CN115661382A - Filtering method based on morphological template angle protection in three-dimensional landform reconstruction - Google Patents
Filtering method based on morphological template angle protection in three-dimensional landform reconstruction Download PDFInfo
- Publication number
- CN115661382A CN115661382A CN202211167831.9A CN202211167831A CN115661382A CN 115661382 A CN115661382 A CN 115661382A CN 202211167831 A CN202211167831 A CN 202211167831A CN 115661382 A CN115661382 A CN 115661382A
- Authority
- CN
- China
- Prior art keywords
- map
- pixel
- landform
- image
- dsm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Image Processing (AREA)
Abstract
The invention provides a filtering method based on morphological template conformal in three-dimensional landform reconstruction, which relates to the technical field of three-dimensional landform reconstruction and solves the problem that the current earth surface filtering method cannot accurately identify and retain the edges and the angular points of a landform, so that the real contour of the landform cannot be accurately recovered.
Description
Technical Field
The invention relates to the technical field of three-dimensional landform reconstruction, in particular to a filtering method based on morphological template angle protection in three-dimensional landform reconstruction.
Background
When the Surface building is restored with the real landform and landform, the building is usually three-dimensionally reconstructed through a multi-view image, the method is to perform corresponding post-processing optimization on a middle result disparity map, a depth map or a Digital Surface Model (DSM map), and the disparity map, the depth map and the DSM map of the landform can be directly back-calculated to a target landform point cloud through a camera photo pose, so that clear, smooth and complete characteristics of the landform in the disparity map and the depth map can provide a better point cloud initial value for a subsequent higher-quality three-dimensional Model, and accordingly, the real three-dimensional reconstruction landform contour characteristics can be restored.
In the traditional method, a disparity map, a depth map or a DSM (digital surface model) map is directly subjected to post-filtering processing by methods such as median filtering and bilateral filtering, but only certain smooth effect of the image at the edge can be ensured, a better real contour is difficult to restore at the corner of a ground object, so that the quality and precision of a subsequent reconstruction model are influenced, in addition, edge enhancement processing needs to be carried out on the corrected image, when the edge information of the corrected image is not accurate, a post-processing result which is more and more deteriorated is obtained, and the real terrain can not be restored. The prior art discloses a ground surface filtering method, which comprises the steps of firstly extracting buildings in a digital orthophoto map, then weighting the building extraction result with geometric information of a DSM (digital projection system) map, calculating 'building factors' in a sliding window of the DSM map, dynamically adjusting filtering parameters in the sliding window according to the terrain, and finally realizing progressive morphological filtering of adaptive parameters.
Disclosure of Invention
In order to solve the problem that the real contour of the ground object cannot be accurately recovered due to the fact that the edge and the angular point of the ground object cannot be accurately identified and reserved by the current ground surface filtering method, the invention provides a filtering method based on morphological template angle preservation in three-dimensional landform reconstruction.
In order to achieve the technical effects, the technical scheme of the invention is as follows:
a filtering method based on morphological template conformal in three-dimensional landform reconstruction comprises the following steps:
s1, determining a target ground object to be processed, and acquiring a landform representation map of the target ground object;
s2, determining the regular shape of the angular point, and constructing a plurality of morphological templates with the same window size based on the regular shape of the angular point;
s3, collecting a specific window of the landform representation image, preprocessing the specific window, and acquiring a pixel code of the specific window;
s4, judging whether the pixel code obtained in the S3 is matched with any one morphological template, if so, taking a specific window corresponding to the current pixel code as an angular point, carrying out angle-preserving weight filtering on the current pixel code, and executing the step S5; otherwise, the specific window corresponding to the current pixel code is the edge, median filtering is carried out on the current pixel code, and the step S5 is executed;
s5, processing the filtered landform representation image to obtain a processed landform representation image;
and S6, reconstructing a three-dimensional model of the target ground object based on the processed landform characterization map obtained in the S5.
