CN115830044A - Image segmentation method and device, electronic equipment and storage medium - Google Patents
Image segmentation method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN115830044A CN115830044A CN202310032599.6A CN202310032599A CN115830044A CN 115830044 A CN115830044 A CN 115830044A CN 202310032599 A CN202310032599 A CN 202310032599A CN 115830044 A CN115830044 A CN 115830044A
- Authority
- CN
- China
- Prior art keywords
- image
- original
- normal vector
- optimal
- target
- 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.)
- Granted
Links
- 238000003709 image segmentation Methods 0.000 title claims abstract description 40
- 238000000034 method Methods 0.000 title claims abstract description 26
- 239000013598 vector Substances 0.000 claims abstract description 83
- 238000004590 computer program Methods 0.000 claims description 12
- 238000005516 engineering process Methods 0.000 claims description 10
- 230000000694 effects Effects 0.000 abstract description 8
- 230000011218 segmentation Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 3
- 238000007689 inspection Methods 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- RTAQQCXQSZGOHL-UHFFFAOYSA-N Titanium Chemical compound [Ti] RTAQQCXQSZGOHL-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Images
Landscapes
- Image Processing (AREA)
Abstract
The invention discloses an image segmentation method, an image segmentation device, electronic equipment and a storage medium, wherein the method comprises the following steps: acquiring all original scanning images of a target area; determining an optimal image in the original scanning image according to the normal vectors of the original scanning image and the target image; establishing a blank canvas according to the size and the resolution ratio of the optimal image; and filling each pixel point of the optimal image into a blank canvas to obtain a segmented image. According to the embodiment of the invention, the optimal image close to the target image is determined, the blank canvas is established, and the pixel points of the optimal image are correspondingly filled in the blank canvas, so that the image can be finely divided, and the image dividing effect can be effectively improved.
Description
Technical Field
The present invention relates to the field of image processing technologies, and in particular, to an image segmentation method and apparatus, an electronic device, and a storage medium.
Background
With the development of scientific technology, the application of image segmentation technology in practical work is more and more extensive. For example, in the field of vehicle technology, a dangerous area on a driving road section can be extracted through image segmentation so as to achieve the effect of safe driving.
Currently, the conventional image segmentation method generally performs panorama segmentation or part segmentation on an image based on an image segmentation model, and performs fusion processing on the segmented image. However, the existing image segmentation method cannot finely segment the image, so that the image segmentation effect is poor.
Disclosure of Invention
The invention provides an image segmentation method, an image segmentation device and a storage medium, which are used for solving the technical problem that the image segmentation effect is poor due to the fact that the existing image segmentation method cannot finely segment an image.
One embodiment of the present invention provides an image segmentation method, including:
acquiring all original scanning images of a target area;
determining an optimal image in the original scanning image according to the normal vectors of the original scanning image and the target image;
establishing a blank canvas according to the size and the resolution of the optimal image;
and filling each pixel point of the optimal image into the blank canvas to obtain a segmented image.
Further, acquiring all original scanning images of the target area, including;
based on the holographic image technology, all original scanning images of the target area are obtained.
Further, the determining an optimal image in the original scanned image according to the normal vectors of the original scanned image and the target image includes:
determining a normal vector of the target image according to the target characteristics;
calculating to obtain a current normal vector of the original scanning image according to the initial normal vector and the attitude angle of the original scanning image;
and calculating an included angle between the normal vector of the target image and the current normal vector of the original scanning image, and determining the original scanning image corresponding to the minimum included angle as the optimal image.
Further, the calculating a current normal vector of the original scanned image according to the initial normal vector and the attitude angle of the original scanned image includes:
calculating to obtain the current normal vector of the scanned image by adopting the following formula:
wherein [ x0, y0, z0]The initial normal vector of the original scanning image; [ x, y, z ]]The current normal vector of the original scanning image; alpha, beta and gamma are attitude angles of the original image.
Further, the calculating an included angle between the normal vector of the target image and the current normal vector of the original scanned image includes:
calculating an included angle between the normal vector of the target image and the current normal vector of the original scanning image by adopting the following formula:
wherein [ a, b, c ] is the normal vector of the target image.
