CN117171384B - High-precision texture path retrieval method, device, computer equipment and storage medium - Google Patents
High-precision texture path retrieval method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN117171384B CN117171384B CN202311442825.4A CN202311442825A CN117171384B CN 117171384 B CN117171384 B CN 117171384B CN 202311442825 A CN202311442825 A CN 202311442825A CN 117171384 B CN117171384 B CN 117171384B
- Authority
- CN
- China
- Prior art keywords
- texture
- diagram
- information
- skeleton
- graph
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000010586 diagram Methods 0.000 claims abstract description 90
- 230000000877 morphologic effect Effects 0.000 claims abstract description 38
- 238000013507 mapping Methods 0.000 claims abstract description 31
- 239000003086 colorant Substances 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 238000005192 partition Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 abstract description 10
- 238000009966 trimming Methods 0.000 abstract description 4
- 210000004209 hair Anatomy 0.000 description 11
- 238000004590 computer program Methods 0.000 description 4
- 238000012216 screening Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 210000003462 vein Anatomy 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a high-precision texture path retrieval method, a device, computer equipment and a storage medium, wherein the method comprises the following steps: acquiring a morphological gradient pointing diagram and a projection mapping diagram of cluster color information; taking the morphological gradient pointing diagram as a guide, traversing the region of interest according to the projection mapping diagram to obtain a texture skeleton diagram; and further filling the texture skeleton diagram to generate a foreground texture interval so as to construct and obtain a texture path diagram. The texture path diagram constructed by the texture path search comprises texture walking venation, the texture extension area also has texture intensity and texture density, and the texture path diagram has great help to image processing such as follow-up portrait matting, edge trimming, problem judgment and the like.
Description
Technical Field
The invention relates to a high-precision texture path retrieval method, a high-precision texture path retrieval device, computer equipment and a storage medium, and belongs to the field of image processing.
Background
Constructing image texture information is always a great difficulty in image processing, especially for the acquisition and processing of portrait images. The human figure has a large number of dense texture areas, hair, clothes and even complex backgrounds, texture information under the environments is very difficult to extract and display, dominant regular and circulated textures are lacking, abstract processing is very difficult, certain difficulties are caused for subsequent human figure matting, edge trimming and problem judgment, and an accurate searching method is needed for the image textures at the moment, so that the image textures can be accurately identified and clearly presented.
Disclosure of Invention
In view of the above, the present invention provides a high-precision texture path searching method, apparatus, computer device and storage medium, wherein a texture path diagram constructed by texture path searching includes texture walking veins, texture extension areas, and texture intensity, which greatly help subsequent portrait matting, edge trimming, problem determination image processing.
A first object of the present invention is to provide a high-precision texture path retrieval method.
A second object of the present invention is to provide a high-precision texture path retrieval apparatus.
A third object of the present invention is to provide a computer device.
A fourth object of the present invention is to provide a computer-readable storage medium.
The first object of the present invention can be achieved by adopting the following technical scheme:
a method of high precision texture path retrieval, the method comprising:
acquiring a morphological gradient pointing diagram and a projection mapping diagram of cluster color information;
taking the morphological gradient pointing diagram as a guide, traversing the region of interest according to the projection mapping diagram to obtain a texture skeleton diagram;
and further filling the texture skeleton diagram to generate a foreground texture interval so as to construct and obtain a texture path diagram.
Further, the obtaining a morphological gradient map specifically includes:
calculating gray level according to the appointed area based on the image gray information, and constructing a trend graph similar to a mountain range through gradient levels of surrounding information of each pixel point;
digitizing the trend graph into a cost graph;
images with values between 0 and 255 are classified according to the cost chart.
Further, the obtaining the projection map of the cluster color information specifically includes:
acquiring basic texture information as a skeleton, and counting color values of an area to be measured and calculated to enable the counted color space to fall on a cubic space of 255 x 255;
abstracting the cube space into a two-dimensional image, and gathering the colors of all the partitions into a plurality of blocks to obtain a cluster map;
dividing the clustering blocks of the clustering graph by a clustering algorithm, and calculating a block data center;
measuring main pixel intervals of each clustering block and main body values of block colors by taking the block data center as a reference point;
and (3) taking the main body numerical value as a standard, establishing a new color space coordinate, mapping the colors of other intervals on the new color space coordinate, and constructing a projection map of clustering color information.
Further, traversing the region of interest according to the projection map to obtain a texture skeleton map, which specifically includes:
traversing the region of interest, and expanding or shrinking scale information of the projection mapping graph to influence the ductility and complexity of the traversed texture skeleton graph;
and after the parameter range is selected, obtaining a texture skeleton diagram, wherein the parameters comprise skeleton information intensity, traversal scale size and traversal advancing step length.
Further, the step of further filling the texture skeleton map to generate a foreground texture interval specifically includes:
comparing the morphological gradient and the color similarity with the framework by taking the standard framework as a reference;
if the morphological gradient, the color similarity and the skeleton similarity are larger than a preset threshold value, and the pixel position is located in a pixel in the skeleton field, gathering the pixels into skeleton information;
and continuously iterating until all pixels which meet the standard around the skeleton are completely included in the skeleton information.
Further, the method further comprises:
obtaining foreground and background main body information to be segmented, and the occurrence position and the information main body of the whole extending process from the foreground to the background according to the texture path diagram;
and generating a trimap image by taking the foreground and background main body information needing to be segmented, the occurrence position and the information main body of the whole extending process from the foreground to the background as standards.
Further, the method further comprises:
and searching texture information of the image to be processed according to the texture path diagram.
The second object of the invention can be achieved by adopting the following technical scheme:
a high precision texture path retrieval apparatus, the apparatus comprising:
the acquisition module is used for acquiring the morphological gradient pointing diagram and acquiring the projection mapping diagram of the clustering color information;
the traversing module is used for traversing the region of interest according to the projection mapping graph by taking the morphological gradient pointing graph as a guide to obtain a texture skeleton graph;
and the filling module is used for further filling the texture skeleton diagram to generate a foreground texture interval so as to construct and obtain a texture path diagram.
The third object of the present invention can be achieved by adopting the following technical scheme:
a computer device comprises a processor and a memory for storing a program executable by the processor, wherein the processor realizes the high-precision texture path retrieval method when executing the program stored by the memory.
The fourth object of the present invention can be achieved by adopting the following technical scheme:
a computer-readable storage medium storing a program which, when executed by a processor, implements the high-precision texture path retrieval method described above.
Compared with the prior art, the invention has the following beneficial effects:
according to the method, the morphological gradient directional images are obtained, the projection mapping images of the clustering color information are obtained, the morphological gradient directional images are used as the guide, the texture skeleton images are obtained by traversing in the region of interest according to the projection mapping images, the texture skeleton images are further filled, foreground texture intervals are generated, so that the texture path images are constructed, the constructed texture path images comprise texture walking venation, texture strength and texture density in the texture extension areas, and the texture path images can be used for solving various problems such as portrait matting, edge trimming and problem judgment in image processing.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to the structures shown in these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a high-precision texture path retrieval method according to embodiment 1 of the present invention.
FIG. 2 is a morphological gradient map of example 1 of the present invention.
FIG. 3 is a visual representation of the digitized morphology gradient map of example 1 of the present invention.
Fig. 4 is a schematic diagram of the abstraction of the cube space into a two-dimensional image according to embodiment 1 of the present invention.
Fig. 5 is a projection map of embodiment 1 of the present invention.
FIG. 6 is a texture skeleton diagram of embodiment 1 of the present invention.
Fig. 7 is a texture interval diagram of embodiment 1 of the present invention.
Fig. 8 is a state diagram showing the substitution of the texture interval range into the real image projection map according to embodiment 1 of the present invention.
FIG. 9 is a trimap diagram of example 1 of the present invention.
FIG. 10 is a graph of division alpha in embodiment 1 of the present invention.
FIG. 11 is a schematic diagram of a texture path diagram according to embodiment 1 of the present invention, which is visualized in an image to be processed.
Fig. 12 is a trend chart of finding background development in the foreground texture interval according to embodiment 1 of the present invention.
Fig. 13 is a block diagram showing the structure of a high-precision texture path retrieval apparatus according to embodiment 2 of the present invention.
Fig. 14 is a block diagram showing the structure of a computer device according to embodiment 3 of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments, and all other embodiments obtained by persons of ordinary skill in the art without making any inventive effort based on the embodiments of the present invention are within the scope of protection of the present invention.
Example 1:
as shown in fig. 1, the present embodiment provides a high-precision texture path searching method, which is a method for searching a high-precision texture path based on a multi-level morphological gradient and a series of practical applications after searching texture path information, and is mainly applied to searching textures of hairs and peripheral areas thereof on portrait images, and has some limitations on non-hair areas, so that information collection and data comparison are only performed on hair information, background information and a small part of skin information, and the method comprises the following steps:
s101, acquiring a morphological gradient pointing map and acquiring a projection mapping map of cluster color information.
Because a large amount of disarranged information exists in the original image information, which is disturbing for extracting specific information, the embodiment adopts two aspects of information and comprehensively considers in order to acquire the information source capable of searching textures; on one hand, a morphological gradient directed graph constructed by the morphological gradient and on the other hand, a projection map on the clustering color information.
For a morphological gradient map, the present embodiment can be obtained by:
1) Based on the image gray information, gray level is calculated for a designated area, and a trend chart similar to a mountain range is constructed by gradient levels of surrounding information of each pixel point, as shown in fig. 2.
2) The trend graph is quantized into a cost chart, and the more intense the edge information is, the higher the cost value represented by the pixel is, and the lower the cost value is, otherwise.
3) According to the cost chart, the images with the numerical value of 0-255 are classified, and then the visual display is shown in fig. 3, the edge reaction is most obvious, the cost value is highest, the foreground and background boundary of the portrait is the information of the hair, the darker the color is, the denser the hair is, and the lighter the color is, the sparser the hair is.
The morphological gradient directional diagram obtained by the steps has the advantages of being not easily influenced by light environment, keeping the integrity of texture information of an image to the greatest extent, avoiding data loss caused by screening information, being easily influenced by background texture to disturb judgment, dividing basic information through deep learning, and being capable of better avoiding background problems by further processing only a region of interest.
For the projection map of the cluster color information, the present embodiment can be obtained by:
1) And acquiring basic texture information as a skeleton, and counting the color values of the areas to be measured and calculated, so that the counted color space falls on a cubic space of 255 x 255.
Acquiring basic texture information as a skeleton, more color information is needed to fill the basic texture information; the color information of the original pixels is rich and complete, but the screening cost required by a special environment is too high, a large amount of information is gushed in to influence texture information which has been abstracted before, so the embodiment utilizes a mode of combining clustering and projection mapping to review the color information of the image again, the image information is shown by the most suitable view angle, the color value of the area to be measured is counted, the color value of each pixel is formed by combining RGB three channel values, and the counted color space is also on a cubic space of 255 x 255.
2) The cube space is abstracted into a two-dimensional image, and as shown in fig. 4, the colors of all the partitions are gathered into a plurality of blocks, so as to obtain a cluster map.
3) And dividing the clustering blocks of the clustering graph by a clustering algorithm, and calculating a block data center.
4) And measuring main pixel intervals of each clustering block and main body values of block colors by taking the block data center as a reference point.
5) With the main body value as a standard, a new color space coordinate is established, colors in other intervals are mapped on the new color space coordinate, and a projection map of clustered color information is constructed, and since the problem of the embodiment is mainly concentrated between hair and background, the projection map constructed by taking hair as a reference is adopted, as shown in fig. 5.
S102, traversing the region of interest according to the projection mapping diagram by taking the morphological gradient pointing diagram as a guide to obtain a texture skeleton diagram.
After the projection map of the morphological gradient pointing map and the clustering color information is obtained, the texture map can be constructed, the morphological gradient pointing map is used as a guide, traversing is carried out on a region of interest, the traversed reference standard is mainly the constructed projection map information of the clustering color information (taking hair as a reference and comprising hair and skin), the extensibility and the complexity degree of the traversed texture map are influenced by expanding or shrinking the scale information of the map, the larger the scale range of the map is, the finer the difference between pixels is, the extensibility of the obtained texture map is reduced, the complexity degree is increased, the smaller the scale range is, the larger the interval difference between pixels is, more pixels are connected, the extensibility is increased, the complexity degree is reduced, meanwhile, the accuracy is also lost, the traversing can be guided by giving the weight of each direction according to the requirement or by externally designating the data direction besides the clustering color information; after a suitable parameter range is selected, the final traversed texture skeleton graph effect is shown in fig. 6, wherein the selected parameters comprise skeleton information intensity, traversal scale size and traversal advancing step length.
And S103, further filling the texture skeleton diagram to generate a foreground texture interval so as to construct and obtain a texture path diagram.
After the texture skeleton diagram is obtained, the texture skeleton diagram is further filled, only the textures of the skeleton have a plurality of defects, the whole generation interval from occurrence to disappearance of the texture information is extended based on the textures of the skeleton, and the interval also represents the interval extending from the foreground area to the background area.
In the embodiment, a standard skeleton is taken as a reference, comparison of morphological gradient and color similarity is started, pixels with high similarity degree with the skeleton and pixel positions in the skeleton field are converged into skeleton information, new iteration is performed on the filled skeleton information, and pixels in a range outside a circle are brought into screening; after a round of iteration, until all pixels meeting the standard around the skeleton are incorporated into the skeleton information, the area presented is the texture interval to be obtained, the obtained texture interval is shown in fig. 7, and the texture interval range is substituted into the state presented by the real image projection diagram, as shown in fig. 8.
After the texture skeleton diagram and the foreground texture interval are obtained, the construction of the texture path diagram is completed, the texture path diagram comprises texture walking venation, and the texture extension area also has texture intensity and texture density, so that the method is greatly helpful for the subsequent image processing.
The present embodiment may apply the constructed texture path map to trimap map generation, so the high-precision texture path retrieval method of the present embodiment may further include the steps of:
s104, obtaining foreground and background main body information to be segmented, and the occurrence position and the information main body of the whole extending process from the foreground to the background according to the texture path diagram.
In the portrait matting, the matting algorithm determines whether the image after the final matting finishes replacing the background is natural and has no segmentation sense, one important parameter in the matting algorithm is a trimap image, and the trimap image guides the cognition of the algorithm to the foreground and the background of the image.
S105, generating a trimap image by taking the foreground and background main body information needing to be segmented, the occurrence position and the information main body of the whole extending process from the foreground to the background as standards.
The trimap graph constructed by the embodiment is more accurate, the calculation requirement on the matrix is lower, the calculation speed is faster, the trimap graph is shown in fig. 9, and the finally generated segmentation alpha graph through the trimap graph is shown in fig. 10.
The present embodiment may also apply the constructed texture path map to texture information lookup, so the high-precision texture path retrieval method of the present embodiment may further include the steps of:
s106, searching texture information of the image to be processed according to the texture path diagram.
After the texture path diagram is constructed, the trend of the image texture is clearly perceived, the qualitative of the texture problem is well referenced, the starting, the extending and the ending of each texture can be measured through the texture path diagram, the extending of a plurality of textures can be sourced from the same starting point, the plurality of textures can also be ended on the same pixel, even the texture path diagram comprises cross branches in the process of extending the texture, the texture path diagram can be stored in the form of a data link list, and the texture path diagram is visually displayed in the image to be processed, as shown in fig. 11; the opposite background extension map can be also reversely deduced from the texture path map, and the background development trend is found in the foreground texture interval, as shown in fig. 12.
It should be noted that although the method operations of the above embodiments are depicted in the drawings in a particular order, this does not require or imply that the operations must be performed in that particular order or that all illustrated operations be performed in order to achieve desirable results. Rather, the depicted steps may change the order of execution. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step to perform, and/or one step decomposed into multiple steps to perform.
Example 2:
as shown in fig. 13, the present embodiment provides a high-precision texture path retrieval device, which includes an acquisition module 1301, a traversal module 1302, and a filling module 1303, where specific functions of the respective modules are as follows:
an obtaining module 1301, configured to obtain a morphological gradient map, and obtain a projection map of cluster color information.
The traversing module 1302 is configured to traverse the region of interest according to the projection map with the morphological gradient pointing map as a guide, and obtain a texture skeleton map.
The filling module 1303 is configured to further fill the texture skeleton map, and generate a foreground texture interval to construct and obtain a texture path map.
Specific implementation of each unit in this embodiment may be referred to embodiment 1, and will not be described in detail herein; it should be noted that, the apparatus provided in this embodiment is only exemplified by the division of the above functional units, and in practical application, the above functional allocation may be performed by different functional units according to needs, that is, the internal structure is divided into different functional units, so as to perform all or part of the functions described above.
Example 3:
the present embodiment provides a computer apparatus, as shown in fig. 14, which includes a processor 1402, a memory, an input device 1403, a display device 1404 and a network interface 1405, which are connected through a system bus 1401, the processor is configured to provide computing and control capabilities, the memory includes a nonvolatile storage medium 1406 and an internal memory 1407, the nonvolatile storage medium 1406 stores an operating system, a computer program and a database, the internal memory 1407 provides an environment for the operating system and the computer program in the nonvolatile storage medium, and the processor 1402 implements the high-precision texture path retrieval method of the above embodiment 1 when executing the computer program stored in the memory, as follows:
acquiring a morphological gradient pointing diagram and a projection mapping diagram of cluster color information;
taking the morphological gradient pointing diagram as a guide, traversing the region of interest according to the projection mapping diagram to obtain a texture skeleton diagram;
and further filling the texture skeleton diagram to generate a foreground texture interval so as to construct and obtain a texture path diagram.
Example 4:
the present embodiment provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements the high-precision texture path retrieval method of embodiment 1 described above, as follows:
acquiring a morphological gradient pointing diagram and a projection mapping diagram of cluster color information;
taking the morphological gradient pointing diagram as a guide, traversing the region of interest according to the projection mapping diagram to obtain a texture skeleton diagram;
and further filling the texture skeleton diagram to generate a foreground texture interval so as to construct and obtain a texture path diagram.
The computer readable storage medium of the present embodiment may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
In summary, the present invention obtains the morphological gradient directional map, obtains the projection mapping map of the clustering color information, uses the morphological gradient directional map as a guide, traverses the region of interest according to the projection mapping map to obtain the texture skeleton map, further fills the texture skeleton map to generate the foreground texture interval, and constructs the texture path map, wherein the constructed texture path map comprises texture walking venation, the texture extension region also comprises texture intensity and texture density, and the texture path map can be applied to various problems of image processing.
The above-mentioned embodiments are only preferred embodiments of the present invention, but the protection scope of the present invention is not limited thereto, and any person skilled in the art can make equivalent substitutions or modifications according to the technical solution and the inventive concept of the present invention within the scope of the present invention disclosed in the present invention patent, and all those skilled in the art belong to the protection scope of the present invention.
Claims (6)
1. A method for retrieving a high-precision texture path, the method comprising:
acquiring a morphological gradient pointing diagram and a projection mapping diagram of cluster color information;
taking the morphological gradient pointing diagram as a guide, traversing the region of interest according to the projection mapping diagram to obtain a texture skeleton diagram;
further filling the texture skeleton diagram to generate a foreground texture interval so as to construct and obtain a texture path diagram;
searching texture information of the image to be processed according to the texture path diagram;
the obtaining a morphological gradient pointing diagram specifically comprises the following steps:
calculating gray level according to the appointed area based on the image gray information, and constructing a trend graph similar to a mountain range through gradient levels of surrounding information of each pixel point;
digitizing the trend graph into a cost graph;
dividing the images into images with values ranging from 0 to 255 according to a cost chart;
the obtaining the projection mapping diagram of the cluster color information specifically includes:
acquiring basic texture information as a skeleton, and counting color values of an area to be measured and calculated to enable the counted color space to fall on a cubic space of 255 x 255;
abstracting the cube space into a two-dimensional image, and gathering the colors of all the partitions into a plurality of blocks to obtain a cluster map;
dividing the clustering blocks of the clustering graph by a clustering algorithm, and calculating a block data center;
measuring main pixel intervals of each clustering block and main body values of block colors by taking the block data center as a reference point;
establishing a new color space coordinate by taking the main body numerical value as a standard, mapping the colors of other intervals on the new color space coordinate, and constructing a projection map of clustering color information;
traversing the region of interest according to the projection mapping diagram to obtain a texture skeleton diagram, which specifically comprises:
traversing the region of interest, and expanding or shrinking scale information of the projection mapping graph to influence the ductility and complexity of the traversed texture skeleton graph;
and after the parameter range is selected, obtaining a texture skeleton diagram, wherein the parameters comprise skeleton information intensity, traversal scale size and traversal advancing step length.
2. The method for retrieving a high-precision texture path according to claim 1, wherein the step of further filling the texture skeleton map to generate a foreground texture interval specifically comprises:
comparing the morphological gradient and the color similarity with the framework by taking the standard framework as a reference;
if the morphological gradient, the color similarity and the skeleton similarity are larger than a preset threshold value, and the pixel position is located in a pixel in the skeleton field, gathering the pixels into skeleton information;
and continuously iterating until all pixels which meet the standard around the skeleton are completely included in the skeleton information.
3. The high precision texture path retrieval method of claim 1, further comprising:
obtaining foreground and background main body information to be segmented, and the occurrence position and the information main body of the whole extending process from the foreground to the background according to the texture path diagram;
and generating a trimap image by taking the foreground and background main body information needing to be segmented, the occurrence position and the information main body of the whole extending process from the foreground to the background as standards.
4. A high precision texture path retrieval apparatus, the apparatus comprising:
the acquisition module is used for acquiring the morphological gradient pointing diagram and acquiring the projection mapping diagram of the clustering color information;
the traversing module is used for traversing the region of interest according to the projection mapping graph by taking the morphological gradient pointing graph as a guide to obtain a texture skeleton graph;
the filling module is used for further filling the texture skeleton diagram to generate a foreground texture interval so as to construct and obtain a texture path diagram;
the retrieval module is used for searching texture information of the image to be processed according to the texture path diagram;
the obtaining a morphological gradient pointing diagram specifically comprises the following steps:
calculating gray level according to the appointed area based on the image gray information, and constructing a trend graph similar to a mountain range through gradient levels of surrounding information of each pixel point;
digitizing the trend graph into a cost graph;
dividing the images into images with values ranging from 0 to 255 according to a cost chart;
the obtaining the projection mapping diagram of the cluster color information specifically includes:
acquiring basic texture information as a skeleton, and counting color values of an area to be measured and calculated to enable the counted color space to fall on a cubic space of 255 x 255;
abstracting the cube space into a two-dimensional image, and gathering the colors of all the partitions into a plurality of blocks to obtain a cluster map;
dividing the clustering blocks of the clustering graph by a clustering algorithm, and calculating a block data center;
measuring main pixel intervals of each clustering block and main body values of block colors by taking the block data center as a reference point;
establishing a new color space coordinate by taking the main body numerical value as a standard, mapping the colors of other intervals on the new color space coordinate, and constructing a projection map of clustering color information;
traversing the region of interest according to the projection mapping diagram to obtain a texture skeleton diagram, which specifically comprises:
traversing the region of interest, and expanding or shrinking scale information of the projection mapping graph to influence the ductility and complexity of the traversed texture skeleton graph;
and after the parameter range is selected, obtaining a texture skeleton diagram, wherein the parameters comprise skeleton information intensity, traversal scale size and traversal advancing step length.
5. A computer device comprising a processor and a memory for storing a program executable by the processor, wherein the processor, when executing the program stored in the memory, implements the high precision texture path retrieval method of any one of claims 1-3.
6. A computer-readable storage medium storing a program, wherein the program, when executed by a processor, implements the high-precision texture path retrieval method of any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311442825.4A CN117171384B (en) | 2023-11-02 | 2023-11-02 | High-precision texture path retrieval method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311442825.4A CN117171384B (en) | 2023-11-02 | 2023-11-02 | High-precision texture path retrieval method, device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117171384A CN117171384A (en) | 2023-12-05 |
| CN117171384B true CN117171384B (en) | 2024-02-23 |
Family
ID=88932032
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311442825.4A Active CN117171384B (en) | 2023-11-02 | 2023-11-02 | High-precision texture path retrieval method, device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117171384B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101615252A (en) * | 2008-06-25 | 2009-12-30 | 中国科学院自动化研究所 | A kind of method for extracting text information from adaptive images |
| WO2015089949A1 (en) * | 2013-12-19 | 2015-06-25 | 成都品果科技有限公司 | Human face clustering method merging lbp and gabor features |
| CN110334692A (en) * | 2019-07-17 | 2019-10-15 | 河南科技大学 | A kind of blind way recognition methods based on image procossing |
| CN112561976A (en) * | 2020-12-09 | 2021-03-26 | 齐鲁工业大学 | Image dominant color feature extraction method, image retrieval method, storage medium and device |
| CN114092486A (en) * | 2021-10-11 | 2022-02-25 | 安庆师范大学 | Automatic segmentation method and device for image texture background |
| CN114202610A (en) * | 2021-12-27 | 2022-03-18 | 武汉大势智慧科技有限公司 | Oblique photography building single model texture mapping method |
-
2023
- 2023-11-02 CN CN202311442825.4A patent/CN117171384B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101615252A (en) * | 2008-06-25 | 2009-12-30 | 中国科学院自动化研究所 | A kind of method for extracting text information from adaptive images |
| WO2015089949A1 (en) * | 2013-12-19 | 2015-06-25 | 成都品果科技有限公司 | Human face clustering method merging lbp and gabor features |
| CN110334692A (en) * | 2019-07-17 | 2019-10-15 | 河南科技大学 | A kind of blind way recognition methods based on image procossing |
| CN112561976A (en) * | 2020-12-09 | 2021-03-26 | 齐鲁工业大学 | Image dominant color feature extraction method, image retrieval method, storage medium and device |
| CN114092486A (en) * | 2021-10-11 | 2022-02-25 | 安庆师范大学 | Automatic segmentation method and device for image texture background |
| CN114202610A (en) * | 2021-12-27 | 2022-03-18 | 武汉大势智慧科技有限公司 | Oblique photography building single model texture mapping method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117171384A (en) | 2023-12-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102154470B1 (en) | 3D Human Hairstyle Generation Method Based on Multiple Feature Search and Transformation | |
| KR102607800B1 (en) | Medical image segmentation methods and devices, electronic devices, and storage media | |
| US11748934B2 (en) | Three-dimensional expression base generation method and apparatus, speech interaction method and apparatus, and medium | |
| CN104616345B (en) | Octree forest compression based three-dimensional voxel access method | |
| Ip et al. | Hierarchical exploration of volumes using multilevel segmentation of the intensity-gradient histograms | |
| CN109558864A (en) | Face critical point detection method, apparatus and storage medium | |
| WO2021093451A1 (en) | Pathological section image processing method, apparatus, system, and storage medium | |
| CN108463823A (en) | A kind of method for reconstructing, device and the terminal of user's Hair model | |
| CN115330940B (en) | Three-dimensional reconstruction method, device, equipment and medium | |
| CN110909680A (en) | Facial expression recognition method and device, electronic equipment and storage medium | |
| DE102021210435A1 (en) | Process and device for the visualization of three-dimensional objects | |
| CN113421204A (en) | Image processing method and device, electronic equipment and readable storage medium | |
| CN109584249B (en) | Three-dimensional volume data segmentation method based on closed form solution | |
| Nousias et al. | A saliency aware CNN-based 3D model simplification and compression framework for remote inspection of heritage sites | |
| CN107481243A (en) | Sheep body chi detection method based on sheep top view | |
| CN108597589B (en) | Model generation method, target detection method and medical imaging system | |
| CN112819767A (en) | Image processing method, apparatus, device, storage medium, and program product | |
| CN117171384B (en) | High-precision texture path retrieval method, device, computer equipment and storage medium | |
| CN112598792A (en) | Multi-fractal quantification method and system for terrain complexity in three-dimensional scene | |
| CN113255514B (en) | Behavior identification method based on local scene perception graph convolutional network | |
| dos Passos et al. | Sample-based synthesis of illustrative patterns | |
| Burdescu et al. | Computational complexity analysis of the graph extraction algorithm for 3D segmentation | |
| CN113239867A (en) | Mask region self-adaptive enhancement-based illumination change face recognition method | |
| CN112825122A (en) | Ethnicity judgment method, ethnicity judgment device, ethnicity judgment medium and ethnicity judgment equipment based on two-dimensional face image | |
| CN113674383A (en) | Method and device for generating text image |
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 |