

[0099] In one embodiment, step A2 performs feature recalibration processing on the feature map to obtain a processed feature map, including: step A21 and step A22.

[0100] Step A21, performing weighted feature enhancement processing on the feature map according to the weight corresponding to each spatial position in the feature map to obtain a first feature map.

[0101] In this embodiment, the attention mechanism includes a spatial attention mechanism and a channel attention mechanism. Among them, the spatial attention mechanism assigns a weight to each spatial position in the feature map, and these weights reflect the importance of the spatial position in the change detection task. The larger the weight, the greater the importance of the spatial position in the change detection task. According to the weight corresponding to each spatial position assigned by the spatial attention mechanism, the feature map is weighted feature enhanced to obtain a first feature map.

[0102] Step A22: performing channel weighted feature processing on the first feature map corresponding to the channel according to the weight corresponding to each channel to obtain a second feature map as the processed feature map.

[0103] In this embodiment, the channel attention mechanism assigns weights to each channel, and these weights indicate the contribution of different feature channels to the change detection task. The channel attention mechanism can perform channel weighted feature processing on the first feature map corresponding to each channel according to the weight corresponding to the channel to obtain a processed feature map.

[0104] Step A3, performing multi-scale feature fusion processing on the processed feature maps corresponding to the multi-scales to obtain the change feature map of the dual-time remote sensing image.

[0105] In this embodiment, an upsampling attention unit may correspond to a feature map of each scale to obtain processed feature maps corresponding to multiple scales, and then these processed feature maps are subjected to multi-scale feature fusion processing to obtain a more comprehensive image representation.

[0106] Multi-scale feature fusion processing can be achieved in many ways, such as feature stacking, feature cascading, or using fusion layers to learn the best combination of features at different scales.

[0107] According to an embodiment of the present application, by performing resolution improvement processing on a feature map of any scale, in the process of performing resolution improvement processing on a feature map of any scale, performing feature recalibration processing on the feature map to obtain a processed feature map, and performing multi-scale feature fusion processing on the processed feature maps corresponding to multiple scales to obtain the change characteristics of the dual-time remote sensing image, which helps to restore the resolution of the image, and also strengthens the change characteristics in the image, thereby improving the accuracy of change detection.

[0108] Step S2204: performing feature fusion processing on the change intensity map and the change feature map to obtain the spatial domain feature.

[0109] In this embodiment, the feature fusion process may be to perform an element-wise multiplication operation on the change intensity map and the change feature map. This multiplication operation may strengthen features that show changes in both images, while suppressing features that are only significant in one of the images.

[0110] The spatial domain feature obtained after feature fusion processing is a feature map that combines two types of information. This map contains both the intensity information of the change and the feature information of the change.

[0111] The fused feature map can provide richer information, which helps to distinguish real changes from false positive changes caused by environmental factors (such as illumination changes and seasonal changes), and improve the robustness of change detection.

[0112] Step S2300: convert the spatial domain features from the spatial domain to the frequency domain to obtain frequency domain features.

[0113] In this embodiment, the spatial domain feature is converted into the frequency domain feature through Fourier transform, that is, the pixel value of the spatial domain feature is converted into the frequency component.

[0114] Those skilled in the art should understand that the method of converting spatial domain features into frequency domain features through Fourier transform is well known in the art and will not be elaborated here.

[0115] By converting spatial domain features into frequency domain features, it is easier to process the features in the frequency domain.

[0116] Step S2400, filtering the frequency domain features to obtain filtered frequency domain features.

[0117] In this embodiment, in the frequency domain, the frequency domain features are multiplied by the parameters of the global filter to enhance or suppress specific frequency components in the image.

[0118] The filter may be a high-pass filter, a low-pass filter, or other types of filters, which are not limited here.

[0119] For example, a high-pass filter can enhance high-frequency components in an image, such as edges and details, while a low-pass filter can smooth an image and reduce noise.

[0120] Step S2500, converting the filtered frequency domain features from the frequency domain to the space domain to obtain filtered space domain features.

[0121] In this embodiment, the filtered frequency domain features are converted from the frequency domain to the spatial domain through inverse Fourier transform.

[0122] Those skilled in the art should understand that the method of converting the filtered frequency domain features from the frequency domain to the spatial domain by inverse Fourier transform is well known in the art and will not be elaborated here.

[0123] Step S2600, determining the changed area in the monitoring area according to the filtered spatial domain features respectively corresponding to the dual-time remote sensing images.

[0124] In this embodiment, a change detection algorithm is used to perform change detection on the filtered spatial domain features to determine the changed area in the monitored area. For example, the feature differences of two filtered spatial domain images can be compared to identify the changed area. The change detection algorithm may use a threshold, a classifier, or other methods to distinguish between changed and non-changed areas, which are not limited here.

[0125] In some embodiments, after determining the change area in the monitoring area, the method further includes classifying the change area according to the physical objects in the change area.

[0126] For example, the changed areas in the monitoring area are determined to be change area 1 and change area 2, and entity object recognition is performed on change area 1 and change area 2 respectively. It is determined that the entity object in change area 1 is the wellhead, and the entity object in change area 2 is the drilling equipment. Then, change area 1 is classified as new drilling, and change area 2 is classified as equipment movement.

[0127] According to the method of this embodiment, by acquiring dual-time remote sensing images of the monitoring area; extracting features from each remote sensing image to obtain spatial domain features; converting the spatial domain features from the spatial domain to the frequency domain to obtain frequency domain features; filtering the frequency domain features to obtain filtered frequency domain features; converting the filtered frequency domain features from the frequency domain to the spatial domain to obtain filtered spatial domain features; and determining the change area in the monitoring area according to the filtered spatial domain features corresponding to the dual-time remote sensing images. It is possible to monitor the construction site based on high-resolution remote sensing images, thereby improving the efficiency, accuracy and safety of construction management.

