CN116740374A - Repeated texture recognition method and device - Google Patents

Abstract

This application provides a repetitive texture recognition method and device. This method horizontally splices two images corresponding to an image pair, and calculates the slope of the connecting line of the matching feature point pairs in the two images. Further statistics are made on the ratio of lines where the deviation between the slope and the slope standard value is greater than or equal to the set value. If the ratio is greater than the corresponding preset value, it indicates that the image pair has a small area of repeated texture in the non-main direction of feature matching, that is, it is determined The matching relationship of the image pair is wrong, and the matching relationship of the image pair is deleted. It can be seen that this method does not need to analyze the content information of the feature points in the image pair. It only identifies repeated textures that are not in the main direction of feature matching through the slope of the feature point connection. Therefore, this method is not sensitive to the input image and improves the robustness of the method. sex.

Description

Repeated texture recognition method and device

Technical field

The present application relates to the field of three-dimensional reconstruction technology, and in particular to a repetitive texture recognition method and device.

Background technique

Three-dimensional reconstruction (3D Reconstruction) is to restore the three-dimensional structure of an object or scene from multiple two-dimensional images, and finally establish a virtual reality that expresses the objective world in the computer. Three-dimensional reconstruction is used in many scenarios, for example, scenarios where three-dimensional digital models are constructed based on real scenes, such as AR augmented reality, and the creation of three-dimensional digital models of cultural relics.

Objects of the same shape and texture may exist in the 2D images used for 3D reconstruction. These object features of the same shape and texture may form erroneous matches, leading to errors in the 3D reconstruction model. For example, the same electrical appliances, such as water dispensers with the same shape and texture, are placed at different locations in a certain space environment. When performing a three-dimensional reconstruction of this scene, it is likely that the water dispensers at different locations will be recognized as the same water dispenser. This leads to errors in the 3D reconstruction model of the environment. Therefore, during the three-dimensional reconstruction process, how to accurately identify and filter out repeated textures is an urgent problem that needs to be solved.

Contents of the invention

In view of this, this application provides a repetitive texture recognition method and device to solve the above technical problems. The disclosed technical solutions are as follows:

In a first aspect, the present application provides a repetitive texture recognition method, which is applied to electronic devices. The method includes: obtaining feature point matching information contained in at least two images with a matching relationship, where the feature point matching information includes at least two images containing information of matching feature points; splice at least two images horizontally, and calculate the slope of the lines connecting all matching feature points in at least two images; count the number of lines connecting the first preset conditions in at least two images , the first preset condition includes that the deviation between the slope and the slope standard value is greater than or equal to the first preset value; it is determined that the number of connected lines meets the second preset condition, and the second preset condition includes those that meet the first preset condition. The ratio of the connecting lines is greater than or equal to the second preset value; it is determined that at least two images contain small areas of repeated texture. It can be seen that this scheme does not need to analyze the content information of the feature points in the image pair. It only identifies repeated textures that are not in the main direction of feature matching through the slope of the feature point connection line. Therefore, this method is not sensitive to the input image and improves the robustness of the method. sex.

In a possible implementation of the first aspect, before horizontally splicing at least two images, the method further includes: counting the number of matching feature points contained in any one of the at least two images; if the number of matching feature points If it is greater than or equal to the first threshold, perform horizontal splicing of at least two images; if the number of matching feature points is less than the first threshold, it is determined that there is a small area of repeated texture in at least two images. In this way, the existence of small-area repeated textures can be initially identified through the number of matching feature points in the image. If the number of matching feature points in the image is less than the first threshold, it is determined that the image pair has small-area repeated textures, and then the image pair is determined. Matching relationship errors improve the efficiency of identifying repeated textures.

In a possible implementation of the first aspect, the process of determining the slope standard value includes: calculating the slope median value corresponding to all the connecting lines included in the image pair, and determining the slope median value as the slope standard value.

In a possible implementation of the first aspect, the method further includes: deleting a matching relationship between at least two images containing repeated textures in small areas.

In a possible implementation of the first aspect, the method further includes: after determining that the number of connections in the at least two images that meet the first preset condition does not meet the second preset condition, counting any of the connections in the at least two images. The distribution range of matching feature points contained in an image; if the distribution range is less than or equal to the preset range, it is determined that at least two images contain small-area repeated textures. In this way, repeated textures that cannot be identified through the slope of the feature point connection line can further be identified by matching the distribution of feature points to identify image pairs with repeated textures. It can be seen that this solution improves the accuracy of identifying repeated textures.