Further, the creating a blank canvas according to the size and resolution of the best image includes:
taking the product of the length of the long edge of the optimal image and the resolution as the length of the long edge of the blank canvas;
taking the product of the length of the short edge of the optimal image and the resolution as the length of the short edge of the blank canvas;
and setting the length and width pixels of the blank canvas as the number of grid rows and columns corresponding to the target image.
Further, the filling each pixel point of the optimal image into the blank canvas to obtain a segmented image includes:
calculating to obtain a geographic coordinate corresponding to each pixel point of the optimal image according to the resolution and the normal vector of the optimal image;
calculating to obtain the original image coordinates of the pixel points corresponding to the geographical coordinates according to the geographical coordinates;
and filling the pixel points to the blank canvas according to the original image coordinates.
An embodiment of the present invention provides an image segmentation apparatus including:
the scanning image acquisition module is used for acquiring all original scanning images of the target area;
the optimal image determining module is used for determining an optimal image in the original scanning image according to the normal vectors of the original scanning image and the target image;
the blank canvas establishing module is used for establishing a blank canvas according to the size and the resolution ratio of the optimal image;
and the image pixel filling module is used for filling each pixel point of the optimal image into the blank canvas to obtain a segmented image.
An embodiment of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the image segmentation method as described above when executing the computer program.
An embodiment of the present invention provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program runs, the apparatus on which the computer-readable storage medium is located is controlled to execute the image segmentation method as described above.
According to the embodiment of the invention, the optimal image close to the target image is determined, the blank canvas is established, and the pixel points of the optimal image are correspondingly filled in the blank canvas, so that the image can be finely divided, and the image dividing effect can be effectively improved.
Drawings
FIG. 1 is a flowchart illustrating an image segmentation method according to an embodiment of the present invention;
fig. 2 is a schematic view of a panoramic point cloud overlay image according to an embodiment of the present invention;
FIG. 3 is a diagram of a newly created blank canvas according to an embodiment of the present invention;
FIG. 4 is a diagram illustrating automatic segmentation parameter setting provided by an embodiment of the present invention;
FIG. 5 is a schematic diagram of a characteristic line selection provided by an embodiment of the present invention;
FIG. 6 is a schematic diagram of a segmented image provided by an embodiment of the present invention;
fig. 7 is a schematic structural diagram of an image segmentation apparatus according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In the description of the present application, it is to be understood that the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or as implying that the number of indicated technical features is indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present application, "a plurality" means two or more unless otherwise specified.
In the description of the present application, it is to be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meaning of the above terms in the present application can be understood in a specific case by those of ordinary skill in the art.
Referring to fig. 1, an embodiment of the present invention provides an image segmentation method, including:
s1, acquiring all original scanning images of a target area;
in the embodiment of the invention, the target area can be set according to actual needs, and the target area can be a certain road section; the original scan image may be a vehicle-mounted scan image. The embodiment of the invention can synchronously acquire the image data and the point cloud data during vehicle-mounted scanning based on the holographic image technology, wherein the image data has accurate position and attitude information and can form a corresponding relation with the point cloud data.
S2, determining an optimal image in the original scanning image according to the normal vectors of the original scanning image and the target image;
in the embodiment of the present invention, the target image may be an image including a preset target, and the preset target may be a person, a vehicle, a road, a building, and the like.
Because the two adjacent original scanning images have a certain overlapping rate and some road sections can be scanned back and forth, the same point cloud data or the same characteristic object may correspond to a plurality of images. In order to enhance the image segmentation effect, the embodiment of the invention determines the best image closest to the target image in all the original scanning images.
S3, establishing a blank canvas according to the size and the resolution of the optimal image;
in the embodiment of the invention, the newly established blank canvas needs to be related to the size and the resolution of the optimal image, and the pixel points of the optimal image are filled in the blank canvas to realize the segmentation of the image.
And S4, filling each pixel point of the optimal image into a blank canvas to obtain a segmented image.
In the embodiment of the invention, each pixel point of the optimal image is filled into the corresponding position of the blank canvas, and after all the pixel points of the optimal image are filled, a new image is generated on the blank canvas, wherein the image is obtained by segmenting the optimal image.
According to the embodiment of the invention, the optimal image close to the target image is determined, the blank canvas is established, and the pixel points of the optimal image are correspondingly filled in the blank canvas, so that the image can be finely divided, the image dividing effect can be effectively improved, and the image dividing result can be widely applied to production work such as quality inspection tour, component investigation, three-dimensional modeling and the like.