In the technical scheme, a target feature to be processed is determined, a landform representation image of the target feature is obtained, a plurality of morphological templates with the same specification are designed based on the regular shapes of corner points, the purpose is to include the shapes of all regular corner points, subsequent corner point matching and preservation are facilitated, then the designed morphological templates are used for detecting the positions of the edge to be processed and the corner points in the landform representation image, whether pixel codes of a specific window of the landform representation image are matched with any one of the morphological templates is judged, if matching is successful, the specific window corresponding to the current pixel codes is the corner point, angle-preserving weighted filtering processing is carried out on the current pixel codes, otherwise, median filtering is carried out on the current pixel codes to guarantee the smoothness of the image at the edge, the filtered landform representation image is processed, a three-dimensional model of the target feature is reconstructed according to the final landform representation image, accurate recognition and preservation of the edge and the corner points of the feature are achieved, and the real contour of the feature is accurately recovered.
Preferably, the topographic characterization map is a depth map, a disparity map or a DSM map, in step S2, each morphological template corresponds to a regular shape of one corner point, and the pixel codes of the color image and the pixel codes of the elevation image in each morphological template are templates of a binary string, and the rules of the binary coding are as follows: taking the difference value between the gray value of the central pixel point of the morphological template and the gray values of the surrounding pixel points as a coded binary template string, and if the difference value between the gray value of the central pixel point and the gray value of the surrounding pixel points is greater than a set depth threshold, the binary bit code of the current pixel point is 1; otherwise, the binary bit code of the current pixel point is 0.
Preferably, in step S3, the size of the specific windows for collecting the geomorphic representation map is the same.
Preferably, in step S3, the specific process of the preprocessing is:
s31, taking the pixel of the landform representation image as a central pixel of a specific window corresponding to the pixel;
s32, if the landform representation image is a depth degree or parallax image, graying; if the landform representation image is a DSM image, graying is not carried out;
and S33, coding the current gray-scale image or DSM image according to a binary coding rule, and outputting pixel codes of a specific window of the landform characterization image.
Preferably, in step S33, in encoding the pixels of the current gray map or the pixels of the DSM map, the current gray map is binary-encoded pixel by pixel or the elevation of the pixels of the current DSM map is binary-encoded pixel by pixel.
Preferably, in step S4, if the binary string of the current pixel code is consistent with the binary string of any one of the morphological templates, the current pixel code is successfully matched with the current morphological template; otherwise, the matching between the current pixel code and the current morphology template fails.
Preferably, in step S4, the specific way of performing the conformal weight filtering is as follows: if the landform characterization map is a parallax map or a depth map, performing conformal filtering based on color distance weight on the parallax map or the depth map; and if the topographic characterization map is a DSM map, performing conformal filtering based on the elevation distance weight on the DSM map.
Preferably, the computational expression of conformal filtering based on color distance weights is:
wherein Δ d represents a color distance, R c 、G c 、B c RGB color values, R, of the central pixel of a particular window representing a disparity map or a depth map, respectively n 、G n 、B n Representing a particular window of a disparity map or depth map, respectivelyThe RGB color values of the neighboring pixels corresponding to the central pixel, and the color weight calculation expression of the neighboring pixels corresponding to the central pixel of the specific window of the disparity map or the depth map are as follows:
where alpha represents a hyper-parameter of 0.5 and gamma represents a hyper-parameter of 60.
Preferably, the calculation expression of conformal filtering based on elevation distance weights is as follows:
wherein Δ d represents the elevation distance, depth c Representing the elevation value, depth, of the center pixel of a particular window of a DSM map n And representing the elevation value of the neighbor pixel corresponding to the center pixel of the specific window of the DSM diagram, wherein the calculation expression of the elevation weight of the neighbor pixel corresponding to the center pixel of the specific window of the DSM diagram is as follows:
where alpha represents a hyper-parameter of 0.5 and gamma represents a hyper-parameter of 60.
Preferably, in step S5, the processing on the filtered geomorphic representation map includes: if the landform representation image is a parallax image, checking the filtered parallax image to obtain a final parallax image; and if the topographic characterization map is a depth map or a DSM map, carrying out bilateral filtering on the filtered depth map or DSM map to obtain a final depth map or DSM map.
Compared with the prior art, the technical scheme of the invention has the beneficial effects that:
the invention provides a filtering method based on morphological template conformal in three-dimensional topographic reconstruction, which comprises the steps of firstly determining a target topographic object to be processed, obtaining a topographic representation of the target topographic object, then designing a plurality of morphological templates with the same specification based on regular shapes of angular points, aiming at containing the shapes of all regular angular points, facilitating the subsequent matching and retaining of the angular points, then detecting the edge to be processed and the angular point position in the topographic representation by using the designed morphological templates, judging whether pixel codes of a specific window of the topographic representation are matched with any one of the morphological templates, if the matching is successful, the specific window corresponding to the current pixel codes is the angular point, carrying out conformal weighted filtering processing on the current pixel codes, otherwise carrying out median filtering on the current pixel codes, ensuring the smoothness of the image at the edge, then processing the filtered topographic representation, reconstructing an angular point model of the target topographic object according to the final topographic representation, realizing the accurate identification and the retaining of the topographic object edge and the accurate restoration of the true contour of the topographic object.
Drawings
Fig. 1 is a schematic flowchart illustrating a filtering method based on morphological template conformal in three-dimensional landform reconstruction according to an embodiment of the present invention;
FIG. 2 is a graph comparing the disparity map results of the present invention with prior art disparity map results;
FIG. 3 shows a graph comparing depth map results of the present invention with prior art depth map results;
FIG. 4 shows a comparison graph of DSM plot results of the present invention and prior art DSM plot results;
FIG. 5 shows a layout of a morphological template of corner points;
FIG. 6 shows a layout of a morphological template of an edge;
FIG. 7 shows a binary-coded pixel window;
fig. 8 shows a schematic diagram of a method of median filtering.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the patent;
for better illustration of the present embodiment, some parts in the drawings may be omitted, enlarged or reduced, and do not represent actual sizes, and the description of the directions of the parts such as "upper" and "lower" is not limited to the patent;
it will be understood by those skilled in the art that certain well-known descriptions of the figures may be omitted;
the positional relationships depicted in the drawings are for illustrative purposes only and are not to be construed as limiting the present patent;
the technical solution of the present invention is further described with reference to the drawings and the embodiments.
Example 1
As shown in fig. 1, a filtering method based on morphological template conformal in three-dimensional landform reconstruction includes the following steps:
s1, determining a target ground object to be processed, and acquiring a landform representation map of the target ground object;
in step S1, the topographic characterization map is a depth map or a disparity map or a DSM map.
S2, determining the regular shape of the angular point, and constructing a plurality of morphological templates with the same window size based on the regular shape of the angular point;
s3, collecting a specific window of the landform representation image, preprocessing the specific window, and acquiring a pixel code of the specific window;
in step S3, the size of the specific windows for collecting the geomorphic representation maps is the same. .
S4, judging whether the pixel code obtained in the S3 is matched with any one morphological template, if so, taking a specific window corresponding to the current pixel code as an angular point, carrying out angle-preserving weight filtering on the current pixel code, and executing the step S5; otherwise, the specific window corresponding to the current pixel code is the edge, median filtering is carried out on the current pixel code, and the step S5 is executed;
s5, processing the filtered landform characteristic diagram to obtain a processed landform characteristic diagram;
in step S5, the processing of the filtered geomorphic representation map includes: if the landform representation image is a parallax image, checking the filtered parallax image in a mode of checking the consistency of the left parallax image and the right parallax image of the filtered parallax image to obtain a final parallax image; if the landform representation image is a depth image or a DSM image, carrying out bilateral filtering on the filtered depth image or the DSM image to obtain a final depth image or the DSM image, wherein the bilateral filtering is a nonlinear filtering method and is compromise processing combining spatial proximity and pixel value similarity of the image, and simultaneously considering spatial domain information and gray scale similarity to achieve the purpose of edge protection and denoising.
And S6, reconstructing a three-dimensional model of the target ground object based on the processed landform characterization map obtained in the S5.
In the embodiment, a target terrain to be processed is determined, a terrain characterization image of the target terrain is obtained, then a plurality of morphological templates with the same specification are designed based on the regular shapes of the corners, the purpose is to include the shapes of all regular corners, so that the subsequent corner matching and retention are facilitated, then the designed morphological templates are utilized to detect the edges to be processed and the corner positions in the terrain characterization image, whether pixel codes of a specific window of the terrain characterization image are matched with any one of the morphological templates is judged, if the matching is successful, the specific window corresponding to the current pixel codes is the corner, and the current pixel codes are subjected to angle-preserving weighted filtering treatment, otherwise, the current pixel codes are subjected to median filtering to ensure the smoothness of the image at the edge, the filtered terrain characterization image is processed, a three-dimensional model of the target terrain is reconstructed according to the final terrain characterization image, namely a final depth image or a parallax image or a DSM image, accurate recognition and corner edge retention and accurate restoration of the terrain true contour are realized; fig. 2 is a diagram showing a comparison between the parallax map result obtained in the present embodiment and the parallax map result in the prior art, and it is obvious from fig. 2 that the building edges obtained by the parallax map in the present embodiment are clearer and more conform to the actual scene; FIG. 3 is a graph comparing the depth map result of the present embodiment with the depth map result of the prior art, and it is apparent from FIG. 3 that the depth map result of the present embodiment is smoother in red house edge and clearer in outline; fig. 4 is a graph showing a comparison between the DSM drawing result of the present embodiment and the DSM drawing result of the prior art, and it is apparent from fig. 4 that the corner contour of the residential building is more clear.
Example 2
Referring to fig. 1, in step S2, the topographic characterization map is a depth map, a disparity map, or a DSM map, each morphological template corresponds to a geometric shape of a regular corner, and a pixel value of a color image or a pixel value of an elevation image in each morphological template is encoded into a template of a binary string, referring to fig. 5 and 6, morphological templates of 40 types of corners are designed, the morphological templates include regular shapes of all corners, and are encoded into 40 types of binary strings as a target feature matching object, and morphological templates of 9 types of edges are also designed, so that matching of corners and edges is performed on pixel encoding of an input depth map, disparity map, or DSM map, and the rule of binary encoding is: taking the difference value between the gray value of the central pixel point of the morphological template and the gray values of the peripheral pixel points as a coded binary template string, and if the difference value between the gray value of the central pixel point and the gray value of the peripheral pixel points is greater than a set depth threshold, coding the binary bit of the current pixel point to be 1; otherwise, the binary bit code of the current pixel point is 0.
Referring to fig. 7, taking a 3 × 3 window as an example, taking the difference between the gray value of the central pixel of the morphological template and the gray values of the 8 surrounding pixels as a coded binary template string, setting a depth threshold, and determining whether the difference between the gray value of the current pixel and the gray value of the surrounding pixels is greater than the set depth threshold, if so, the corresponding binary bit code is 1; otherwise, the corresponding binary bit code is 0, the gray value of the central pixel in fig. 4 is 10, the gray values of the surrounding 8 pixels are 34, 25, 12, 15, 46, 32, 11 and 25 from left to right and from top to bottom respectively, the difference values between the gray value of the central pixel and the gray values of the surrounding 8 pixels are 14, 15, 2, 5, 36, 22, 1 and 15 respectively, the depth threshold is 3, wherein the binary bit code whose difference value is greater than the depth threshold 3 is 1, the binary bit code whose difference value is less than the depth threshold 3 is 0, and the binary code pixel of the 3 × 3 window image is 11011101.
Example 3
Referring to fig. 1, in step S3, the specific process of the preprocessing is:
s31, taking the pixel of the landform representation image as a central pixel of a specific window corresponding to the pixel;
s32, if the landform representation image is a depth degree or parallax image, graying; if the landform representation map is a DSM map, graying is not carried out;
in step S32, the step of graying the depth map or the disparity map includes: first, an RGB image corresponding to the depth map or the disparity map is determined, and then the current RGB image is converted into a grayscale map.
And S33, encoding the current gray level image or DSM image according to the binary encoding rule, and outputting the pixel code of a specific window of the landform characterization image.
In step S33, the current gray map is binary coded pixel by pixel or the elevation of the pixel of the current DSM map is binary coded pixel by pixel during the coding of the pixel of the current gray map or the pixel of the DSM map.
In step S4, if the binary string of the current pixel code is consistent with the binary string of any one of the morphological templates, the matching between the current pixel code and the current morphological template is successful, that is, each binary string is consistent, for example, the binary string of the morphological template in fig. 7 is 11011101, and the binary string of the current pixel code needs to be the same as 11011101, which proves that the matching between the current pixel code and the binary string is successful; otherwise, the matching between the current pixel code and the current morphological template fails, and the median filtering of edge smoothing is needed to be carried out; the concrete way of carrying out the weight filtering of the angle is as follows: for the disparity map or the depth map, conformal filtering based on color distance weights needs to be performed, and the computational expression of the conformal filtering based on the color distance weights is as follows:
wherein Δ d represents a color distance, R c 、G c 、B c RGB color values, R, of a center pixel of a particular window representing a disparity map or a depth map, respectively n 、G n 、B n Representing RGB colour values of a neighbour pixel corresponding to a central pixel of a particular window of a disparity or depth map, respectively, a particular disparity or depth mapColor weights of neighbor pixels corresponding to a center pixel of a windowweightThe calculation expression is:
where alpha represents a hyper-parameter of 0.5 and gamma represents a hyper-parameter of 60.
Example 4
For the DSM chart, conformal filtering based on the elevation distance weight is required to be carried out, and the calculation expression of the conformal filtering based on the elevation distance weight is as follows:
wherein Δ d represents the elevation distance, depth c Elevation value, depth, representing a center pixel of a particular window of a DSM map n And representing the elevation value of the neighbor pixel corresponding to the center pixel of the specific window of the DSM diagram, wherein the calculation expression of the elevation weight of the neighbor pixel corresponding to the center pixel of the specific window of the DSM diagram is as follows:
where alpha represents a hyper-parameter of 0.5 and gamma represents a hyper-parameter of 60.
Referring to fig. 8, a method for performing median filtering with edge smoothing is to select a 3 × 3 pixel image window, the depth value or the elevation value of the center pixel of the left morphological template is 10, the depth values or the elevation values of the surrounding 8 pixel points are 6, 3, 2, 5, 4, and 6 from left to right, respectively, sort the depth values or the elevation values, 2, 3, 4, 5, 6, and 10, then take the median value 4 in the sequence, update the depth value or the elevation value of the center pixel of the left morphological template, then take the median value 4 in the sequence as the depth value or the elevation value to update the center pixel, the median filtering method in this embodiment is a nonlinear smoothing technique, which sets the gray value of each pixel point as the median value of all pixel points in a certain neighborhood window of the point, the median filtering is a nonlinear signal processing technique capable of effectively suppressing noise based on the ordering statistical theory, the median filtering is a two-dimensional sliding filtering method, which takes the median value of one point in the real point as the median value of the two-dimensional image data, and eliminates the two-dimensional noise of the two-dimensional sliding pixel data.
It should be understood that the above-described embodiments of the present invention are merely examples for clearly illustrating the present invention and are not intended to limit the embodiments of the present invention. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. This need not be, nor should it be exhaustive of all embodiments. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the claims of the present invention.