In a possible implementation of the first aspect, counting the distribution range of matching feature points contained in any one of the at least two images includes: dividing any one of the at least two images into multiple grids, and counting The number of grids containing matching feature points in any image; connect the grids containing matching feature points and adjacent locations into a connected area; determine the distribution range of matching feature points based on the parameters of the connected area contained in any image, parameters Include at least one of the number and area of regions. It can be seen that this scheme counts the distribution of feature points by dividing the grid in the image, which is simple and effective.

In a possible implementation of the first aspect, determining the distribution range of the matching feature points based on the parameters of the area included in any image includes: if the number of all connected areas included in any image is less than or equal to a third threshold, determining The distribution range of matching feature points is smaller than the preset range; if the number of all connected areas contained in any image is greater than the third threshold, determine whether the total area of all connected areas in any image is less than or equal to the fourth threshold; if the total area is less than Or equal to the fourth threshold, it is determined that at least two images contain repeated textures in small areas; if the total area is greater than the fourth threshold, it is determined that the matching relationship between at least two images is correct. In this way, the distribution of matching feature points is determined by counting the number or area of connected regions contained in the image, which improves the accuracy of repeated texture recognition results.

In a possible implementation of the first aspect, before counting the distribution range of matching feature points included in any one of the at least two image pairs, the method further includes: counting the matching included in any one of the at least two images. The number of feature points; if the number of matching feature points is less than or equal to the second threshold, perform the step of counting the distribution range of matching feature points contained in any image in at least two image pairs; if the number of matching feature points is greater than the second threshold , confirm that the matching relationship between at least two images is correct. It can be seen that this scheme initially identifies images without repeated textures in small areas through the number of matching feature points contained in the images, and reduces the number of images required to identify whether repeated textures in small areas exist. Therefore, it improves the efficiency of identifying repeated textures.

In a second aspect, this application also provides a repetitive texture recognition method, which is applied to electronic devices. The method includes: obtaining feature point matching information contained in at least two image pairs with a matching relationship, and the feature point matching information includes at least two image pairs. Information about matching feature points contained in the image; counting the distribution range of matching feature points contained in any one of at least two images; determining that at least two images whose distribution range is less than or equal to a preset range contain small-area repeated textures. It can be seen that this scheme does not need to analyze image feature information, but identifies repeated textures in small areas through the distribution of matching feature point pairs contained in image pairs. Therefore, this method is insensitive to the input image and improves the robustness of the method.

In a possible implementation of the second aspect, counting the distribution range of matching feature points contained in any image in the image pair includes: dividing any one of at least two images into multiple grids, and counting any one The number of grids containing matching feature points in the image; connect grids containing matching feature points and adjacent locations into a connected area; determine the distribution range of matching feature points based on the parameters of the connected area contained in any image, and the parameters include areas At least one of quantity and area.

In a possible implementation of the second aspect, determining the distribution range of the matching feature points based on parameters of the areas included in any image includes: if the number of all connected areas included in any image is less than or equal to a third threshold, determining The distribution range of matching feature points is smaller than the preset range; if the number of all connected areas contained in any image is greater than the third threshold, determine whether the total area of all connected areas in any image is less than or equal to the fourth threshold; if the total area is less than Or equal to the fourth threshold, it is determined that at least two images contain repeated textures in small areas; if the total area is greater than the fourth threshold, it is determined that the matching relationship between at least two images is correct.

In a possible implementation of the second aspect, before counting the distribution range of matching feature points included in any one of the at least two image pairs, the method further includes: counting the matching included in any one of the at least two images. The number of feature points; if the number of matching feature points is less than or equal to the second threshold, perform the step of counting the distribution range of matching feature points contained in any image in at least two image pairs; if the number of matching feature points is greater than the second threshold , confirm that the matching relationship between at least two images is correct.

In a third aspect, this application also provides an electronic device. The electronic device includes: one or more processors, memories and touch screens; the memory is used to store program codes; the processor is used to run the program codes, so that the electronic device implements the first A repetitive texture identification method for either aspect or the second aspect.

In a fourth aspect, the present application also provides a computer-readable storage medium on which instructions are stored. When the instructions are run on an electronic device, the electronic device causes the electronic device to perform the repetitive texture recognition of any one of the first aspect or the second aspect. method.

In a fifth aspect, the present application also provides a computer program product with execution stored thereon. When the computer program product is run on an electronic device, the electronic device implements the repetitive texture of any one of the first aspect or the second aspect. recognition methods.

It should be understood that the description of technical features, technical solutions, beneficial effects or similar language in this application does not imply that all features and advantages can be achieved in any single embodiment. On the contrary, it can be understood that the description of features or beneficial effects means that specific technical features, technical solutions or beneficial effects are included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions or beneficial effects in this specification do not necessarily refer to the same embodiments. Furthermore, the technical features, technical solutions and beneficial effects described in this embodiment can also be combined in any appropriate manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions or beneficial effects of a specific embodiment. In other embodiments, additional technical features and beneficial effects may also be identified in specific embodiments that do not embody all embodiments.

Description of drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are: For some embodiments of the present invention, those of ordinary skill in the art can also obtain other drawings based on these drawings without exerting creative efforts.

Figure 1 is a flow chart of a repetitive texture recognition method provided by an embodiment of the present application;

Figure 2 is a schematic diagram of an image pair provided by an embodiment of the present application;

Figure 3 is a schematic diagram of another image pair provided by an embodiment of the present application;

Figure 4 is a flow chart of another repetitive texture recognition method provided by an embodiment of the present application;

Figure 5 is a flow chart of yet another repetitive texture recognition method provided by an embodiment of the present application;

FIG. 6 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

Detailed ways

The terms “first”, “second”, “third”, etc. in the description, claims and drawings of this application are used to distinguish different objects, rather than to limit a specific order.

In the embodiments of this application, words such as "exemplary" or "for example" are used to represent examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "such as" in the embodiments of the present application is not to be construed as preferred or advantageous over other embodiments or designs. Rather, use of the words "exemplary" or "such as" is intended to present the concept in a concrete manner.

Please refer to Figure 1, which shows a flow chart of a repetitive texture recognition method provided by an embodiment of the present application. This method can be applied to electronic equipment such as servers. The electronic equipment includes a feature extraction module, a feature matching module, a first filter removal module and a second filtering module.

As shown in Figure 1, the method includes the following steps:

S11, the feature extraction module obtains multiple views of the three-dimensional reconstructed object.

For example, images of multiple different perspectives of the three-dimensional reconstructed object, that is, multiple views, can be collected through terminal devices (such as cameras, smartphones, virtual reality devices, etc.).

Multiple views of the three-dimensional reconstruction object collected by the terminal device are uploaded to the electronic device for the three-dimensional reconstruction process.

S12, the feature extraction module extracts the feature points of each view to obtain a feature map.

In an exemplary embodiment of the present application, a scale-invariant feature transform (SIFT) method can be used to extract feature points from each image. The purpose of SIFT's feature point screening is to find extreme points in different scale spaces and ensure that these feature points exist under zoom-in or zoom-out conditions.

S13, the feature extraction module transfers all feature maps to the feature matching module.

S14: The feature matching module performs feature matching on all feature maps to obtain image pairs with matching relationships, and transfers all image pairs with matching relationships to the first filtering module.

In an exemplary embodiment, each image has a unique identifier, and the image pair in this article includes the unique identifiers of two matching images and their mapping relationship. For example, after the feature matching module performs feature matching on image A and image B, it is confirmed that the successfully matched features in the two images meet the preset conditions (for example, the number of successfully matched features is greater than the preset value).

In an exemplary embodiment, a random sample consensus (RANSAC) algorithm can be used to perform feature matching on all feature maps to obtain image pairs with matching relationships, and feature point pairs with matching relationships contained in the image pairs ( Or called matching feature point pairs) information. The RANSAC algorithm is a simple and effective way to remove the influence of noise and estimate the model. The algorithm uses as few points as possible to estimate the model parameters, and then expands the influence range of the obtained model parameters as much as possible.

S15, the first filtering module obtains the feature point matching information of the image pair.

For example, in an exemplary embodiment, the feature point matching information may include the number of feature points with matching relationships, location information, feature values, etc.

S16: Determine whether the number of matching feature points contained in one image in the image pair is greater than or equal to the first threshold; if so, perform S17; if not, perform S20.

Matching feature points refer to feature points that have a matching relationship between the two images contained in the image pair.

As shown in Figure 2, image 30 and image 40 are two images that are determined to have a matching relationship after feature matching by the feature matching module. The scenes shown in images 30 and 40 include objects such as tables.

In this example, there is a matching relationship between feature point A contained in image 30 and feature point B contained in Figure 20 , that is, point A and point B are a pair of matching feature points. This example shows only one matching feature point pair.

Count the number of feature points in any image of the image pair that have a matching relationship with the other image. If the number of feature points that have a matching relationship is greater than or equal to the first threshold, it indicates that the matching relationship of the image pair may be correct.

The first threshold can be statistically obtained based on a limited number of test results. In this example, the value of the first threshold can be 15. This application does not limit the numerical range of the first threshold.

In one application scenario, for image pairs corresponding to successfully matched feature points (or called matching feature points), the proportion of the slopes of the lines connecting all matching feature points in the two images within a slope range is greater than or equal to the preset Scale, the slope range in such a scene can be called the main direction of feature matching for this image pair. For example, the slope of the line connecting successfully matched feature points in images A and B exceeds 80% and is about 60°, and the main direction of feature matching in images A and B is about 60°.

However, there are still matching feature points that are not in the main direction in the two images. Such feature points are mismatched feature points. If there are mismatched feature points in the two images, it can be determined that the two images are mismatched. image. Such situations can be identified through the process shown in steps S17 to S19 below.

S17, splice the image pairs horizontally, and calculate the slope and slope median of all matching feature point pairs.

Among them, horizontal splicing refers to splicing two images in the horizontal direction (such as the X-axis direction).

FIG. 2 shows a schematic diagram of the image 30 and the image 40 spliced in the X-axis direction, and the calculation of the slope of the line AB connecting the feature point B and the feature point A.

The median slope is the median of the slopes of the lines connecting all matching feature points in the image pair. For example, the image pair contains 20 matching feature point pairs, that is, the image pair can obtain 20 feature point connecting lines. Calculate these 20 features separately. The slope of the line connecting the points is then calculated to obtain the median of the slope of the line connecting the 20 feature points.

S18: Count the number of connected lines in which the deviation between the slope of the connecting line corresponding to the same image pair and the median slope is greater than or equal to the first preset value.

In the embodiment of this application, the first preset value can be determined based on the statistical results of a limited number of tests. For example, the first preset value is 8°. This application does not limit the preset value corresponding to the slope deviation.

S19: Determine whether the proportion of qualified connections is greater than or equal to the second preset value; if yes, execute S20; if not, execute S22.

In the embodiment of the present application, the second preset value may be determined based on experimental statistical results. For example, the numerical range of the second preset value may be 5% to 15%.

The ratio of the number of qualified connection lines refers to the ratio of the number of qualified feature point connections to the number of all feature point connections included in the image pair. For example, assuming that the second preset value is 5%, an image pair contains 8 feature point connections that meet the conditions, and the image pair contains a total of 100 feature point connections, that is, the proportion of feature point connections that meet the conditions is 8%. , obviously, the connection ratio of the image to the load condition contained is greater than the second preset value.

S20: Delete the matching relationship of the image pair.

For example, if image A and image B are two images with an incorrect matching relationship, delete the matching mapping relationship between image A and image B.

The process described in S15 to S20 above can screen out image pairs whose matching feature points are inconsistent with the main direction of feature matching. Such images are usually images with wrong matching relationships.

In another scenario, as shown in Figure 3, image 10 and image 20 are image pairs with a matching relationship. In Figure 3, image 10 and image 20 are spliced horizontally. Moreover, the matching feature points contained in images 10 and 20 are concentrated in the area where the word "福" is located. The dotted line in the "福" area in Figure 3 represents the connection line between the feature points with a matching relationship in image 10 and Figure 20. As shown in Figure 3, the slopes of the lines connecting the feature points of the two images are basically consistent, but the matching feature points of the two images are concentrated in a certain small area. This type of repetitive texture cannot be identified by the slope of the above-mentioned feature point connection line. Therefore, embodiments of the present application also provide another solution for filtering out repeated textures. This solution identifies images in which feature points are concentrated in a small area of the image by matching the distribution of feature points in the image. This process can include the following steps:

S21, determine whether the number of matching feature points is less than or equal to the second threshold; if so, execute S22; if not, execute S26.

In the embodiment of the present application, the second threshold is greater than the first threshold. The second threshold can also be statistically obtained based on the results of a limited number of tests. For example, the value of the second threshold in this example can be 500. The value of the second threshold in this application is The scope is not limited.

If the number of matching feature points contained in the image is greater than the first threshold and less than or equal to the second threshold, it is necessary to further identify whether there is a small area of repeated texture in the image. If the number of matching feature points is greater than the second threshold, it is determined that the matching relationship of the image pair is correct, and the image pair is directly retained.

S22: Divide any image corresponding to the image pair into N*M grids, and count the distribution of grids containing matching feature points in the image.

It can be understood that M and N can be adjusted according to the size of the image. In an exemplary embodiment, it is necessary to ensure that each grid contains a certain number of pixels, for example, each grid contains 40×40 pixels. .

S23: Connect adjacent grids containing matching feature points in any image into a region (or called a connected region).

Matching feature points refer to feature points in any image of an image pair that have a matching relationship with feature points of the other image. For example, feature point A in image 30 in Figure 2 matches feature point B in image 40, so , both point A and point B can be called matching feature points.

In an exemplary embodiment, N*M grids in the image are first traversed to screen out grids containing feature points, and then the grids containing feature points are traversed to screen out another grid containing features that are paired with the image. The image has a grid of matching feature points, and finally the grids containing matching feature points and adjacent positions are connected into a region.

S24, determine whether the number of areas contained in the image is less than the third threshold; if so, execute S26; if not, execute S25.

Count the number of areas obtained by connecting the grids contained in any image in the image pair. If the number is less than a certain preset value, such as 3, it is determined that the matching feature points of the image pair are distributed in a smaller area, and then Determine the presence of repeating texture in this small area.

Usually, the matching feature points of a correctly matched image pair are relatively evenly distributed on the image. Therefore, if the matching feature points of an image pair are concentrated in a small area, it can be determined that there is an error in the matching relationship of the image pair.

As shown in Figure 3, image 10 and image 20 are image pairs with a matching relationship. In Figure 3, image 10 and image 20 are spliced horizontally. Moreover, the matching feature points contained in images 10 and 20 are concentrated in the area where the word "福" is located. Therefore, by judging whether the number of areas contained in the image is less than the third threshold, it can be determined that the image pair is an image pair containing repeated textures. There is an error in its matching relationship.

S25: Determine whether the maximum number of grids contained in the area is less than the fourth threshold; if so, execute S26; if not, execute S27.

If the number of regions contained in any image in the image pair is greater than the third threshold, continue to determine whether the sum of the number of grids contained in all connected regions in the image is less than the fourth threshold. If so, it indicates that the matching feature points of the image are concentrated in a small area. Inside.

In an exemplary embodiment of the present application, the fourth threshold may be determined based on the total number of grids contained in the image. For example, in an example, the fourth threshold may be set to 0.05×the total number of grids.

S26: Delete the matching relationship of the image pair.

If the number of areas is less than the third threshold, or the number of grids contained in the largest area is less than the fourth threshold, it is determined that the matching feature points of the image pair are concentrated in a small area. As mentioned before, the matching feature points of a correctly matched image pair are usually evenly distributed throughout the image. Therefore, if the matching feature points of the two images are concentrated in a small area, it is wrong to determine the matching relationship between the two images. , need to remove incorrectly matched image pairs.

For example, if image A and image B are an image pair with an incorrect matching relationship, delete the matching mapping relationship between image A and image B.

S27, retain the matching relationship of the image pair.

If the sum of the number of grids contained in all connected areas in the image is greater than the fourth threshold, it indicates that the sum of the connected areas is larger. In other words, the matching feature points of the image pair have a larger distribution range, and the matching relationship of the image pair is finally determined. Correct, so the matching relationship of this image pair is preserved.

S28, determine whether there are unprocessed image pairs.

In an exemplary embodiment, parameter i can be set to represent the number of currently unprocessed image pairs corresponding to the same three-dimensional reconstruction object, and the value of parameter i is updated every time an image pair is processed, such as i=i-1. In this scenario, if the value of i is equal to 0, it indicates that there is no unprocessed image pair. If the value of i is greater than 0, it indicates that there is an unprocessed image pair.

In another exemplary embodiment of the present application, the parameter j can be set to represent the number of processed image pairs corresponding to the same three-dimensional reconstruction object, and j is increased by 1 for each image pair processed. In this scenario, if the value of j is less than the number of all image pairs corresponding to the three-dimensional reconstruction object, it indicates that there are unprocessed image pairs; if the value of j is equal to the number of all image pairs corresponding to the three-dimensional reconstruction object, it indicates that there are no Unprocessed image pairs.

If there are unprocessed image pairs, return to execution S15 to continue processing the next image pair. If there are no unprocessed image pairs, execute S29.

S29, output image pairs that retain matching relationships.

The image pairs with matching relationships are retained, that is, the image pairs with correct matching relationships are filtered out, and finally the image pairs with correct matching relationships are output and subsequent processing continues.

It can be understood that only the processes shown in S15 to S20 can be used to filter out image pairs with repeated textures in small areas in non-main directions, or only the processes shown in S21 to S27 can be used to filter out image pairs concentrated in small areas. An image pair of repeating textures.

The repetitive texture recognition method provided by this embodiment calculates the slope of the connecting line of the matching feature point pair contained in the image pair and the median slope of the image pair, and counts the number of connecting lines whose deviation from the slope median is greater than the first preset value. , if the proportion of connections that meet this condition is greater than the second preset value, it indicates that the matching relationship of the image pair is wrong, and the matching relationship of the image pair is deleted. Furthermore, this method can identify whether the matching relationship of the image pair is correct based on the distribution of matching feature point pairs contained in the image pair. Specifically, the image can be divided into multiple grids, and the distribution of feature points on the grids can be counted. Connect adjacent grids containing matching feature point pairs into a region, and count the number of regions contained in the image and the number of grids contained in each region. If the number of regions contained in the image is less than or equal to the third threshold, or, the region The maximum number of grids contained in the grid is less than or equal to the fourth threshold, indicating that the matching feature point pairs contained in the image pair are concentrated in a small area, and the matching relationship of the image pair is deleted. If the maximum number of grids contained in the area is greater than the fourth threshold, the matching relationship of the image pair is retained. Finally, the image pairs retaining the matching relationship are output, and subsequent processing continues. This scheme does not need to analyze image feature information, but identifies small-area repetitive textures through the slope or distribution of matching feature point pairs contained in the image. Therefore, the method is insensitive to the input image and improves the robustness of the method.

Figure 4 is a flow chart of another repetitive texture identification method provided by an embodiment of the present application. This embodiment determines whether there is a repetitive texture in an image pair by using the slope of the connection line between the matching feature point pairs in the image pair. The method is applied to electronic equipment, and the electronic equipment includes a feature extraction module, a feature matching module and a first filtering module.

As shown in Figure 4, the method may include the following steps:

S31. The feature extraction module obtains multiple views of the three-dimensional reconstructed object.

S32, the feature extraction module extracts the feature points of each view to obtain a feature map.

S33. The feature extraction module transfers all feature maps to the feature matching module.

S34: The feature matching module performs feature matching on all feature maps to obtain image pairs with matching relationships, and transfers all image pairs with matching relationships to the first filtering module.

S35. The first filtering module obtains feature point matching information of the image pair.

S36: Determine whether the number of matching feature points contained in one image in the image pair is greater than or equal to the first threshold; if so, perform S37; if not, perform S310.

S37, splice the image pairs horizontally, and calculate the slope and median slope of the lines connecting all matching feature point pairs.

S38: Count the number of connected lines in the same image pair whose deviation between the slope of the connecting line and the median slope is greater than or equal to the first preset value.

S39: Determine whether the proportion of qualified connections is greater than or equal to the second preset value; if yes, execute S310; if not, execute S311.

S310: Delete the matching relationship of the image pair.

In this embodiment, the implementation process of S31 to S310 is the same as the implementation process of S11 to S20 in the embodiment shown in FIG. 1 , and will not be described again here.

S311, retain the matching relationship of the image pair.

In this embodiment, if the first filtering module determines that the proportion of qualified connections is less than the second preset value, it indicates that the proportion of feature points that are not in the main direction of feature matching is small and can be ignored, and then it can be determined that the matching relationship of the image pair is correct. Keep the matching relationship between the image pairs.

S312. The first filtering module determines whether there are unprocessed image pairs. If yes, return to S35; if no, execute S313.

In this embodiment, the process of the first filtering module determining whether there is an unprocessed image pair is the same as the implementation of S28 in the embodiment shown in FIG. 1, and will not be described again here.

S313. The first filtering module outputs image pairs that retain matching relationships.

The implementation process of each step in this embodiment is basically the same as the implementation process of the relevant steps in the embodiment shown in Figure 1, and will not be described in detail in this embodiment.

The repetitive texture recognition method provided in this embodiment horizontally splices two images corresponding to an image pair, calculates the slope of the connection line of the matching feature point pair in the image pair, and the median slope of all the connection lines in the image pair. Then count the proportion of lines where the deviation between the slope and the median slope is greater than or equal to the set value. If the proportion is greater than the corresponding preset value, it indicates that the image pair has a small area of repeated texture in the non-main direction of feature matching, that is, it is determined The matching relationship of the image pair is wrong, and the matching relationship of the image pair is deleted. It can be seen that this method does not need to analyze the content information of the feature points in the image pair. It only identifies repeated textures that are not in the main direction of feature matching through the slope of the feature point connection. Therefore, this method is not sensitive to the input image and improves the robustness of the method. sex.

Figure 5 shows a flow chart of yet another repetitive texture identification method provided by an embodiment of the present application. This embodiment analyzes the distribution of matching feature points in an image pair to identify whether there is a repetitive texture in the image pair. The method is applied to electronic equipment, and the electronic equipment includes a feature extraction module, a feature matching module and a second filtering module.

As shown in Figure 5, the method may include the following steps:

S41. The feature extraction module obtains multiple views of the three-dimensional reconstructed object.

S42, the feature extraction module extracts the feature points of each view to obtain a feature map.

S43, the feature extraction module transfers all feature maps to the feature matching module.

S44: The feature matching module performs feature matching on all feature maps to obtain image pairs with matching relationships, and transfers all image pairs with matching relationships to the first filtering module.

S45: The second filtering module obtains feature point matching information of the image pair.

S46: The second filtering module determines whether the number of matching feature points is less than or equal to the second threshold; if so, execute S47; if not, execute S412.

S47: Divide any image in the image pair into N*M grids, and count the corresponding grid distribution of feature points contained in the image.

S48: Connect adjacent grids containing matching feature points in the image into a region.

S49: Determine whether the number of areas contained in the image is less than the third threshold; if so, execute S411; if not, execute S410.

S410: Determine whether the maximum number of grids contained in all connected areas is less than the fourth threshold; if so, execute S411; if not, execute S412.

S411. Delete the matching relationship of the image pair.

S412, retain the matching relationship of the image pair.

S413, determine whether there are unprocessed image pairs. If there are unprocessed image pairs, return to execution S45 to continue processing the next image pair. If there are no unprocessed image pairs, perform S414.

S414. Output image pairs that retain matching relationships.

In this embodiment, the processes described in S45 to S414 are all run in the second filtering module.

The processes described in S41 to S414 in this embodiment are the same as the implementation processes of the same steps in the embodiment shown in FIG. 1 , and will not be described again here.

The repetitive texture identification method provided in this embodiment identifies whether there is a small area repeated texture in the image pair based on the distribution of matching feature point pairs contained in the image pair. If it exists, the matching relationship of the image pair is deleted, that is, matching relationship errors are filtered out. matching relationship between image pairs. This scheme does not need to analyze image feature information, but identifies repeated textures in small areas through the distribution of matching feature point pairs contained in the image. Therefore, the method is insensitive to the input image and improves the robustness of the method.

On the other hand, the present application also provides an electronic device suitable for the repeated texture recognition method provided by the present application.

FIG. 6 is a schematic structural diagram of an electronic device provided by an embodiment of the present application. For example, the electronic device may be an electronic device such as a server or a terminal device. Terminal devices may include mobile phones, tablet computers, desktops, laptops, notebook computers, ultra-mobile personal computers (UMPC), handheld computers, netbooks, personal digital assistants (Personal Digital Assistant, PDA), Wearable electronic devices, smart watches and other devices.

As shown in FIG. 6 , the electronic device may include a processor 101, a memory 102, a bus 103, and a communication interface 104. The number of processors 101 may be 1 to N, and N is an integer greater than 1.

The processor 101 and the memory 102 communicate with each other through the bus 103 . The processor 101 can communicate with external devices through the bus 103 and the communication interface 104. For example, the communication interface 104 includes a sending unit and a receiving unit. The communication interface 104 receives data sent by the peripheral device through the receiving unit, and the data is transmitted to the processor 101 via the bus 103 . The data sent by the processor 101 is transmitted to the communication interface 104 via the bus 103, and the communication interface 104 sends it to the peripheral device through the sending unit. The processor 101 is used to call program instructions in the memory 102 to execute the repetitive texture identification method shown in FIG. 1, FIG. 4 or FIG. 5.

Through the above description of the embodiments, those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above functional modules is used as an example. In actual applications, the above functions can be allocated as needed. It is completed by different functional modules, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. For the specific working processes of the systems, devices and units described above, reference can be made to the corresponding processes in the foregoing method embodiments, which will not be described again here.

In the several embodiments provided in this embodiment, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of modules or units is only a logical function division. In actual implementation, there may be other division methods, for example, multiple units or components may be The combination can either be integrated into another system, or some features can be ignored, or not implemented. On the other hand, the coupling or direct coupling or communication connection between each other shown or discussed may be through some interfaces, and the indirect coupling or communication connection of the devices or units may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of this embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated units can be implemented in the form of hardware or software functional units.

If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this embodiment is essentially or contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to cause a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment. The aforementioned storage media include: flash memory, mobile hard disk, read-only memory, random access memory, magnetic disk or optical disk and other media that can store program codes.

The above are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application shall be covered by the protection scope of the present application. . Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims (14)

1. A method of repeating texture recognition, for use with an electronic device, the method comprising:

acquiring feature point matching information contained in at least two images with a matching relationship, wherein the feature point matching information comprises information of matched feature points contained in the at least two images;

transversely splicing the at least two images, and calculating the slopes of all the matching characteristic point connecting lines in the at least two images;

counting the number of connecting lines meeting a first preset condition in the at least two images, wherein the first preset condition comprises that the deviation between the slope and the standard value of the slope is larger than or equal to a first preset value;

determining that the number of the connecting lines meets a second preset condition, wherein the second preset condition comprises that the proportion of the connecting lines meeting the first preset condition is larger than or equal to a second preset value;

Determining that the at least two images contain small region repeating textures.

2. The method of claim 1, wherein prior to laterally stitching the at least two images, the method further comprises:

counting the number of matching feature points contained in any one of the at least two images; if the number of the matching feature points is greater than or equal to a first threshold value, executing the transverse stitching of the at least two images;

and if the number of the matching feature points is smaller than the first threshold value, determining that the at least two images have small-area repeated textures.

3. The method according to claim 1, wherein the slope criterion value determination process includes:

and calculating slope median values corresponding to all connecting lines included in the image pairs, and determining the slope median values as the slope standard values.

4. A method according to any one of claims 1-3, wherein the method further comprises: and deleting the matching relation between at least two images containing the small-area repeated textures.

5. The method according to any one of claims 1-4, further comprising:

after determining that the number of connecting lines meeting the first preset condition in the at least two images does not meet the second preset condition, counting the distribution range of matching feature points contained in any image in the at least two images;

And if the distribution range is smaller than or equal to a preset range, determining that the at least two images contain small-area repeated textures.

6. The method of claim 5, wherein said counting the distribution range of the matching feature points included in any one of the at least two images comprises:

dividing any one image of the at least two images into a plurality of grids, and counting the number of grids containing the matching characteristic points in any one image;

the grids which contain the matching characteristic points and are adjacent in position are communicated into a communication area;

and determining the distribution range of the matching feature points based on parameters of connected areas contained in any image, wherein the parameters comprise at least one of the number and the area of the areas.

7. The method of claim 6, wherein determining the distribution range of the matching feature points based on the parameters of the region included in the arbitrary image comprises:

if the number of all the connected areas contained in any one image is smaller than or equal to a third threshold value, determining that the distribution range of the matched feature points is smaller than the preset range;

if the number of all the communication areas contained in any one image is larger than the third threshold value, judging whether the total area of all the communication areas in any one image is smaller than or equal to a fourth threshold value;

If the total area is smaller than or equal to the fourth threshold value, determining that the at least two images contain small-area repeated textures;

and if the total area is larger than the fourth threshold value, determining that the matching relationship of the at least two images is correct.

8. The method according to any one of claims 5-7, wherein prior to said counting the distribution range of matching feature points comprised by any one of said at least two image pairs, the method further comprises:

counting the number of matching feature points included in any one of the at least two images;

if the number of the matching feature points is smaller than or equal to a second threshold value, executing the step of counting the distribution range of the matching feature points contained in any image in the at least two image pairs;

and if the number of the matching feature points is larger than the second threshold value, determining that the matching relationship of the at least two images is correct.

9. A method of repeating texture recognition, for use with an electronic device, the method comprising:

acquiring feature point matching information contained in at least two image pairs with a matching relationship, wherein the feature point matching information comprises information of matched feature points contained in the at least two images;

Counting the distribution range of the matching feature points contained in any image of the at least two images;

and determining that at least two images with the distribution range smaller than or equal to a preset range contain small-area repeated textures.

10. The method of claim 9, wherein said counting the distribution range of matching feature points contained in any one of the pair of images comprises:

dividing any one image of the at least two images into a plurality of grids, and counting the number of grids containing the matching characteristic points in any one image;

the grids which contain the matching characteristic points and are adjacent in position are communicated into a communication area;

and determining the distribution range of the matching feature points based on parameters of connected areas contained in any image, wherein the parameters comprise at least one of the number and the area of the areas.

11. The method according to claim 10, wherein determining the distribution range of the matching feature points based on the parameters of the region included in the arbitrary image includes:

if the number of all the connected areas contained in any one image is smaller than or equal to a third threshold value, determining that the distribution range of the matched feature points is smaller than the preset range;

If the number of all the communication areas contained in any one image is larger than the third threshold value, judging whether the total area of all the communication areas in any one image is smaller than or equal to a fourth threshold value;

if the total area is smaller than or equal to the fourth threshold value, determining that the at least two images contain small-area repeated textures;

and if the total area is larger than the fourth threshold value, determining that the matching relationship of the at least two images is correct.

12. The method according to any one of claims 9-11, wherein prior to said counting the distribution range of matching feature points comprised by any one of said at least two image pairs, the method further comprises:

counting the number of matching feature points included in any one of the at least two images;

if the number of the matching feature points is smaller than or equal to a second threshold value, executing the step of counting the distribution range of the matching feature points contained in any image in the at least two image pairs;

and if the number of the matching feature points is larger than the second threshold value, determining that the matching relationship of the at least two images is correct.

13. An electronic device, the electronic device comprising: one or more processors, memory, and a touch screen; the memory is used for storing program codes; the processor is configured to execute the program code to cause the electronic device to implement the repetitive texture recognition method as defined in any one of claims 1 to 12.

14. A computer readable storage medium having instructions stored thereon which, when executed on an electronic device, cause the electronic device to perform the repetitive texture recognition method as defined in any one of claims 1 to 12.