Keyboard circuit and keyboard for labeling remote sensing image

Abstract

The utility model provides a keyboard circuit for remote sensing image labeling, which relates to the technical field of keyboard circuits and comprises a break point moving key, a function control key and a single chip microcomputer used for controlling terminal equipment to perform remote sensing image labeling according to input commands of the break point moving key and the function control key. One end of the break point moving key and one end of the function control key are electrically connected with the input end of the single-chip microcomputer, the other end of the break point moving key and the other end of the function control key are electrically connected with the grounding end of the single-chip microcomputer, and the output end of the single-chip microcomputer is connected with the terminal equipment. According to the utility model, the labeling operation is effectively simplified, and the label manufacturing speed and the marking accuracy of the folding point position are improved.

Description

Technical Field

[0001] This utility model relates to the technical field of keyboard circuits, and in particular to a keyboard circuit and keyboard for tagging remote sensing images. Background Technology

[0002] Semantic segmentation labeling of remote sensing images refers to assigning a specific semantic category to each pixel in a remote sensing image, so that each pixel in the image corresponds to a certain type of land cover. Semantic segmentation focuses on the fine processing of local features in the image. Each pixel in a remote sensing image usually has semantic information about the land cover category. Therefore, by creating semantic segmentation labels for remote sensing images, accurate detection and identification of surface objects can be achieved. This pixel-level labeling classification method is of great significance for fields such as land use analysis, urban planning, environmental monitoring, and disaster assessment.

[0003] Semantic segmentation label creation for remote sensing images is typically done manually by researchers using a mouse and displayed as a visual mask or label map. The category information for each pixel is stored through digital encoding. The process of creating semantic segmentation labels is complex, involving understanding image data, geospatial understanding, and precise mouse manipulation, resulting in low efficiency. To address these issues, existing methods for creating semantic segmentation labels involve registering the remote sensing image in ArcMap software and then creating vector polygons along the pixel boundaries of semantic segmentation through visual observation and mouse manipulation. Specifically, the process involves: enabling the vector "create element" mode; using the mouse wheel to zoom in on the remote sensing image to a scale where the pixel boundaries are visible and identifiable; visually identifying the semantic segmentation boundaries; precisely moving the mouse to the crosshair point of the pixel boundary (within approximately 1 POS distance from the screen) and left-clicking; then zooming in again to find the next polygon boundary inflection point to be labeled; and zooming in again to a visible pixel boundary scale; and visually identifying the semantic segmentation boundaries. To segment the boundary, precisely move the mouse to the crosshair intersection of the pixel boundary and left-click, repeating this process until the closed polygon is constructed. However, this labeling operation is complex and slow. When using a mouse to label in ArcMap, the mouse wheel needs to be used to zoom out when moving the vertex, focus on the next vertex, and then zoom in to the maximum display scale. The most crucial aspect of mouse control is precisely moving the mouse to the crosshair intersection of four pixels and left-clicking. If the image is not zoomed in to the maximum scale or the click is inaccurate, the semantic label segmentation will be inaccurate, thus compromising the accuracy of the vertex position marking. Summary of the Invention

[0004] To address the problems of complex labeling operations, slow label production speed, and inaccurate fold point marking in existing technologies, this invention proposes a keyboard circuit and keyboard for labeling remote sensing images, which effectively simplifies the labeling operation and improves label production speed and the accuracy of fold point marking.

[0005] To achieve the above-mentioned technical effects, the technical solution of this utility model is as follows:

[0006] A keyboard circuit for tagging remote sensing images includes: a pivot point movement key, a function control key, and a microcontroller for controlling a terminal device to tag remote sensing images according to input commands from the pivot point movement key and the function control key; one end of the pivot point movement key and the function control key are electrically connected to the input terminal of the microcontroller, the other end of the pivot point movement key and the function control key are electrically connected to the ground terminal of the microcontroller, and the output terminal of the microcontroller is connected to the terminal device.

[0007] Preferably, the pivot point movement keys include an up key 7, a down key 4, a left key 3, and a right key 5.

[0008] Preferably, one end of the upper button key7 is connected to the upper button pin KEY07 of the microcontroller, one end of the lower button key4 is electrically connected to the lower button pin KEY04 of the microcontroller, one end of the left button key3 is electrically connected to the lower button pin KEY03 of the microcontroller, one end of the right button key5 is electrically connected to the lower button pin KEY05 of the microcontroller, and the ground terminal GND of the microcontroller is electrically connected to the other ends of the upper button key7, lower button key4, left button key3 and right button key5 respectively.

[0009] Preferably, the upper button constructs a new inflection point n grid units above the previous inflection point, and simultaneously moves n grid pixels upward in the terminal device's screen view, ensuring that the screen display area moves with the constructed new inflection point; the lower button constructs a new inflection point n grid units below the previous inflection point, and simultaneously moves n grid pixels downward in the terminal device's screen view, ensuring that the screen display area moves with the constructed new inflection point; the left button constructs a new inflection point n grid units to the left of the previous inflection point, and simultaneously moves n grid pixels to the left in the terminal device's screen view, ensuring that the screen display area moves with the constructed new inflection point; the right button constructs a new inflection point n grid units to the right of the previous inflection point, and simultaneously moves n grid pixels to the right in the terminal device's screen view, ensuring that the screen display area moves with the constructed new inflection point.

[0010] Preferably, when the upper button key7 is pressed, the microcontroller is triggered to send a first keyboard and mouse operation message corresponding to the upper button key7 to the terminal device; when the lower button key4 is pressed, the microcontroller is triggered to send a second keyboard and mouse operation message corresponding to the lower button key4 to the terminal device; when the left button key3 is pressed, the microcontroller is triggered to send a third keyboard and mouse operation message corresponding to the left button key3 to the terminal device; and when the right button key5 is pressed, the microcontroller is triggered to send a fourth keyboard and mouse operation message corresponding to the right button key5 to the terminal device.

[0011] Preferably, the function control buttons include a delete button (key8), a complete button (key6), an upshift button (key1), and a downshift button (key2).

[0012] Preferably, one end of the delete button key8 is connected to the delete button pin KEY08 of the microcontroller, one end of the complete button key6 is electrically connected to the complete button pin KEY06 of the microcontroller, one end of the upshift button key1 is electrically connected to the upshift button pin KEY01 of the microcontroller, one end of the downshift button key2 is electrically connected to the downshift button pin KEY02 of the microcontroller, and the ground terminal GND of the microcontroller is electrically connected to the other ends of the delete button key8, the complete button key6, the upshift button key1, and the downshift button key2, respectively.

[0013] Preferably, the delete button is used to delete the previous vertex of the construction; the complete button is used to complete the creation of the vector polygon, forming a semantic segmentation vector boundary layer; the upshift button is used to increase the shift by adjusting the number of grid units moved each time, completing the scale scaling corresponding to the shifted grid multiplier; the downshift button is used to decrease the shift by adjusting the number of grid units moved each time, completing the scale scaling corresponding to the shifted grid multiplier.

[0014] Preferably, when the delete key 8 is pressed, the microcontroller is triggered to send a fifth keyboard and mouse operation message corresponding to the delete key 8 to the terminal device; when the complete key 6 is pressed, the microcontroller is triggered to send a sixth keyboard and mouse operation message corresponding to the complete key 6 to the terminal device; when the upshift key 1 is pressed, the microcontroller is triggered to send a seventh keyboard and mouse operation message corresponding to the upshift key 1 to the terminal device; and when the downshift key 2 is pressed, the microcontroller is triggered to send an eighth keyboard and mouse operation message corresponding to the downshift key 2 to the terminal device.

[0015] This invention also proposes a keyboard for tagging remote sensing images, including the keyboard circuit for tagging remote sensing images as described above.

[0016] Compared with the prior art, the beneficial effects of this utility model's technical solution are:

[0017] This invention proposes a keyboard circuit and keyboard for labeling remote sensing images. First, a microcontroller receives input commands from the inflection point movement keys and function control keys to control the terminal device for labeling remote sensing images. Second, the scaling and precise clicking operations required for constructing each inflection point during remote sensing image labeling are integrated into the inflection point movement keys and function control keys. This simplifies the labeling operation, reduces the time spent on scaling, moving, and finding pixel boundary points, and alleviates the effort required to accurately find four pixel boundary points. It can accurately identify the boundaries of semantic segmentation content in remote sensing images over a large scale. While accelerating label production and reducing effort, it also significantly improves the accuracy of boundary construction, ensuring accurate inflection point marking—something that cannot be guaranteed by clicking each pixel individually with a mouse. Furthermore, it shortens the remote sensing image labeling process. Attached Figure Description

[0018] Figure 1 This is a block diagram of a keyboard circuit for tagging remote sensing images, as proposed in an embodiment of this utility model.

[0019] Figure 2 This diagram shows the specific structural connection of a keyboard circuit for tagging remote sensing images, as proposed in an embodiment of the present invention.

[0020] Figure 3 This is a flowchart illustrating the labeling process proposed in the embodiments of this utility model.

[0021] Figure 4 This is a flowchart illustrating the implementation process of the keyboard circuit proposed in this embodiment of the present invention.

[0022] Figure 5 This diagram illustrates a keyboard structure for tagging remote sensing images, as proposed in an embodiment of the present invention.

[0023] 1. Inverted point movement button; 2. Function control button; 3. Microcontroller. Detailed Implementation

[0024] The accompanying drawings are for illustrative purposes only and should not be construed as limiting the scope of this patent.

[0025] To better illustrate this embodiment, some parts of the accompanying drawings may be omitted, enlarged, or reduced, and do not represent actual dimensions. The descriptions of directions such as "up" and "down" are not intended to limit this patent.

[0026] It is understandable to those skilled in the art that some well-known details may be omitted from the accompanying drawings;

[0027] The terms used to describe positional relationships in the accompanying drawings are for illustrative purposes only and should not be construed as limiting this patent.

[0028] The technical solution of this utility model will be further described below with reference to the accompanying drawings and embodiments.

[0029] Example 1

[0030] like Figure 1 As shown, this embodiment proposes a keyboard circuit for tagging remote sensing images, including: a pivot point movement key 1, a function control key 2, and a microcontroller 3 for controlling a terminal device to tag remote sensing images according to input commands from the pivot point movement key 1 and the function control key 2; one end of the pivot point movement key 1 and the function control key 2 are electrically connected to the input terminal of the microcontroller 3, the other end of the pivot point movement key 1 and the function control key 2 are electrically connected to the ground terminal of the microcontroller 3, and the output terminal of the microcontroller 3 is connected to the terminal device.

[0031] The circuit works as follows: First, press the movement button 1 or the function control button 2 to input a command to the microcontroller 3, which drives the microcontroller 3 to output control signals to the terminal device, thereby controlling the terminal device to tag remote sensing images.

[0032] See Figure 2 and Figure 3 The pivot point movement button 1 includes an up button (key7), a down button (key4), a left button (key3), and a right button (key5). The microcontroller 3 is a CH558T circuit board, and the wiring method of the microcontroller 3 is as follows: one end of the up button (key7) is connected to the up button pin (KEY07) of the microcontroller 3; one end of the down button (key4) is electrically connected to the down button pin (KEY04) of the microcontroller 3; one end of the left button (key3) is electrically connected to the down button pin (KEY03) of the microcontroller 3; one end of the right button (key5) is electrically connected to the down button pin (KEY05) of the microcontroller 3; and the ground terminal (GND) of the microcontroller 3 is electrically connected to the other ends of the up button (key7), down button (key4), left button (key3), and right button (key5), respectively.

[0033] The up button creates a new inflection point n grid units above the previous inflection point, and simultaneously moves the screen view of the terminal device upwards by n grid pixels, ensuring that the screen display area moves with the newly created inflection point. The down button creates a new inflection point n grid units below the previous inflection point, and simultaneously moves the screen view of the terminal device downwards by n grid pixels, ensuring that the screen display area moves with the newly created inflection point. The left button creates a new inflection point n grid units to the left of the previous inflection point, and simultaneously moves the screen view of the terminal device to the left by n grid pixels, ensuring that the screen display area moves with the newly created inflection point. The right button creates a new inflection point n grid units to the right of the previous inflection point, and simultaneously moves the screen view of the terminal device to the right by n grid pixels, ensuring that the screen display area moves with the newly created inflection point.

[0034] The microcontroller 3 has a built-in control program module. The control program in the control program module is written in C language, and the parameters are stored in a txt file. The parameters can be adjusted according to the actual screen display of the terminal device. These parameters include: the resolution of the terminal device screen (m, n), the POS value (x1, y1) of the terminal device where the "drag" hand is located in the ArcMap software interface, the POS (x2, y2) of the "face" in the construction tool, the screen POS distance a1 corresponding to the side length of one pixel on a 1 / 500 scale, and the default pixel unit length pixel_size. After the program runs, it reads the corresponding parameters and converts them into values ​​and assigns them to variables in the program.

[0035] If the up button (key7), down button (key4), left button (key3), and right button (key5) are pressed, the control program module in microcontroller 3 will receive the pin message indicating that the corresponding button has been pressed and send the corresponding keyboard and mouse operation message to the terminal device.

[0036] Pressing the up key (key7), down key (key4), left key (key3), and right key (key5) sends the following keyboard and mouse operation messages: Simultaneously press Ctrl+G to open the direction extension mode, sequentially send the corresponding azimuth number 'TAB' key codes to input the azimuth and switch to the length input grid, simultaneously press Ctrl+V to input the extension distance (pixal_size*current_gear) stored in the clipboard, and send the Enter key code to confirm the extension is complete. Move the mouse to the POS value (x1, y1) on the computer screen where the "drag" hand is located, send the left-click code, move the mouse to the center of the screen POS (m / 2, n / 2), hold down the left mouse button, move to (m / 2 + a1, n / 2), and release the left mouse button. Move the mouse to the construction surface button POS (x2, y2), left-click to switch back to the construction surface vector mode, and move the mouse to the center of the screen POS (m / 2, n / 2).

[0037] More specifically, when the upper key 7 is pressed, the control program module in the microcontroller 3 is triggered to send the first keyboard and mouse operation message corresponding to the upper key 7 to the terminal device as follows: 'ctrl+G', '2'7'0' 'TAB', 'ctrl'+'v''enter' Mouse moves to POS(x1,y1), delay 2 milliseconds, "left mouse button", moves to POS(m / 2,n / 2), "left mouse button" is held down, moves to (m / 2-a1,n / 2), delay 2 milliseconds, "left mouse button" is released, mouse moves to POS(x2,y2), delay 2 milliseconds, "left mouse button", moves to POS(m / 2,n / 2).

[0038] When the lower key 4 is pressed, the control program module of the microcontroller 3 is triggered to send the second keyboard and mouse operation message corresponding to the lower key 4 to the terminal device as follows: 'ctrl+G' '9' '0' 'TAB', 'ctrl'+'v' 'enter' The mouse moves to POS(x1,y1), delays for 2 milliseconds, "left mouse button", moves to POS(m / 2,n / 2), "left mouse button" is held down, moves to (m / 2,n / 2+a1), delays for 2 milliseconds, "left mouse button" is released, the mouse moves to POS(x2,y2), delays for 2 milliseconds, "left mouse button", moves to POS(m / 2,n / 2).

[0039] When the left key 3 is pressed, the control program module of the microcontroller 3 is triggered to send the third keyboard and mouse operation message corresponding to the left key 3 to the terminal device as follows: 'ctrl+G' '1' '8' '0' 'TAB', 'ctrl'+'v' 'enter' The mouse moves to POS(x1,y1), delays for 2 milliseconds, "left mouse button", moves to POS(m / 2,n / 2), "left mouse button" is held down, moves to (m / 2+ a1 ,n / 2), delays for 2 milliseconds, "left mouse button" is released, the mouse moves to POS(x2,y2), delays for 2 milliseconds, "left mouse button", moves to POS(m / 2,n / 2).

[0040] When the right key 5 is pressed, the control program module of the microcontroller 3 is triggered to send the fourth keyboard and mouse operation message corresponding to the right key 5 to the terminal device as follows: 'ctrl+G', '0' 'TAB', 'ctrl'+'v' 'enter' The mouse moves to POS(x1,y1), delays for 2 milliseconds, "left mouse button", moves to POS(m / 2,n / 2), "left mouse button" is held down, moves to (m / 2+,n / 2-a1), delays for 2 milliseconds, "left mouse button" is released, the mouse moves to POS(x2,y2), delays for 2 milliseconds, "left mouse button", moves to POS(m / 2,n / 2).

[0041] In this embodiment, the microcontroller first receives input commands from the vertices movement buttons and function control buttons to control the terminal device for remote sensing image tagging. Secondly, the scaling and precise clicking of each vertices during tagging are integrated into the vertices movement buttons and function control buttons, simplifying the tagging process and reducing the time spent scaling, moving, and finding pixel boundary points. This also reduces the effort required to accurately find four pixel boundary points, enabling accurate identification of the boundaries of semantic segmentation content in remote sensing images over a large scale. While accelerating tagging speed and reducing effort, it also significantly improves the accuracy of boundary construction, ensuring accurate vertices are marked—something impossible to guarantee with individual mouse clicks. Furthermore, it shortens the tagging time; for example, it reduces the time for creating semantic segmentation tags for north-south extending plots to 50%, and for plots not extending north-south to a quarter of the time. Additionally, converting mouse operations to keyboard operations increases the fun of tagging, reminiscent of the Snake game. At the same time, it lowers the technical threshold for tagging, making it easy for people without relevant professional backgrounds to get started, which also provides convenience for related technical work.

[0042] Example 2

[0043] See Figure 2 and Figure 3 The function control buttons 2 include a delete button (key8), a complete button (key6), an upshift button (key1), and a downshift button (key2).

[0044] One end of the delete button key8 is connected to the delete button pin KEY08 of the microcontroller 3. One end of the complete button key6 is electrically connected to the complete button pin KEY06 of the microcontroller 3. One end of the upshift button key1 is electrically connected to the upshift button pin KEY01 of the microcontroller 3. One end of the downshift button key2 is electrically connected to the downshift button pin KEY02 of the microcontroller 3. The ground terminal GND of the microcontroller 3 is electrically connected to the other ends of the delete button key8, the complete button key6, the upshift button key1, and the downshift button key2, respectively.

[0045] The delete button is used to delete the previous vertex of the construction; the complete button is used to complete the creation of the vector polygon, forming a semantic segmentation vector boundary layer; the upshift button is used to increase the shift by adjusting the number of grid units moved each time, completing the scale scaling corresponding to the shifted grid multiplier; the downshift button is used to decrease the shift by adjusting the number of grid units moved each time, such as 1, 2, 3, 5, completing the scale scaling corresponding to the shifted grid multiplier.

[0046] The gear shifting mechanism utilizes the principle that the change in map scale is inversely proportional to the change in actual distance, and that each pixel is a square of equal size. By scaling the map, the number of nodes that advance can be multiplied without changing the POS value of each screen range movement. Through observation and testing, the following scale was chosen: 1 / 500 for gear 1, scrolling 3 increments with the mouse to 1 / 1000 for gear 2, scrolling 2 increments with the mouse wheel to 1 / 1500 for gear 3, and scrolling 2 increments with the mouse wheel to 1 / 2500 for gear 5. The current gear is stored using `current_gear`.

[0047] When the delete key (key8) is pressed, the control program module of the microcontroller 3 sends the fifth keyboard and mouse operation message corresponding to the delete key (key8) to the terminal device, which is as follows: "Right mouse button" "C". The previous inflection point can be deleted by calling the "Change Length" option in ArcGIS.

[0048] When the completion button (key6) is pressed, the control program module of the microcontroller 3 sends the sixth keyboard and mouse operation message corresponding to the completion button (key6) to the terminal device. The message is as follows: "F2" can quickly invoke the ArcGIS shortcut key to complete a sketch. This completes the construction of a field surface vector.

[0049] When the shift key (key1) is pressed, the control program module of the microcontroller 3 sends the seventh keyboard and mouse operation message corresponding to the shift key (key1) to the terminal device as follows: First, move the mouse to the center point of the screen (m / 2, n / 2). If the key is pressed at shift 5, continue skipping this loop and waiting for the next key press. If the key is pressed at shift 1, send 3 "-" signals to adjust the display scale. If the key is pressed at shift 2 or 3, send 2 "-" signals to adjust the display scale. Then, multiply the actual pixel size by the corresponding shift number (pixal_size*current_gear) and write it to the clipboard to change the actual distance of the constructed vector movement in the map.

[0050] When the downshift button key2 is pressed, the control program module of the microcontroller 3 sends the eighth keyboard and mouse operation message corresponding to the downshift button key2 to the terminal device as follows: First, move the mouse to the center point of the screen (m / 2, n / 2). If it is pressed at level 1, continue skipping this loop and waiting for the next key press. If it is pressed at level 2, send 3 "+" signals to adjust the display scale. If it is pressed at levels 2 or 3, send 2 "+" signals to adjust the display scale. Then, multiply the actual pixel size by the corresponding level (pixel_size*current_gear) and write it to the clipboard to change the actual distance of the constructed vector movement in the map.

[0051] It should be specifically stated that, see [link to relevant document] Figure 4 The system interacts with the terminal device via the keyboard circuit used for labeling remote sensing images, reads user-set parameters, sends continuous keyboard and mouse signals, and controls the clipboard content, thereby automating vector polygon construction operations in ArcGIS software. The up, down, left, and right arrow keys send continuous key codes to invoke ArcGIS shortcuts for direction input and construct polygon vectors by inputting extended lengths through the clipboard. Mouse control signals are sent to switch to the panning tool and translate the point a corresponding distance along the construction direction to ensure the construction point is centered on the screen. Simultaneously, the map scale is adjusted using the zoom level to ensure the view follows the construction point, and the program writes the extended lengths from the map to the clipboard at the corresponding scale multiple. The delete key invokes ArcGIS's delete vertex shortcut, and the complete key invokes the software's complete construction shortcut. By sending continuous key codes and clipboard data, multiple continuous but repetitive keyboard and mouse operations are integrated into the keyboard, enabling one-click generation of vertexes and adjustment of the distance between each generated vertex.

[0052] Example 3

[0053] See Figure 5This embodiment proposes a keyboard for tagging remote sensing images, including a keyboard circuit for tagging remote sensing images as described in the above embodiment. The circuit includes: a pivot point movement key 1, a function control key 2, and a microcontroller 3 for controlling a terminal device to tag remote sensing images according to the input commands of the pivot point movement key 1 and the function control key 2. One end of the pivot point movement key 1 and the function control key 2 are electrically connected to the input terminal of the microcontroller 3, and the other end of the pivot point movement key 1 and the function control key 2 are electrically connected to the ground terminal of the microcontroller 3. The output terminal of the microcontroller 3 is connected to the terminal device. First, the microcontroller receives input commands from the inflection point movement buttons and function control buttons to control the terminal device for labeling remote sensing images. Second, the process of scaling each inflection point and precisely clicking the crosshair with a mouse during remote sensing image labeling is integrated into the inflection point movement buttons and function control buttons. This simplifies the labeling operation, reduces the time spent on scaling, moving, and finding pixel boundary points, and also reduces the effort required to accurately find four-pixel boundary points. It can accurately identify the boundaries of semantic segmentation content in remote sensing images within a large scale. While speeding up label production and reducing effort, it also significantly improves the accuracy of boundary construction, ensuring accurate inflection point marking, which cannot be guaranteed by clicking the mouse one by one. It also shortens the remote sensing image labeling time.

[0054] The keyboard is designed using the characteristics of raster images: each grid cell is the same size and a regular square, with two opposite sides parallel to the two opposite sides of the display screen. Unlike vector layers, it is not distorted by the coordinate system. The keyboard also utilizes the characteristic that the scaling factor is inversely proportional to the actual distance scaling factor.

[0055] The pivot point movement key 1 includes an up key 7, a down key 4, a left key 3, and a right key 5; the function control key 2 includes a delete key 8, a complete key 6, an up key 1, and a down key 2; therefore, the keyboard has 8 key covers corresponding to the 8 keys. The right hand controls 6 key covers, namely "up," "down," "left," "right," "delete," and "complete." The left hand controls 2 key covers, namely "up" and "down." The "up," "down," "left," and "right" key covers are arranged according to the up, down, left, and right keys, with the "delete" key in the upper right corner. The layout is quite similar to a real keyboard and is easy to use.

[0056] The keyboard layout is shown in the image below. It uses a 1.5mm thick acrylic sheet as the base and outer shell. Six anti-slip pads are attached to the bottom of the shell, allowing it to be placed on a desktop or laptop touchpad for operation. It measures 13.5cm in length and 5.5cm in width, and can also be held with both hands like a game controller for tagging.

[0057] This embodiment also uses the keyboard to create semantic segmentation labels for fields in an arid region. These fields are approximately 200m x 800m apart, with a pixel size of 2.5012784m, extending along a true north-south and east-west direction. Researchers skilled in semantic label creation and related research would need nearly 40 minutes to create labels. However, a non-technical person, with simple visual guidance, can create semantic labels using the keyboard in about 15 minutes. For fields not extending north-south, due to numerous pixel inflections, constructing surface vectors requires many vertices, and sloping fields are generally not used for semantic segmentation labeling. Professional researchers using a mouse to construct surface vectors would take over an hour, which is inefficient. Using the keyboard, however, it's easy to "climb stairs" along pixel boundaries to construct semantic segmentation labels for sloping fields. The time required is similar to that for fields extending in a true north-south and east-west direction; a non-professional can complete the semantic segmentation label creation for a single field in about twenty minutes. Since only the first vertices need to be accurately located, the accuracy of all subsequent vertices in that vector is guaranteed. Using a keyboard can significantly improve the accuracy of data entry and the quality of semantic segmentation tag creation. Furthermore, compared to the tedious, laborious, and time-consuming mouse operation, keyboard clicking offers a far more engaging and immersive experience than traditional mouse-based tag creation.

[0058] Obviously, the above embodiments of this utility model are merely examples for clearly illustrating this utility model, and are not intended to limit the implementation of this utility model. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this utility model should be included within the protection scope of the claims of this utility model.

Claims

1. A keyboard circuit for tagging remote sensing images, characterized in that, include: The device includes a pivot point movement button (1), a function control button (2), and a microcontroller (3) for controlling the terminal device to tag remote sensing images according to the input commands of the pivot point movement button (1) and the function control button (2). One end of the pivot point movement button (1) and the function control button (2) are electrically connected to the input terminal of the microcontroller (3), and the other end of the pivot point movement button (1) and the function control button (2) are electrically connected to the ground terminal of the microcontroller (3). The output terminal of the microcontroller (3) is connected to the terminal device.

2. The keyboard circuit for tagging remote sensing images according to claim 1, characterized in that, The fold point movement button (1) includes an up button (key7), a down button (key4), a left button (key3), and a right button (key5).

3. The keyboard circuit for tagging remote sensing images according to claim 2, characterized in that, One end of the upper button key7 is connected to the upper button pin KEY07 of the microcontroller (3), one end of the lower button key4 is electrically connected to the lower button pin KEY04 of the microcontroller (3), one end of the left button key3 is electrically connected to the lower button pin KEY03 of the microcontroller (3), one end of the right button key5 is electrically connected to the lower button pin KEY05 of the microcontroller (3), and the ground terminal GND of the microcontroller (3) is electrically connected to the other ends of the upper button key7, the lower button key4, the left button key3 and the right button key5 respectively.

4. The keyboard circuit for tagging remote sensing images according to claim 2, characterized in that, The up button creates a new inflection point n grid units above the previous inflection point, and simultaneously moves the screen view of the terminal device upwards by n grid pixels, ensuring that the screen display area moves with the newly created inflection point. The down button creates a new inflection point n grid units below the previous inflection point, and simultaneously moves the screen view of the terminal device downwards by n grid pixels, ensuring that the screen display area moves with the newly created inflection point. The left button creates a new inflection point n grid units to the left of the previous inflection point, and simultaneously moves the screen view of the terminal device to the left by n grid pixels, ensuring that the screen display area moves with the newly created inflection point. The right button creates a new inflection point n grid units to the right of the previous inflection point, and simultaneously moves the screen view of the terminal device to the right by n grid pixels, ensuring that the screen display area moves with the newly created inflection point.

5. The keyboard circuit for tagging remote sensing images according to claim 2, characterized in that, When the upper key 7 is pressed, the microcontroller (3) is triggered to send the first keyboard and mouse operation message corresponding to the upper key 7 to the terminal device; when the lower key 4 is pressed, the microcontroller (3) is triggered to send the second keyboard and mouse operation message corresponding to the lower key 4 to the terminal device; when the left key 3 is pressed, the microcontroller (3) is triggered to send the third keyboard and mouse operation message corresponding to the left key 3 to the terminal device; when the right key 5 is pressed, the microcontroller (3) is triggered to send the fourth keyboard and mouse operation message corresponding to the right key 5 to the terminal device.

6. The keyboard circuit for tagging remote sensing images according to claim 1, characterized in that, The function control buttons (2) include a delete button (key8), a complete button (key6), an upshift button (key1), and a downshift button (key2).

7. The keyboard circuit for tagging remote sensing images according to claim 6, characterized in that, One end of the delete button key8 is connected to the delete button pin KEY08 of the microcontroller (3), one end of the complete button key6 is electrically connected to the complete button pin KEY06 of the microcontroller (3), one end of the upshift button key1 is electrically connected to the upshift button pin KEY01 of the microcontroller (3), one end of the downshift button key2 is electrically connected to the downshift button pin KEY02 of the microcontroller (3), and the ground terminal GND of the microcontroller (3) is electrically connected to the other end of the delete button key8, the complete button key6, the upshift button key1 and the downshift button key2 respectively.

8. The keyboard circuit for tagging remote sensing images according to claim 6, characterized in that, The delete button is used to delete the previous vertex of the construction; the complete button is used to complete the creation of the vector polygon, forming a semantic segmentation vector boundary layer; the upshift button is used to increase the shift by adjusting the number of raster units moved each time, completing the scale scaling corresponding to the shift raster multiplier; the downshift button is used to decrease the shift by adjusting the number of raster units moved each time, completing the scale scaling corresponding to the shift raster multiplier.

9. The keyboard circuit for tagging remote sensing images according to claim 6, characterized in that, When the delete key 8 is pressed, the microcontroller (3) is triggered to send the fifth keyboard and mouse operation message corresponding to the delete key 8 to the terminal device; when the complete key 6 is pressed, the microcontroller (3) is triggered to send the sixth keyboard and mouse operation message corresponding to the complete key 6 to the terminal device; when the upshift key 1 is pressed, the microcontroller (3) is triggered to send the seventh keyboard and mouse operation message corresponding to the upshift key 1 to the terminal device; when the downshift key 2 is pressed, the microcontroller (3) is triggered to send the eighth keyboard and mouse operation message corresponding to the downshift key 2 to the terminal device.

10. A keyboard for tagging remote sensing images, characterized in that, Includes a keyboard circuit for tagging remote sensing images as described in any one of claims 1-9.

Images