Claims (10)
1. A filtering method based on morphological template conformal in three-dimensional landform reconstruction is characterized by comprising the following steps:
s1, determining a target ground object to be processed, and acquiring a landform representation map of the target ground object;
s2, determining the regular shape of the angular point, and constructing a plurality of morphological templates with the same window size based on the regular shape of the angular point;
s3, collecting a specific window of the landform representation image, preprocessing the specific window, and acquiring a pixel code of the specific window;
s4, judging whether the pixel code obtained in the S3 is matched with any one morphological template, if so, taking a specific window corresponding to the current pixel code as an angular point, carrying out angle-preserving weight filtering on the current pixel code, and executing the step S5; otherwise, the specific window corresponding to the current pixel code is the edge, median filtering is carried out on the current pixel code, and the step S5 is executed;
s5, processing the filtered landform characteristic diagram to obtain a processed landform characteristic diagram;
and S6, reconstructing a three-dimensional model of the target ground object based on the processed landform characterization map obtained in the S5.
2. The filtering method based on morphology template conformal in three-dimensional landform reconstruction according to claim 1, wherein the landform characterization map is a depth map or a disparity map or a DSM map, in step S2, each morphology template corresponds to a regular shape of a corner point, the pixel coding of the color image and the pixel coding of the elevation image in each morphology template are templates of a binary string, and the rules of binary coding are as follows: taking the difference value between the gray value of the central pixel point of the morphological template and the gray values of the peripheral pixel points as a coded binary template string, and if the difference value between the gray value of the central pixel point and the gray value of the peripheral pixel points is greater than a set depth threshold, coding the binary bit of the current pixel point to be 1; otherwise, the binary bit code of the current pixel point is 0.
3. The filtering method based on morphology template conformal in three-dimensional landform reconstruction according to claim 1, wherein in step S3, the size of the specific windows for collecting the landform characterization maps is the same.
4. The filtering method based on morphological template conformal in three-dimensional landform reconstruction as claimed in claim 2, wherein in step S3, the specific process of preprocessing is:
s31, taking the pixel of the landform representation image as a central pixel of a specific window corresponding to the pixel;
s32, if the landform representation image is a depth degree or parallax image, graying is carried out; if the landform representation image is a DSM image, graying is not carried out;
and S33, coding the current gray-scale image or DSM image according to a binary coding rule, and outputting pixel codes of a specific window of the landform characterization image.
5. The filtering method based on morphological template conformal in three-dimensional landform reconstruction according to claim 4, wherein in step S33, in the process of encoding the pixels of the current gray-scale map or the pixels of the DSM map, the current gray-scale map is binary encoded pixel by pixel or the elevations of the pixels of the current DSM map are binary encoded pixel by pixel.
6. The filtering method based on morphology template conformal in three-dimensional landform reconstruction according to claim 5, characterized in that, in step S4, if the binary string of the current pixel code is consistent with the binary string of any one morphology template, the matching between the current pixel code and the current morphology template is successful; otherwise, the matching between the current pixel code and the current morphology template fails.
7. The filtering method based on morphological template conformal in three-dimensional landform reconstruction as claimed in claim 6, wherein in step S4, the specific way of performing the weight filtering of the conformal is: if the landform characterization map is a parallax map or a depth map, performing conformal filtering based on color distance weight on the parallax map or the depth map; and if the topographic characterization map is a DSM map, performing conformal filtering based on the elevation distance weight on the DSM map.
8. The filtering method based on the morphological template conformal in the three-dimensional landform reconstruction of claim 7, wherein the computational expression of conformal filtering based on the color distance weight is as follows:
wherein Δ d represents a color distance, R c 、G c 、B c RGB color values, R, of a center pixel of a particular window representing a disparity map or a depth map, respectively n 、G n 、B n Respectively representing RGB color values of a neighbor pixel corresponding to a central pixel of a particular window of a disparity map or a depth map, color weights of a neighbor pixel corresponding to a central pixel of a particular window of a disparity map or a depth mapHeavy weight meter Calculating tableThe expression is as follows:
where alpha represents a hyper-parameter of 0.5 and gamma represents a hyper-parameter of 60.
9. The filtering method based on morphological template conformal in three-dimensional geomorphology reconstruction as claimed in claim 7, wherein the calculation expression of conformal filtering based on elevation distance weight is:
wherein Δ d represents the elevation distance, depth c Representing the elevation value, depth, of the center pixel of a particular window of a DSM map n And representing the elevation value of the neighbor pixel corresponding to the center pixel of the specific window of the DSM diagram, wherein the calculation expression of the elevation weight of the neighbor pixel corresponding to the center pixel of the specific window of the DSM diagram is as follows:
where alpha represents a hyper-parameter of 0.5 and gamma represents a hyper-parameter of 60.
10. The filtering method based on morphology template conformal in three-dimensional landform reconstruction according to claim 8, wherein in step S5, the processing of the filtered landform characterization map comprises: if the landform representation image is a parallax image, checking the filtered parallax image to obtain a final parallax image; and if the topographic characterization map is a depth map or a DSM map, carrying out bilateral filtering on the filtered depth map or DSM map to obtain a final depth map or DSM map.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211167831.9A CN115661382A (en) | 2022-09-23 | 2022-09-23 | Filtering method based on morphological template angle protection in three-dimensional landform reconstruction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211167831.9A CN115661382A (en) | 2022-09-23 | 2022-09-23 | Filtering method based on morphological template angle protection in three-dimensional landform reconstruction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115661382A true CN115661382A (en) | 2023-01-31 |
Family
ID=84985396
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211167831.9A Pending CN115661382A (en) | 2022-09-23 | 2022-09-23 | Filtering method based on morphological template angle protection in three-dimensional landform reconstruction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115661382A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115965855A (en) * | 2023-02-14 | 2023-04-14 | 成都睿芯行科技有限公司 | Method and device for improving tray identification precision |
-
2022
- 2022-09-23 CN CN202211167831.9A patent/CN115661382A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115965855A (en) * | 2023-02-14 | 2023-04-14 | 成都睿芯行科技有限公司 | Method and device for improving tray identification precision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108921799B (en) | Remote sensing image thin cloud removing method based on multi-scale collaborative learning convolutional neural network | |
| CN108986119B (en) | Image segmentation method and device, computer equipment and readable storage medium | |
| CN110866924B (en) | Line structured light center line extraction method and storage medium | |
| CN109242791B (en) | Batch repair method for damaged plant leaves | |
| CN110310320B (en) | Binocular vision matching cost aggregation optimization method | |
| CN110348263B (en) | Two-dimensional random code image identification and extraction method based on image identification | |
| US20100008576A1 (en) | System and method for segmentation of an image into tuned multi-scaled regions | |
| CN103530889A (en) | Method for decoding coding mark points of measuring stick based on skeleton extraction | |
| CN104517110A (en) | Binarization method and system of two-dimensional code image | |
| CN110544300B (en) | Method for automatically generating three-dimensional model based on two-dimensional hand-drawn image characteristics | |
| CN109961416B (en) | Business license information extraction method based on morphological gradient multi-scale fusion | |
| CN107122713B (en) | Analog property detection method based on deep learning | |
| CN109978848A (en) | Method based on hard exudate in multiple light courcess color constancy model inspection eye fundus image | |
| CN109754440A (en) | A kind of shadow region detection method based on full convolutional network and average drifting | |
| AU2020103470A4 (en) | Shadow Detection for High-resolution Orthorectificed Imagery through Multi-level Integral Relaxation Matching Driven by Artificial Shadows | |
| CN110751680A (en) | Image processing method with fast alignment algorithm | |
| CN115661382A (en) | Filtering method based on morphological template angle protection in three-dimensional landform reconstruction | |
| CN112819688A (en) | Conversion method and system for converting SAR (synthetic aperture radar) image into optical image | |
| CN116721121A (en) | Plant phenotype color image feature extraction method | |
| CN115861409A (en) | Soybean leaf area measuring and calculating method, system, computer equipment and storage medium | |
| CN113920147A (en) | Remote sensing image building extraction method and device based on deep learning | |
| CN107358635B (en) | Color morphological image processing method based on fuzzy similarity | |
| CN115841618A (en) | Remote sensing image coastline extraction method based on clustering and edge detection | |
| CN112329674A (en) | Frozen lake detection method and device based on multi-texture feature fusion | |
| Jain et al. | Shadow removal for umbrageous information recovery in aerial images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |