Garden landscape tree construction progress monitoring and identifying system
By adaptively obtaining the green detection threshold from RGB images of landscape construction, the problem of missed detections and false detections caused by improper threshold settings in RGB image segmentation is solved, and accurate monitoring of the progress of landscape construction is achieved.
Patent Information
- Application Number
- CN202411456713.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-18
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-10-18
AI Technical Summary
In the monitoring of construction progress in garden landscape projects, existing RGB image segmentation methods are prone to missed or false detection of green pixels due to improperly set thresholds, which affects the accuracy of construction progress identification.
By obtaining the difference and mean of the green channel and other channel components of each pixel in the RGB image of the landscape construction, a template area is selected. The green judgment threshold is adaptively obtained by using the weight and standard difference value within the template area to identify the landscape area.
It enables accurate monitoring of the construction progress of garden landscape projects, reduces missed detections of green pixels and false detections of non-green pixels, and ensures the accuracy of intelligent identification of construction progress and project arrangement.
Smart Images

Figure CN118967844A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of graphic recognition, and in particular to a garden landscape tree construction progress monitoring and recognition system. Background Art
[0002] The landscaping project is a project to build landscape gardens and green spaces. Landscaping provides people with a good place to rest, culturally entertain, get close to nature, and satisfy people's desire to return to nature. It is an important measure to protect the ecological environment and improve the urban living environment. Compared with manual on-site supervision, the intelligent monitoring system has become the main way to supervise the construction progress of garden landscape. The garden landscape is mainly based on the construction of trees and shrubs. The construction progress is mainly judged by identifying the distribution of green plants in the monitoring image of the construction site, and the image is entered into the system; while the trees and shrubs are mainly green, graphic recognition is required to identify the green part in the image to obtain the construction progress, and the image data is stored for subsequent review and control of the construction progress.
[0003] The commonly used RGB image color extraction method is color segmentation, that is, obtaining the channel components of the image pixel points, setting the threshold, and performing image segmentation. If the threshold is set too small, it is easy to cause over-detection of non-target pixels; if the threshold is set too large, it is easy to miss the target pixels, resulting in target loss. At the same time, affected by noise, some pixels will produce color errors, resulting in poor final image segmentation results. Summary of the invention
[0004] In order to solve the above problems, the present invention provides a system for monitoring and identifying the construction progress of landscape trees, the system comprising: Image capture module, to obtain RGB images of landscape construction and design planning drawings; The template region extraction module obtains the difference between the green channel and the red channel components and the difference between the green channel and the blue channel components of each pixel in the garden landscape construction RGB image; obtains the mean of all channel components of each pixel in the garden landscape construction RGB image; obtains suspected obvious green points based on the difference between the green channel and the red channel components of the pixel, the difference between the green channel and the blue channel components and the mean of all channel components; obtains the template region based on the suspected obvious green points; The landscape area extraction module calculates the first weight and the second weight of each pixel point according to the difference between the green channel and the red channel, and the difference between the green channel and the blue channel of all the pixels in the template area; obtains the first standard difference value and the second standard difference value of each pixel point according to the first weight and the second weight of each pixel point, and obtains the red channel difference threshold and the blue channel difference threshold of each template area according to the first standard difference value and the second standard difference value of each pixel point; obtains the landscape area in the RGB image of the garden landscape construction according to the red channel difference threshold and the blue channel difference threshold; The construction progress monitoring module monitors the progress of garden landscape construction according to the area of the landscape area.
[0005] Preferably, the step of determining the difference between the green channel component and the red channel component, and the difference between the green channel component and the blue channel component of the pixel point comprises: in For the The difference between the green channel and other channel components of each pixel; For the The difference between the green channel and the red channel components of each pixel; For the The difference between the green and blue channel components of each pixel.
[0006] Preferably, the suspected obvious green point is obtained by using the mean of all channel components, including the steps of: When the difference between the green channel of a pixel and other channel components is greater than or equal to the mean of all channel components of the pixel, the pixel is considered as a suspected obvious green point. Conversely, when the difference between the green channel of a pixel and other channel components is less than the mean of all channel components of the pixel, the pixel is not considered as a suspected obvious green point.
[0007] Preferably, the step of obtaining the template area according to the suspected obvious green point comprises: When a pixel is a suspected obvious green point and there are at least 4 suspected obvious green points in the eight-neighborhood, the pixel is regarded as an obvious green point; all obvious green points are obtained; Mark the obvious green points as 1 and mark the other pixels as 0 to obtain a binary image of the obvious green points. Perform connected domain analysis on the obvious green points in the binary image to obtain all connected domains. Select any area in the most obvious green connected domain as the template area.
[0008] Preferably, the method for obtaining the most obvious green connected domain is: Get the mean of the sum of the differences between the green channel and other channel components of all pixels in each connected domain, and take the connected domain with the largest mean as the most obvious green connected domain.
[0009] Preferably, the first weight and the second weight of each pixel are calculated according to the difference between the green channel and the red channel, and the difference between the green channel and the blue channel of all the pixels in the template area, respectively, including the steps of: in, In the template area The first weight of pixels; In the template area The second weight of pixels; In the template area The difference between the green channel and the red channel components of each pixel; In the template area The difference between the green channel and blue channel components of each pixel; is a natural constant.
[0010] Preferably, the step of respectively obtaining the first standard difference value and the second standard difference value of each pixel point according to the first weight value and the second weight value of each pixel point comprises the following steps: The standard deviation of the difference between the green channel and the red channel components of all pixels in the template area and the product of the preset hyperparameters are recorded as the first standard difference value; the standard deviation of the difference between the green channel and the blue channel components of all pixels in the template area and the product of the preset hyperparameters are recorded as the second standard difference value.
[0011] Preferably, the step of respectively obtaining the red channel difference threshold and the blue channel difference threshold of each template area according to the first standard difference value and the second standard difference value of each pixel point comprises the following steps: in, is the red channel difference threshold; In the template area The first weight of pixels; In the template area The difference between the green channel and the red channel components of each pixel; is the number of pixels in the template area; is the first standard difference value of all pixels in the template area; is the blue channel difference threshold; In the template area The second weight of pixels; In the template area The difference between the green channel and blue channel components of each pixel; is the second standard difference of all pixels in the template area; Preferably, the step of acquiring the landscape area in the garden landscape construction RGB image according to the red channel difference threshold and the blue channel difference threshold comprises the following steps: When the difference between the green channel and the red channel components of a pixel is greater than the red channel difference threshold in the green judgment threshold, and the difference between the green channel and the blue channel components of the pixel is greater than the blue channel difference threshold in the green judgment threshold, the pixel is regarded as a green pixel; The area formed by all green pixels in the garden landscape construction RGB image is taken as the landscape area.
[0012] Preferably, the monitoring of the progress of garden landscape construction according to the area of the landscape area comprises the following steps: Compare the area of the landscape area with the design plan to monitor and identify the construction progress of the garden landscape trees.
[0013] The present invention has the following beneficial effects: the present invention captures the RGB image of garden landscape construction, obtains suspected obvious green points according to the difference between the green channel of each pixel and other channel components and the mean of all channel components, and then screens out the template area, obtains the first weight and the second weight according to the difference between the green channel of the pixel point and other channel components in the template area, and then obtains the green judgment threshold, obtains the landscape area in the RGB image of garden landscape construction according to the green judgment threshold, and monitors the progress of garden landscape construction according to the area of the landscape area. The present invention adaptively obtains the green judgment threshold. At present, the color segmentation is usually performed by artificially setting the threshold. When the artificially set threshold is too large or too small, it will cause the green pixel point to be missed or the non-green pixel point to be misdetected, thereby causing the landscape area to be incomplete or the landscape area to contain other non-green pixels, thereby affecting the intelligent monitoring of the progress of garden landscape construction. Compared with the prior art, the present invention combines the most obvious green area in the RGB image of garden landscape construction, that is, the template area, and adaptively obtains the green judgment threshold. The obtained landscape area is complete and accurate, ensuring the accuracy of the intelligent recognition of the progress of garden landscape construction, which is conducive to the control of the construction progress and the arrangement of subsequent projects. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0015] Figure 1 A system block diagram of a system for monitoring and identifying the construction progress of landscape trees provided by one embodiment of the present invention; Figure 2 RGB images for landscape construction; Figure 3 It is a landscape area image with a relatively small green judgment threshold; Figure 4 This is the landscape area image when the green judgment threshold is large; Figure 5 It is the landscape area image corresponding to the adaptively acquired green judgment threshold. DETAILED DESCRIPTION
[0016] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a garden landscape tree construction progress monitoring and identification system proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures, or characteristics in one or more embodiments may be combined in any suitable form.
[0017] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
[0018] The specific scheme of a garden landscape tree construction progress monitoring and identification system provided by the present invention is described in detail below with reference to the accompanying drawings.
[0019] See also Figure 1 , which shows a system for monitoring and identifying the construction progress of landscape trees provided by an embodiment of the present invention, the system comprising the following modules: The image capturing module 101 captures the RGB image of the garden landscape construction.
[0020] Build an image capture device, which is composed of an industrial camera. Through this image capture device, RGB images and design planning drawings of the landscape construction site are captured in real time. The RGB images of the landscape construction site can be found in Figure 2 .
[0021] At this point, the garden landscape construction RGB image and design plan are obtained.
[0022] The template region extraction module 102 extracts the template region in the garden landscape construction RGB image.
[0023] It should be noted that during the construction of garden landscape, most of the plants planted are green. Therefore, the green part in the RGB image of garden landscape construction is the landscape area, which is an important criterion for judging the construction progress. In the segmentation of RGB images, due to the influence of noise, color errors will occur in the color of pixels, which will affect color recognition and lead to poor image segmentation. At the same time, most of the existing RGB image segmentation adopts threshold segmentation, and the difference standard between the green channel and other channel components is artificially set as the threshold. When the artificially set threshold is too small, that is, the channel component difference standard is too small, it is easy to identify the pixel points with a small difference from the green channel component as green pixel points, resulting in over-detection of non-target points; when the artificially set threshold is too large, only the pixel points with a large difference between other channels and the green channel component are identified as green, resulting in missed detection of target points, resulting in incomplete extraction of the landscape area in the RGB image of garden landscape construction, affecting the detection of construction progress. Therefore, it is necessary to adaptively obtain the green judgment threshold according to the characteristics of the RGB image of garden landscape construction. In order to obtain an accurate green judgment threshold, it is first necessary to obtain a landscape area with more obvious green performance in the garden landscape construction RGB image, and use this as a standard to obtain the green judgment threshold to obtain a more accurate threshold, thereby reducing over-detection of non-target points and missed detection of target points.
[0024] In the embodiment of the present invention, the difference between the green channel and other channel components of each pixel in the garden landscape construction RGB image is obtained: in For the The difference between the green channel and other channel components of each pixel; For the The difference between the green channel and the red channel components of each pixel; For the The difference between the green channel and blue channel components of each pixel; For the The green component of the pixel, i.e. The value of the green channel of each pixel; For the The red component of the pixel, i.e. The value of the red channel of each pixel; For the The blue component of the pixel, i.e. The value of the blue channel of the pixel; when and When both are larger, The difference between the green channel and other channel components of the pixel The bigger, the The green component of the pixel is more significant than the red and blue components. pixels are more likely to appear green.
[0025] Get the mean of all channel components of each pixel in the landscape construction RGB image: in, For the The green component of each pixel; For the The red component of each pixel; For the The blue component of each pixel.
[0026] When The difference between the green channel and other channel components of the pixel Greater than The mean of all channel components of pixels At that time, The green component of the pixel is very significant compared to the red and blue components. The pixel shows obvious green characteristics. pixel points are regarded as suspected obvious green points; when the The difference between the green channel and other channel components of the pixel Less than or equal to The mean of all channel components of pixels At that time, The green component of the pixel is less significant than the red and blue components. pixels as non-obvious green points.
[0027] Similarly, all the suspected obvious green points in the garden landscape construction RGB image are obtained.
[0028] It should be noted that due to the interference of noise points, some non-green pixels will appear green, or some green pixels will appear other colors, causing color errors and affecting the subsequent acquisition of the green judgment threshold. Therefore, it is necessary to remove the noise points that are mistakenly identified as suspected obvious green points. The green landscape area in the RGB image of landscape construction is continuous, that is, the eight neighborhoods of green pixels are almost all green pixels. Therefore, the color of the pixel in the eight neighborhoods can be used to determine whether the pixel is a noise point.
[0029] In an embodiment of the present invention, if a pixel point is a suspected obvious green point and there are at least 4 suspected obvious green points in its eight-neighborhood area, the pixel point is considered to be an obvious green point; conversely, if the pixel point is a suspected obvious green point and there are less than 4 suspected obvious green points in its eight-neighborhood area, the pixel point is considered to be a noise point.
[0030] Get all the obvious green points. Mark the obvious green points as 1 and the other pixels as 0 to get a binary image of the obvious green points. Perform a connected domain analysis on the obvious green points in the binary image to get all the connected domains. Get the mean of the difference between the green channel and other channel components of all pixels in each connected domain, and take the connected domain with the largest mean as the most obvious green connected domain. Randomly select one in the most obvious green connected domain. The area is used as a template area. In other embodiments, the implementer can set The value of .
[0031] At this point, the template area is obtained.
[0032] It should be noted that the embodiment of the present invention obtains suspected obvious green points by using the difference between the green channel and other channel components of the pixel point, and combines the distribution of the suspected obvious green points to fully utilize the continuity of the landscape area to eliminate noise points, obtain actual obvious green points, and then obtain the template area. Subsequently, according to the component distribution of each channel of the pixel point in the template area, the green judgment threshold is adaptively obtained, which fully considers the connection between the pixel points in the landscape area and avoids the adverse effects of artificially setting too small or too large thresholds on target point recognition.
[0033] The landscape region extraction module 103 extracts the landscape region in the garden landscape construction RGB image according to the template region.
[0034] It should be noted that the template area is part of the most obvious green connected domain, and the most obvious green connected domain is the area where the green is more obvious in the RGB image of the landscape construction. The green judgment threshold is calculated based on this standard, and the obtained green judgment threshold is a more accurate typical green judgment standard. The difference between the green channel and other channels reflects the possibility of the pixel being green. The greater the difference, the greater the possibility that the pixel is green. The smaller the green judgment threshold is set, the greater the possibility that the pixel is judged as green, that is, when the difference between the green channel of the green pixel and other channels is relatively small, the pixel may also be identified as green.
[0035] In the embodiment of the present invention, the difference between the green channel and other channel components of each pixel in the template area is calculated to obtain the first weight and the second weight of each pixel in the template area: in, In the template area The first weight of the pixel point, that is, the first The weight corresponding to the difference between the green channel and the red channel of each pixel; In the template area The second weight of the pixel point, that is, the The weight corresponding to the difference between the green channel and the blue channel of each pixel; In the template area The difference between the green channel and the red channel components of each pixel; In the template area The difference between the green channel and blue channel components of each pixel; is a natural constant; when The greater the difference between the green channel and the red channel components of the pixel, the The first weight of the pixel point is larger, and vice versa; when the first The greater the difference between the green channel and the blue channel components of the pixel, the The smaller the second weight of the pixel, the larger the value.
[0036] Get the standard deviation of the difference between the green channel and the red channel components of all pixels in the template area and the standard deviation of the difference between the green channel and the blue channel components of all pixels in the template area.
[0037] The standard deviation of the difference between the green channel and the red channel components of all pixels in the template area and the product of the preset hyperparameters are recorded as the first standard difference value; the standard deviation of the difference between the green channel and the blue channel components of all pixels in the template area and the product of the preset hyperparameters are recorded as the second standard difference value.
[0038] According to the first weight and the second weight of each pixel in the template area, and the first standard difference value and the second standard deviation value in the template area, the red channel difference threshold and the blue channel difference threshold are obtained: in, is the red channel difference threshold; In the template area The first weight of pixels; In the template area The difference between the green channel and the red channel components of each pixel; is the number of pixels in the template area; is the first standard difference value of all pixels in the template area; is the blue channel difference threshold; In the template area The second weight of pixels; In the template area The difference between the green channel and blue channel components of each pixel; is the second standard difference value of all pixels in the template area; To preset hyper parameters, in the embodiment of the present invention In other embodiments, the implementer can set the preset hyperparameters according to the actual implementation situation. The value of When the red channel difference threshold and the blue channel difference threshold are smaller, the possibility that the pixel is judged as green is greater. When the red channel difference threshold and the blue channel difference threshold are larger, the possibility that the pixel is judged as green is smaller. Each pixel in the template image is an obvious green point. In order to ensure that green pixels are not missed, a larger weight is set for pixels with smaller differences between the green channel and other channel components in the template image, and a smaller weight is set for pixels with larger differences between the green channel and other channel components in the template image. The red channel difference threshold and the blue channel difference threshold are obtained according to the difference between the green channel and other channel components of all pixels in the template image, so as to reduce missed detection of green pixels and ensure the integrity of landscape area extraction.
[0039] Threshold the red channel difference and the blue channel difference threshold Together they serve as the green judgment threshold. Each pixel in the landscape construction RGB image is judged according to the green judgment threshold: When the difference between the green channel and the red channel components of the pixel is greater than the red channel difference threshold in the green judgment threshold , and the difference between the green channel and the blue channel components of the pixel is greater than the blue channel difference threshold in the green judgment threshold , the pixel is regarded as a green pixel.
[0040] All green pixels in the garden landscape construction RGB image are obtained, and the area formed by all green pixels in the garden landscape construction RGB image is the landscape area.
[0041] At this point, the landscape area in the garden landscape construction RGB image is obtained.
[0042] Figure 3 This is the landscape area image obtained when a smaller green judgment threshold is artificially set. Figure 4 This is the landscape area image obtained when a larger green judgment threshold is artificially set. Figure 5 It is a landscape area image obtained by adaptively obtaining the green judgment threshold in the embodiment of the present invention. Figure 3 The green judgment threshold set in is too small, so that the obtained landscape area contains more non-green pixels. Figure 4 The green judgment threshold set in is too large, so the obtained landscape area is incomplete. The landscape area obtained by adaptively obtaining the green judgment threshold in the embodiment of the present invention is complete and does not contain non-green pixels, and the effect of landscape area extraction is better.
[0043] The construction progress monitoring module 104 monitors the progress of the garden landscape construction according to the green area.
[0044] The area of the landscape area in the RGB image of the garden landscape construction is obtained, and compared with the area of the landscape area in the design plan, and the difference between the two is obtained, and the difference reflects the progress of the garden landscape construction. In this way, the progress of the garden landscape construction is monitored.
[0045] In summary, the system of the present invention includes an image capture module, a template area extraction module, a landscape area extraction module, and a construction progress monitoring module. The present invention captures the RGB image of the garden landscape construction, obtains suspected obvious green points according to the difference between the green channel of each pixel and other channel components and the mean of all channel components, and then screens out the template area, obtains the first weight and the second weight according to the difference between the green channel of the pixel point and other channel components in the template area, and then obtains the green judgment threshold, obtains the landscape area in the RGB image of the garden landscape construction according to the green judgment threshold, and monitors the progress of the garden landscape construction according to the area of the landscape area. The present invention adaptively obtains the green judgment threshold. At present, the color segmentation is usually performed by artificially setting the threshold. When the artificially set threshold is too large or too small, it will cause the green pixel point to be missed or the non-green pixel point to be misdetected, thereby causing the landscape area to be incomplete or the landscape area to contain other non-green pixels, thereby affecting the intelligent monitoring of the progress of the garden landscape construction. Compared with the prior art, the present invention combines the most obvious green area in the captured garden landscape construction RGB image, namely the template area, and adaptively obtains the green judgment threshold. The obtained landscape area is complete and accurate, ensuring the accuracy of intelligent identification of the garden landscape construction progress, which is beneficial to the control of the construction progress and the arrangement of subsequent projects.
[0046] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A system for monitoring and identifying the construction progress of landscape trees, characterized in that: The system comprises: Image capture module, to obtain RGB images of landscape construction and design planning drawings; The template region extraction module obtains the difference between the green channel and the red channel components and the difference between the green channel and the blue channel components of each pixel in the garden landscape construction RGB image; obtains the mean of all channel components of each pixel in the garden landscape construction RGB image; obtains suspected obvious green points based on the difference between the green channel and the red channel components of the pixel, the difference between the green channel and the blue channel components and the mean of all channel components; obtains the template region based on the suspected obvious green points; The landscape area extraction module calculates the first weight and the second weight of each pixel point according to the difference between the green channel and the red channel, and the difference between the green channel and the blue channel of all the pixels in the template area; obtains the first standard difference value and the second standard difference value of each pixel point according to the first weight and the second weight of each pixel point, and obtains the red channel difference threshold and the blue channel difference threshold of each template area according to the first standard difference value and the second standard difference value of each pixel point; obtains the landscape area in the RGB image of the garden landscape construction according to the red channel difference threshold and the blue channel difference threshold; The construction progress monitoring module monitors the progress of garden landscape construction according to the area of the landscape area.
2. A system for monitoring and identifying construction progress of landscape trees according to claim 1, characterized in that: The method according to the difference between the green channel and the red channel components of the pixel point, and the difference between the green channel and the blue channel components, comprises the following steps: in For the The difference between the green channel and other channel components of each pixel; For the The difference between the green channel and the red channel components of each pixel; For the The difference between the green and blue channel components of each pixel.
3. A system for monitoring and identifying construction progress of landscape trees according to claim 2, characterized in that: The steps of obtaining the suspected obvious green points using the mean of all channel components are as follows: When the difference between the green channel of a pixel and other channel components is greater than or equal to the mean of all channel components of the pixel, the pixel is considered as a suspected obvious green point. Conversely, when the difference between the green channel of a pixel and other channel components is less than the mean of all channel components of the pixel, the pixel is not considered as a suspected obvious green point.
4. A system for monitoring and identifying construction progress of landscape trees according to claim 2, characterized in that: The step of obtaining the template area according to the suspected obvious green points includes: When a pixel is a suspected obvious green point and there are at least 4 suspected obvious green points in the eight-neighborhood, the pixel is regarded as an obvious green point; Get all the obvious green points; Mark the obvious green points as 1 and mark the other pixels as 0 to obtain a binary image of the obvious green points. Perform connected domain analysis on the obvious green points in the binary image to obtain all connected domains. Select any area in the most obvious green connected domain as the template area.
5. A system for monitoring and identifying construction progress of landscape trees according to claim 4, characterized in that: The method for obtaining the most obvious green connected domain is: Get the mean of the sum of the differences between the green channel and other channel components of all pixels in each connected domain, and take the connected domain with the largest mean as the most obvious green connected domain.
6. A system for monitoring and identifying construction progress of landscape trees according to claim 1, characterized in that: The method of calculating the first weight and the second weight of each pixel point according to the difference between the green channel and the red channel and the difference between the green channel and the blue channel of all the pixels in the template area comprises the following steps: in, In the template area The first weight of pixels; In the template area The second weight of pixels; In the template area The difference between the green channel and the red channel components of each pixel; In the template area The difference between the green channel and blue channel components of each pixel; is a natural constant.
7. A system for monitoring and identifying construction progress of landscape trees according to claim 1, characterized in that: The step of respectively obtaining the first standard difference value and the second standard difference value of each pixel point according to the first weight value and the second weight value of each pixel point comprises the following steps: The standard deviation of the difference between the green channel and the red channel components of all pixels in the template area and the product of the preset hyperparameters are recorded as the first standard difference value; the standard deviation of the difference between the green channel and the blue channel components of all pixels in the template area and the product of the preset hyperparameters are recorded as the second standard difference value.
8. The system for monitoring and identifying the construction progress of landscape trees according to claim 1 is characterized in that: The method of obtaining the red channel difference threshold and the blue channel difference threshold of each template area according to the first standard difference value and the second standard difference value of each pixel point respectively includes the following steps: in, is the red channel difference threshold; In the template area The first weight of pixels; In the template area The difference between the green channel and the red channel components of each pixel; is the number of pixels in the template area; is the first standard difference value of all pixels in the template area; is the blue channel difference threshold; In the template area The second weight of pixels; In the template area The difference between the green channel and blue channel components of each pixel; is the second standard difference of all pixels in the template area.
9. The system for monitoring and identifying the construction progress of landscape trees according to claim 1 is characterized in that: The step of obtaining the landscape area in the garden landscape construction RGB image according to the red channel difference threshold and the blue channel difference threshold comprises the following steps: When the difference between the green channel and the red channel components of a pixel is greater than the red channel difference threshold in the green judgment threshold, and the difference between the green channel and the blue channel components of the pixel is greater than the blue channel difference threshold in the green judgment threshold, the pixel is regarded as a green pixel; The area formed by all green pixels in the garden landscape construction RGB image is taken as the landscape area.
10. A system for monitoring and identifying construction progress of landscape trees according to claim 8, characterized in that: The progress monitoring of the garden landscape construction according to the area of the landscape area includes the following steps: Compare the area of the landscape area with the design plan to monitor and identify the construction progress of the garden landscape trees.
Citation Information
Patent Citations
Method for using BIM data in an augmented reality application and computer program product for carrying out the method.
CH718751A2
Garden landscape greening area intelligent division method based on remote sensing image
CN117291934A
Image processing device and image processing method
US20160063305A1
Image processing-based insulator defect detection method and system
WO2018086299A1
Cited By
Road construction progress tracking and monitoring method
CN120125843A
A road construction progress tracking and monitoring method
CN120125843B