In one embodiment, all raw scan images of the target area are acquired, including;
based on the holographic image technology, all original scanning images of the target area are obtained.
In the embodiment of the invention, the original scanning image can be obtained by scanning the road section where the vehicle runs through the holographic image technology.
The embodiment of the invention can also adopt a vehicle-mounted laser scanning system to acquire point cloud data, and solve the acquired point cloud data and extract the object vector.
In the embodiment of the invention, the camera and the laser scanner can be rigidly connected, the line eccentricity and the angle eccentricity between the camera and the laser scanner are calibrated, and the image acquired by the camera and the point cloud data acquired by the laser scanner are registered to obtain the superposed original scanning image shown in fig. 2.
In one embodiment, the step S2 of determining the best image in the original scanned image according to the normal vectors of the original scanned image and the target image may further include the following sub-steps:
s21, determining a normal vector of the target image according to the target characteristics;
in the embodiment of the present invention, the normal vector [ a, b, c ] of the target image may be determined according to the actual position of the target feature, or the size of the target image may be determined according to the size of the target feature.
S22, calculating to obtain a current normal vector of the original scanning image according to the initial normal vector and the attitude angle of the original scanning image;
in the embodiment of the invention, when the original scanning images are obtained by the camera, the attitude file of the camera stores the attitude information corresponding to each original scanning image, and the attitude information comprises data such as the position, the course angle, the pitch angle, the roll angle, the attitude angle and the like of the camera at the exposure time. According to the embodiment of the invention, the current normal vector of each original scanning image can be obtained by calculation according to the initial normal vector and the attitude angle of the original scanning image.
And S23, calculating an included angle between the normal vector of the target image and the current normal vector of the original scanning image, and determining the original scanning image corresponding to the minimum included angle as the optimal image.
In the embodiment of the invention, the original scanning image closest to the target image can be determined as the optimal image according to the normal vector included angle between the two images.
In one embodiment, calculating the current normal vector of the original scanned image according to the initial normal vector and the attitude angle of the original scanned image comprises:
calculating the current normal vector of the scanned image by adopting the following formula:
wherein [ x0, y0, z0] is the initial normal vector of the original scanning image; [ x, y, z ] is the current normal vector of the original scanned image; alpha, beta and gamma are attitude angles of the original image.
In the embodiment of the present invention, before the current normal vector of the original scanned image is obtained by performing the calculation, the original scanned image obtained by scanning may be further screened to obtain an original scanned image that meets the preset condition, which may specifically be:
according to the central position of the geometric figure and the position of the exposure point, calculating to obtain an original scanned image with the distance from the target image within a preset threshold range according to a distance formula between the two points, and then calculating the current normal vector of the original scanned image.
In one embodiment, calculating an angle between a normal vector of the target image and a current normal vector of the original scanned image comprises:
calculating an included angle between the normal vector of the target image and the current normal vector of the original scanning image by adopting the following formula:
wherein [ a, b, c ] is the normal vector of the target image.
In the embodiment of the invention, the normal vector included angle between each original scanning image and the target image is obtained through calculation, and the original scanning image corresponding to the minimum included angle is selected as the optimal image in all the normal vector included angles.
In an embodiment, the step S3 of creating a blank canvas according to the size and resolution of the best image may further include the following sub-steps:
s31, taking the product of the length of the long edge of the optimal image and the resolution as the length of the long edge of the blank canvas;
s32, taking the product of the short edge length and the resolution of the optimal image as the short edge length of the blank canvas;
referring to fig. 3, in the embodiment of the invention, when the size of the best image is mxn (m) and the resolution is P, the size of the blank canvas is Pm × Pn (pixel).
And S33, setting the length and width pixels of the blank canvas as the grid row number corresponding to the target image.
According to the embodiment of the invention, the size of the blank canvas and the number of the grid rows and columns corresponding to the target image are set according to the size and the resolution ratio of the optimal image, and when the pixel points of the optimal image are sequentially filled into the blank canvas, the pixel points of the optimal image can be completely filled into the newly-built blank canvas, so that a complete segmentation image can be formed.
In an embodiment, the step S4 of filling each pixel point of the optimal image into a blank canvas to obtain a segmented image may further include the following substeps:
s41, calculating to obtain a geographic coordinate corresponding to each pixel point of the optimal image according to the resolution and the normal vector of the optimal image;
s42, calculating to obtain the original image coordinates of the corresponding pixel points according to the geographic coordinates;
in the embodiment of the invention, the original image coordinates corresponding to the pixel points can be calculated by adopting a collinear equation:
wherein, (x 0, y0, z 0) is the projection center coordinate of the original image; (xA, yA, zA) is the geographical coordinate corresponding to the pixel point in the best image; f is the main distance of the original image; wherein:
And S43, filling the pixel points to the blank canvas according to the original image coordinates.
In the embodiment of the invention, after the pixel points are filled in the blank canvas, the RGB values of the pixel points are also assigned, so that the image obtained by segmentation is more accurate.
Referring to fig. 4, in an embodiment, a data processing program may be programmed according to a correlation algorithm, and the automatic segmentation of the image corresponding to the feature information may be implemented according to the data processing program, and the generated segmented image and the feature data may be stored synchronously.
Referring to fig. 5, in an embodiment of the present invention, based on the vehicle-mounted laser point cloud and the image data, the corresponding image segmentation is performed on the entity object extracted from the point cloud, the hardware environment is configured as NVIDIA Titan Xp, the video memory of the video card is 12G, the video memory speed is 11.4Gbps, the bit width is 384 bits, and the bandwidth is 547.7GB/s. In the range of 450KM2 in Guangdong province, the image segmentation method provided by the embodiment of the invention is adopted for image segmentation. The total number of the feature information is 718, and 36 error data are deleted through man-machine interaction, so that 712 effective data are obtained finally. Wherein, the dotted line can be used as the selected characteristic line in the figure.
Please refer to fig. 6, which is a diagram illustrating an image segmentation result according to an embodiment of the present invention.
The embodiment of the invention has the following beneficial effects:
according to the embodiment of the invention, the optimal image close to the target image is determined, the blank canvas is established, and the pixel points of the optimal image are correspondingly filled in the blank canvas, so that the image can be finely divided, the image dividing effect can be effectively improved, and the image dividing result can be widely applied to production work such as quality inspection tour, component investigation, three-dimensional modeling and the like.
Referring to fig. 7, based on the same inventive concept as the above embodiment, an embodiment of the present invention provides an image segmentation apparatus, including:
a scanned image obtaining module 10, configured to obtain all original scanned images of a target area;
an optimal image determining module 20, configured to determine an optimal image in the original scanned image according to normal vectors of the original scanned image and the target image;
a blank canvas establishing module 30, configured to establish a blank canvas according to the size and resolution of the optimal image;
and the image pixel filling module 40 is configured to fill each pixel point of the optimal image into a blank canvas to obtain a segmented image.
In one embodiment, the scan image acquisition module 10 is further configured to:
based on the holographic image technology, all original scanning images of the target area are obtained.
In one embodiment, the best image determination module 20 is further configured to:
determining a normal vector of a target image according to the target characteristics;
calculating to obtain a current normal vector of the original scanning image according to the initial normal vector and the attitude angle of the original scanning image;
and calculating an included angle between the normal vector of the target image and the current normal vector of the original scanning image, and determining the original scanning image corresponding to the minimum included angle as the optimal image.
In one embodiment, calculating the current normal vector of the original scanned image according to the initial normal vector and the attitude angle of the original scanned image comprises:
calculating to obtain the current normal vector of the scanned image by adopting the following formula:
wherein [ x0, y0, z0] is the initial normal vector of the original scanning image; [ x, y, z ] is the current normal vector of the original scanned image; alpha, beta and gamma are attitude angles of the original image.
In one embodiment, calculating the angle between the normal vector of the target image and the current normal vector of the original scanned image comprises:
calculating an included angle between the normal vector of the target image and the current normal vector of the original scanning image by adopting the following formula:
wherein [ a, b, c ] is the normal vector of the target image.
In one embodiment, the blank canvas creation module 30 is further operable to:
taking the product of the length of the long edge of the optimal image and the resolution as the length of the long edge of the blank canvas;
taking the product of the length of the short edge of the optimal image and the resolution as the length of the short edge of the blank canvas;
and setting the length and width pixels of the blank canvas as the number of grid rows and columns corresponding to the target image.