[0128] <Storage Medium Embodiment>

[0129] An embodiment of the present application provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method described in any of the above method embodiments are implemented.

[0130] <Device Example>

[0131] Figure 3 4 is a structural block diagram of a remote sensing image change detection device 3000 according to an embodiment of the present invention.

[0132] In this embodiment, if Figure 3 As shown, the remote sensing image change detection device 3000 includes a memory 3001 and a processor 3002, wherein the memory 3001 is used to store executable instructions, and the processor 3002 is used to operate according to the control of the instructions to execute the method described in any of the above embodiments.

[0133] <System Embodiment>

[0134] Figure 4 4 is a structural block diagram of a remote sensing image change detection system 4000 according to an embodiment of the present invention.

[0135] The remote sensing image change detection system 4000 comprises a remote sensing image acquisition device 4001 and a remote sensing image change detection device 4002 . The remote sensing image acquisition device 4001 is used to acquire dual-time remote sensing images of a monitoring area and send the images to the remote sensing image change detection device 4002 .

[0136] In one example, the remote sensing image change detection device 4002 may be as follows: Figure 1 or Figure 3 The device shown.

[0137] The present invention may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present invention.

[0138] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the above. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the above. The computer-readable storage medium used herein is not to be interpreted as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0139] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0140] The computer program instructions for performing the operation of the present invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect through the Internet). In some embodiments, by using the state information of the computer-readable program instructions to personalize an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), the electronic circuit may execute the computer-readable program instructions, thereby implementing various aspects of the present invention.

[0141] Various aspects of the present invention are described herein with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to embodiments of the present invention. It should be understood that each box of the flowchart and / or block diagram and the combination of the boxes in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0142] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0143] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0144] The flow charts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a part of a module, a program segment or an instruction, and a part of the module, a program segment or an instruction contains one or more executable instructions for realizing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions. It is well known to those skilled in the art that it is equivalent to implement it by hardware, implement it by software, and implement it by combining software and hardware.

[0145] Embodiments of the present invention have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements in the marketplace, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein. The scope of the present invention is defined by the appended claims.

Claims

1. A remote sensing image change detection method, characterized in that: include: Acquire dual-time remote sensing images of the monitoring area; wherein the dual-time remote sensing images are remote sensing images of the same monitoring area at two different time points; Perform feature extraction on each remote sensing image to obtain spatial domain features; Converting the spatial domain features from the spatial domain to the frequency domain to obtain frequency domain features; Performing filtering on the frequency domain features to obtain filtered frequency domain features; Converting the filtered frequency domain features from the frequency domain to the spatial domain to obtain filtered spatial domain features; Determine the change area in the monitoring area according to the filtered spatial domain features respectively corresponding to the dual-time remote sensing images; The feature extraction of each remote sensing image to obtain spatial domain features includes: obtaining a difference image according to the dual-time remote sensing image; Performing enhancement processing on the changes in the difference image to obtain a change intensity map; wherein the change intensity map reflects the position and intensity of the changes; Extracting a change feature map of the dual-time remote sensing image; Performing feature fusion processing on the change intensity map and the change feature map to obtain the spatial domain feature; Wherein, extracting the change feature map of the dual-time remote sensing image comprises: Extracting local features of the dual-time remote sensing image to obtain a multi-scale feature map; The multi-scale feature map is subjected to multi-scale feature enhancement processing to obtain a change feature map of the dual-time remote sensing image.

2. The method according to claim 1, characterized in that The step of obtaining a difference image according to the dual-time remote sensing image comprises: The difference image is obtained according to the pixel values ​​of corresponding pixel points in the dual-time remote sensing image; wherein the pixel value of each pixel point in the difference image is the difference of the pixel values ​​of the corresponding pixel points in the dual-time remote sensing image.

3. The method according to claim 2, characterized in that The step of enhancing the changes in the difference image comprises: The changed pixel values ​​in the difference image are enhanced to obtain the change intensity map.

4. The method according to claim 1, characterized in that: The performing multi-scale feature enhancement processing on the multi-scale feature map to obtain the change feature map of the dual-time remote sensing image includes: Perform resolution enhancement processing on feature maps of any scale; In the process of performing resolution enhancement processing on a feature map of any scale, performing feature recalibration processing on the feature map to obtain a processed feature map; The processed feature maps corresponding to the multiple scales are subjected to multi-scale feature fusion processing to obtain the change feature map of the dual-time remote sensing image.

5. The method according to claim 4, characterized in that Performing feature recalibration processing on the feature map to obtain a processed feature map includes: According to the weight corresponding to each spatial position in the feature map, the feature map is subjected to weighted feature enhancement processing to obtain a first feature map; According to the weight corresponding to each channel, channel weighted feature processing is performed on the first feature map corresponding to the channel to obtain a second feature map as the processed feature map.

6. The method according to claim 1, characterized in that Before extracting features from each remote sensing image to obtain spatial domain features, the method further includes: Performing image enhancement processing on the dual-time remote sensing image.

7. A remote sensing image change detection device, comprising a memory and a processor, wherein the memory is used to store executable instructions; and the processor is used to operate according to the control of the instructions to execute the method as claimed in any one of claims 1 to 6.

8. A remote sensing image change detection system, comprising a remote sensing image acquisition device and the remote sensing image change detection device as claimed in claim 7, wherein the remote sensing image acquisition device is used to acquire dual-time remote sensing images of a monitoring area and send them to the remote sensing image change detection device.

Images