In one embodiment, the image pixel filling module 40 is further configured to:
calculating to obtain a geographic coordinate corresponding to each pixel point of the optimal image according to the resolution and the normal vector of the optimal image;
calculating according to the geographic coordinates to obtain original image coordinates of pixel points corresponding to the geographic coordinates;
and filling the pixel points on the blank canvas according to the original image coordinates.
An embodiment of the present invention provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the image segmentation method as described above is implemented.
An embodiment of the present invention provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program runs, a device in which the computer-readable storage medium is located is controlled to execute the image segmentation method as described above.
The foregoing is a preferred embodiment of the present invention, and it should be noted that it would be apparent to those skilled in the art that various modifications and enhancements can be made without departing from the principles of the invention, and such modifications and enhancements are also considered to be within the scope of the invention.
Claims (10)
1. An image segmentation method, comprising:
acquiring all original scanning images of a target area;
determining an optimal image in the original scanning image according to the normal vectors of the original scanning image and the target image;
establishing a blank canvas according to the size and the resolution ratio of the optimal image;
and filling each pixel point of the optimal image into the blank canvas to obtain a segmented image.
2. The image segmentation method according to claim 1, wherein the obtaining of all original scan images of the target region comprises;
based on the holographic image technology, all original scanning images of the target area are obtained.
3. The method of image segmentation according to claim 1, wherein the determining an optimal image in the original scanned images according to the normal vectors of the original scanned images and the target image comprises:
determining a normal vector of the target image according to the target characteristics;
calculating to obtain a current normal vector of the original scanning image according to the initial normal vector and the attitude angle of the original scanning image;
and calculating an included angle between the normal vector of the target image and the current normal vector of the original scanning image, and determining the original scanning image corresponding to the minimum included angle as the optimal image.
4. The image segmentation method according to claim 3, wherein the calculating a current normal vector of the original scanned image according to the initial normal vector and the attitude angle of the original scanned image comprises:
calculating to obtain the current normal vector of the scanned image by adopting the following formula:
wherein [ x0, y0, z0] is the initial normal vector of the original scanning image; [ x, y, z ] is the current normal vector of the original scanned image; alpha, beta and gamma are attitude angles of the original image.
5. The image segmentation method of claim 3, wherein the calculating an angle between the normal vector of the target image and the current normal vector of the original scanned image comprises:
calculating an included angle between the normal vector of the target image and the current normal vector of the original scanning image by adopting the following formula:
wherein [ a, b, c ] is the normal vector of the target image.
6. The image segmentation method of claim 1, wherein the creating a blank canvas according to the size and resolution of the best image comprises:
taking the product of the length of the long edge of the optimal image and the resolution as the length of the long edge of the blank canvas;
taking the product of the length of the short edge of the optimal image and the resolution as the length of the short edge of the blank canvas;
and setting the length and width pixels of the blank canvas as the number of grid rows and columns corresponding to the target image.
7. The image segmentation method of claim 1, wherein the filling of each pixel of the optimal image into the blank canvas to obtain a segmented image comprises:
calculating to obtain a geographic coordinate corresponding to each pixel point of the optimal image according to the resolution and the normal vector of the optimal image;
calculating to obtain the original image coordinates of the pixel points corresponding to the geographical coordinates according to the geographical coordinates;
and filling the pixel points to the blank canvas according to the original image coordinates.
8. An image segmentation apparatus, comprising:
the scanning image acquisition module is used for acquiring all original scanning images of the target area;
the optimal image determining module is used for determining an optimal image in the original scanning image according to the normal vectors of the original scanning image and the target image;
the blank canvas establishing module is used for establishing a blank canvas according to the size and the resolution ratio of the optimal image;
and the image pixel filling module is used for filling each pixel point of the optimal image into the blank canvas to obtain a segmented image.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the image segmentation method according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, comprising a stored computer program, wherein when the computer program is run, the computer-readable storage medium is controlled by an apparatus to perform the image segmentation method according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310032599.6A CN115830044B (en) | 2023-01-10 | 2023-01-10 | Image segmentation method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310032599.6A CN115830044B (en) | 2023-01-10 | 2023-01-10 | Image segmentation method and device, electronic equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115830044A true CN115830044A (en) | 2023-03-21 |
CN115830044B CN115830044B (en) | 2024-04-05 |
Family
ID=85520528
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310032599.6A Active CN115830044B (en) | 2023-01-10 | 2023-01-10 | Image segmentation method and device, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115830044B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111914781A (en) * | 2020-08-10 | 2020-11-10 | 杭州海康威视数字技术股份有限公司 | Method and device for processing face image |
CN112258386A (en) * | 2020-10-26 | 2021-01-22 | 中国科学院微电子研究所 | Image deformation acceleration processing method and device, electronic equipment and readable storage medium |
CN113034570A (en) * | 2021-03-09 | 2021-06-25 | 北京字跳网络技术有限公司 | Image processing method and device and electronic equipment |
CN115100151A (en) * | 2022-06-27 | 2022-09-23 | 武汉兰丁智能医学股份有限公司 | Result-oriented cell image high-definition identification marking method |
-
2023
- 2023-01-10 CN CN202310032599.6A patent/CN115830044B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111914781A (en) * | 2020-08-10 | 2020-11-10 | 杭州海康威视数字技术股份有限公司 | Method and device for processing face image |
CN112258386A (en) * | 2020-10-26 | 2021-01-22 | 中国科学院微电子研究所 | Image deformation acceleration processing method and device, electronic equipment and readable storage medium |
CN113034570A (en) * | 2021-03-09 | 2021-06-25 | 北京字跳网络技术有限公司 | Image processing method and device and electronic equipment |
CN115100151A (en) * | 2022-06-27 | 2022-09-23 | 武汉兰丁智能医学股份有限公司 | Result-oriented cell image high-definition identification marking method |
Non-Patent Citations (1)
Title |
---|
王学: "基于多视角影像的三维模型纹理映射研究", no. 03, pages 32 - 33 * |
Also Published As
Publication number | Publication date |
---|---|
CN115830044B (en) | 2024-04-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20240153143A1 (en) | Multi view camera registration | |
CN110211043B (en) | Registration method based on grid optimization for panoramic image stitching | |
CN111815707A (en) | Point cloud determining method, point cloud screening device and computer equipment | |
WO2012126500A1 (en) | 3d streets | |
CN116778094B (en) | Building deformation monitoring method and device based on optimal viewing angle shooting | |
CN112367474A (en) | Self-adaptive light field imaging method, device and equipment | |
CN113345084B (en) | Three-dimensional modeling system and three-dimensional modeling method | |
CN111445513B (en) | Plant canopy volume acquisition method and device based on depth image, computer equipment and storage medium | |
CN111009011A (en) | Method, device, system and storage medium for predicting vehicle direction angle | |
CN114372992A (en) | Edge corner point detection four-eye vision algorithm based on moving platform | |
CN117336620B (en) | Adaptive video stitching method and system based on deep learning | |
CN117274514A (en) | Remote sensing image generation method and device based on ground-air visual angle geometric transformation | |
CN115830044B (en) | Image segmentation method and device, electronic equipment and storage medium | |
CN116092035A (en) | Lane line detection method, lane line detection device, computer equipment and storage medium | |
CN113077504B (en) | Large scene depth map generation method based on multi-granularity feature matching | |
CN111982015B (en) | Method for monitoring structure geometry | |
CN114998496A (en) | Orthoimage rapid generation method based on scene aerial photography image and sparse point cloud | |
CN110223250B (en) | SAR geometric correction method based on homography transformation | |
CN113379899A (en) | Automatic extraction method for regional images of construction engineering working face | |
Brunken et al. | Incorporating Plane-Sweep in Convolutional Neural Network Stereo Imaging for Road Surface Reconstruction. | |
CN113593026B (en) | Lane line labeling auxiliary map generation method, device and computer equipment | |
CN116538953B (en) | Intelligent detection method and system for elevation targets and readable storage medium | |
US20240242318A1 (en) | Face deformation compensating method for face depth image, imaging device, and storage medium | |
CN116485636B (en) | Point cloud elevation imaging method, system and readable storage medium | |
CN116958449B (en) | Urban scene three-dimensional modeling method and device and electronic equipment |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |