CN117975179A - Image processing method, device and electronic device - Google Patents

Abstract

The embodiment of the invention discloses an image processing method, an image processing device and electronic equipment; an image processing method, comprising: determining at least two feature points corresponding to an image to be processed; determining a target feature screening scale according to the first feature screening scale and/or the second feature screening scale; the first feature screening scale is related to the image information to be processed and the feature point information, and the second feature screening scale is related to the optical parameters of the image acquisition equipment; and screening the at least two feature points according to the target feature screening scale to obtain screened feature points. The image processing method, the device and the electronic equipment are used for realizing reasonable screening of high-quality characteristic points with uniform spatial distribution.

Description

Image processing method, device and electronic device

Technical Field

The present disclosure relates to the field of image processing technology, and in particular to an image processing method, device and electronic device.

Background technique

Current vision-based positioning and navigation technologies use the matching relationship between images to estimate motion states. The matching effect between images is not only affected by feature detection algorithms and matching methods, but also by the quality of feature points. Extracting high-quality feature points helps improve the accuracy, efficiency, and stability of image matching.

One of the conditions for high-quality feature points is that their distribution needs to be as uniform as possible, that is, high-quality feature points cannot be concentrated in a certain area of the image, and a certain distance must be maintained between points to avoid dense distribution. Therefore, after feature extraction, it is necessary to screen the feature points to retain high-quality feature points with a relatively uniform distribution.

Summary of the invention

This disclosure section is provided to introduce concepts in a brief form, which will be described in detail in the detailed description section below. This disclosure section is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

The embodiments of the present disclosure provide an image processing method, an apparatus, and an electronic device for realizing reasonable screening of high-quality feature points with uniform spatial distribution.

In a first aspect, an embodiment of the present disclosure provides an image processing method, comprising: determining at least two feature points corresponding to an image to be processed; determining a target feature screening scale based on a first feature screening scale and/or a second feature screening scale; wherein the first feature screening scale is related to the image information to be processed and the feature point information, and the second feature screening scale is related to the optical parameters of an image acquisition device; screening the at least two feature points according to the target feature screening scale to obtain screened feature points.

In a second aspect, an embodiment of the present disclosure provides another image processing method, comprising: determining at least two feature points and a feature screening scale corresponding to the image to be processed; using the feature screening scale as the standard deviation to generate a Gaussian function; and screening the at least two feature points multiple times according to the Gaussian function.

In a third aspect, an embodiment of the present disclosure provides an image processing device, comprising: a determination unit, used to determine at least two feature points corresponding to an image to be processed; the determination unit is also used to determine a target feature screening scale based on a first feature screening scale and/or a second feature screening scale; wherein the first feature screening scale is related to the image information to be processed and the feature point information, and the second feature screening scale is related to the optical parameters of an image acquisition device; a screening unit, used to screen the at least two feature points according to the target feature screening scale to obtain screened feature points.

In a fourth aspect, an embodiment of the present disclosure provides another image processing device, comprising: a determination unit, used to determine at least two feature points and a feature screening scale corresponding to an image to be processed; a generation unit, used to generate a Gaussian function using the feature screening scale as a standard deviation; and a screening unit, used to perform multiple screening on the at least two feature points according to the Gaussian function.

In a fifth aspect, an embodiment of the present disclosure provides an electronic device, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the image processing method as described in the first aspect or the second aspect.

In a sixth aspect, an embodiment of the present disclosure provides a computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the image processing method as described in the first aspect or the second aspect.

The image processing method, device and electronic device provided by the embodiments of the present disclosure take into account that the image itself is restricted by the resolution, and the resolution is related to the optical parameters of the image acquisition device, so that the optical parameters of the image acquisition device affect the distance between the feature points. Based on this, the feature screening scale related to the image information and the feature point information, and/or the feature screening scale related to the optical parameters of the image acquisition device are used as prior information to determine the screening scale of the feature points, and the feature points are screened based on the screening scale of the feature points; so that the screening of the feature points avoids the influence caused by the resolution, and then realizes the reasonable screening of high-quality feature points with uniform spatial distribution.

Furthermore, at least two feature points are screened multiple times by using a Gaussian function generated based on the feature screening scale to achieve iterative screening of feature points and retain the distribution of feature points as evenly as possible; feature points with sufficiently large responses and slightly close distances can also be retained to avoid being roughly screened out, thereby reasonably screening out high-quality feature points with uniform spatial distribution.

BRIEF DESCRIPTION OF THE DRAWINGS

The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

FIG1 is a flow chart of an embodiment of an image processing method according to the present disclosure;

FIG2 is a flow chart of another embodiment of an image processing method according to the present disclosure;

FIG3 is a flow chart of another embodiment of an image processing method according to the present disclosure;

FIG4 is a flow chart of another embodiment of an image processing method according to the present disclosure;

FIG5 is a schematic structural diagram of an image processing device according to an embodiment of the present disclosure;

FIG6 is a schematic structural diagram of another embodiment of an image processing device according to the present disclosure;

FIG. 7 is an exemplary system architecture in which the image processing method according to an embodiment of the present disclosure may be applied;

FIG8 is a schematic diagram of a basic structure of an electronic device provided according to an embodiment of the present disclosure.

Detailed ways

Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and/or in parallel. In addition, the method embodiments may include additional steps and/or omit the steps shown. The scope of the present disclosure is not limited in this respect.

The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

The technical solution disclosed in the present invention can be applied to various image processing scenarios involving feature point extraction. In some image processing scenarios, it may be necessary to match images, such as vision-based positioning scenarios, navigation scenarios, etc., and the matching between images requires the use of image feature points, so that the image feature points are obtained by processing the image. In other image processing scenarios, it may be necessary to identify the image, such as the identification of the target object, etc. In order to achieve more accurate identification of the target object, it is necessary to extract feature points from the image.

Regardless of the image processing scenario, when extracting feature points, feature points need to be screened to ensure that the final feature points are high-quality feature points.

In the related art, feature point extraction algorithms are used to extract feature points in the entire image. Feature point extraction algorithms include, for example, Harris (a corner point extraction algorithm proposed by Harris), Fast (Features from accelerated segment test, corner point detection algorithm), and other feature point extraction algorithms.

In these related technologies, depending on the image content, the detected feature points may be concentrated in a certain area of the image. Furthermore, in related application scenarios, if the distance between feature points is close, the calculated descriptors may have a high similarity, which makes it very easy to search for mismatched feature points in the image to be matched; on the other hand, too much concentration of feature points will also lead to fewer feature points in other parts of the whole image, resulting in greater deviation in the solution of the overall pose of the whole image.

Therefore, the related technology extracts feature points in the entire image, resulting in uneven distribution of the extracted feature points, and cannot achieve reasonable extraction or screening of high-quality feature points with uniform spatial distribution.

Based on this, the feature point screening scheme provided by the embodiment of the present disclosure, on the one hand, takes into account that the image itself is restricted by the resolution, and the resolution is related to the optical parameters of the image acquisition device, so that the optical parameters of the image acquisition device affect the distance between the feature points. Then, the feature screening scale related to the image information and the feature point information, and the feature screening scale related to the optical parameters of the image acquisition device are used as prior information to determine the screening scale of the feature points, and the feature points are screened based on the screening scale of the feature points; so that the screening of feature points avoids the influence caused by the resolution, and then realizes the reasonable screening of high-quality feature points with uniform spatial distribution.

On the other hand, based on the Gaussian function generated by the feature screening scale, at least two feature points are screened multiple times to achieve iterative screening of feature points and retain the distribution of feature points as evenly as possible; feature points with sufficiently large responses and slightly closer distances can also be retained to avoid being roughly screened out, thereby reasonably screening out high-quality feature points with uniform spatial distribution.

Please refer to FIG. 1, which shows a flow chart of an embodiment of an image processing method according to the present disclosure. The image processing method can be applied to a terminal device. As shown in FIG. 1, the image processing method includes the following steps:

Step 101: determine at least two feature points corresponding to the image to be processed.

In some application scenarios, the image to be processed is an image that needs to be matched. In other application scenarios, the image to be processed is an image that needs to be identified as a target object.

In some embodiments, if the number of images to be processed is 1, then the image to be processed is processed accordingly. In other embodiments, if the number of images to be processed is greater than 1, then each image to be processed is processed separately according to the same processing method.

In some embodiments, the multiple feature points in step 101 are original feature points extracted from the image to be processed, that is, feature points that have not been processed. In other embodiments, the multiple feature points in step 101 are feature points that have been preliminarily screened. For example, 100 feature points are extracted from the image to be processed, and 10 of the 100 feature points are deleted by preliminarily screening the 100 feature points. Then, the preliminarily screened feature points are the remaining 90 feature points.

Therefore, as an optional implementation, step 101 includes: obtaining at least two initial feature points extracted from the image to be processed; determining feature point response values corresponding to the at least two initial feature points; performing preliminary screening of the at least two initial feature points according to the feature point response values corresponding to the at least two initial feature points, and determining the preliminary screened feature points as at least two feature points of the image to be processed.

In some embodiments, at least two initial feature points are extracted based on a preset feature extraction algorithm, including but not limited to FAST, Harris, SIFT (Scale-invariant feature transform), SURF (Speeded Up Robust Features), and other algorithms.

Furthermore, the feature point response values corresponding to the at least two initial feature points may be response values corresponding to the corresponding feature extraction algorithms. In other embodiments, no matter which feature extraction algorithm is used, the feature point response value uses the Harris corner point response value.

In some embodiments, a feature point response value threshold is preset, and feature points whose feature point response values are less than or equal to the feature point response value threshold are deleted; feature points whose feature point response values are greater than the feature point response value threshold are retained. The feature point response value threshold is determined according to the corresponding feature point response value determination method, for example, for the Harris corner point response value, the corresponding Harris corner point response threshold is set, and the specific value is not limited here.

Therefore, as an optional implementation, the feature point response value is a Harris corner point response value. Compared with other response values, the Harris corner point response value can better reflect the robustness and quality of the feature points, so as to improve the quality of the feature points finally screened out.

In some application scenarios, in order to improve the extraction efficiency of feature points, before performing feature extraction based on the image to be processed, the image is first divided into blocks, and then feature extraction is performed based on the divided image.

Therefore, as an optional implementation, the image to be processed is an image that has been processed by image block processing, and the at least two initial feature points are feature points extracted based on at least two image blocks.

In some embodiments, at least two image blocks are sorted, for example, at least two image blocks are numbered starting from 0; thus, when extracting feature points, feature points are extracted in sequence according to the numbering order of the image blocks to improve the efficiency and accuracy of feature point extraction and avoid duplication or erroneous extraction.

Step 102: determining a target feature screening scale according to the first feature screening scale and/or the second feature screening scale.

The first feature screening scale is related to the image information to be processed and the feature point information, and the second feature screening scale is related to the optical parameters of the image acquisition device.

In some embodiments, the image information to be processed includes: the size of the image to be processed, and the feature point information includes: the number of preset feature points and the number of at least two feature points.

The preset number of feature points may be a preset threshold value of the number of feature points, which is preset in combination with a specific application scenario, and no specific value is limited here.

Therefore, the image processing method further includes: determining a first feature screening scale according to the size of the image to be processed, the number of preset feature points and the number of at least two feature points.

In some embodiments, the size of the image to be processed may include: the width and height of the image to be processed.

In some embodiments, the first feature screening metric is expressed as: Wherein, w is the width of the image to be processed, h is the height of the image to be processed, f _p is the number of preset feature points, and f ₀ is the number of at least two feature points.

In some embodiments, the image acquisition device is a camera, and the optical parameters of the image acquisition device include: point spread function and half-height width; the optical parameters can be provided by the camera module manufacturer. In addition, the optical parameters can be understood as feature point screening prior information given from the perspective of optical camera resolution.

Furthermore, the image processing method further includes: determining a second feature screening scale according to the point spread function and the half-height width.

In some embodiments, the second feature screening scale can be expressed as: FWHM (PSF); wherein PSF stands for pointspread function, which represents point spread function, and FWHM stands for Full width half maximum, which represents half-height width.

It can be understood that if the target feature screening scale is determined according to the first feature screening scale, only the first feature screening scale needs to be determined. If the target feature screening scale is determined according to the second feature screening scale, only the second feature screening scale needs to be determined. If the target feature screening scale is determined according to the first feature screening scale and the second feature screening scale, the first feature screening scale and the second feature screening scale need to be determined.

Furthermore, the target feature screening scale may be determined by combining the first feature screening scale and/or the second feature screening scale.

As an optional implementation, step 102 includes: determining a maximum feature screening scale between the first feature screening scale and the second feature screening scale; and determining a target feature screening scale according to the maximum feature screening scale.

In some embodiments, the target feature screening dimension is equal to the maximum feature screening dimension.

In other embodiments, the target feature screening scale is the product of the maximum feature screening scale and the preset scale gain. The preset scale gain can be reasonably set according to different application scenarios. For example, it can be set according to experience or demand, and the basis can be the time consumption of the algorithm, the memory usage, etc. It can be understood that the larger the scale gain, the faster the feature point screening speed; similarly, the smaller the scale gain, the slower the feature point screening speed.

As another optional implementation, step 102 includes: determining the first feature screening scale as the target feature screening scale. Alternatively, the target feature screening scale is the product of the first feature screening scale and a preset scale gain. The preset scale gain refers to the introduction of the above embodiment.

As another optional implementation, step 102 includes: determining the second feature screening scale as the target feature screening scale. Alternatively, the target feature screening scale is the product of the second feature screening scale and a preset scale gain. The preset scale gain refers to the introduction of the above embodiment.

Step 103 , screening at least two feature points according to the target feature screening scale to obtain screened feature points.

It can be understood that the filtered feature point is a feature point retained after filtering among at least two feature points.

In the embodiments of the present disclosure, it is considered that the image itself is restricted by the resolution, and the resolution is related to the optical parameters of the image acquisition device, so that the optical parameters of the image acquisition device affect the distance between the feature points. Based on this, the feature screening scale related to the image information and the feature point information, and/or the feature screening scale related to the optical parameters of the image acquisition device are used as prior information to determine the screening scale of the feature points, and the feature points are screened based on the screening scale of the feature points; so that the screening of the feature points avoids the influence caused by the resolution, and then realizes the reasonable screening of high-quality feature points with uniform spatial distribution.

In some embodiments, the distance between the target feature point and the feature points within the neighborhood may be calculated, and then the feature points whose distance is less than the target feature screening scale may be deleted, and the feature points whose distance is greater than the target feature screening scale may be retained.

The distance between feature points may be Manhattan distance or Euclidean distance, etc. Correspondingly, when calculating the distance between feature points, the corresponding distance calculation method is used based on the coordinates of the feature points, which is not limited in the disclosed embodiments.

The target feature point may be the feature point with the largest feature point response value among the currently remaining feature points; then, before each screening, the feature point with the largest feature point response value among the currently remaining feature points is first determined, and then feature screening is performed based on the feature point and the target feature screening scale. The feature point response value of the feature point can be determined before screening or during the screening process, and it can be a Harris corner point response value.

In other embodiments, iterative screening can also be implemented using a target feature screening scale.

Therefore, as an optional implementation, step 103 includes: using the target feature screening scale as the standard deviation to generate a Gaussian function; and screening at least two feature points multiple times according to the Gaussian function.

In some embodiments, the Gaussian function may be a two-dimensional Gaussian function, which is a symmetric two-dimensional Gaussian function with a mean of 0, and the standard deviation is required.

In some embodiments, after the Gaussian function is generated, maximum value normalization processing may be performed, that is, the capping of the Gaussian function may be subjected to maximum value normalization processing to facilitate subsequent processing.

Therefore, the size of the two-dimensional Gaussian function can generally be greater than or equal to 2 times the standard deviation plus 1. For example, assuming the standard deviation is 3, the Gaussian window size can be 7*7, and the Gaussian window is used as prior information for the next step of feature screening.

It can be understood that the related technology of Gaussian function can refer to the mature technology in the field and will not be introduced in detail here.

Furthermore, as an optional implementation, at least two feature points are screened multiple times according to a Gaussian function, including: obtaining a feature point response value matrix corresponding to at least two feature points, the feature point response value matrix including feature point response values of each feature point; determining a Gaussian matrix according to the Gaussian function; and screening at least two feature points multiple times according to the feature point response value matrix, the Gaussian matrix and a preset gain.

Among them, the feature point response value matrix corresponding to at least two feature points can also be called a feature point response map, which includes the feature point response values of each feature point, and each feature point is represented by the horizontal and vertical coordinates of the pixel of the feature point, which is a matrix with the same size as the image to be processed.

In some embodiments, the Gaussian matrix, ie, the Gaussian window introduced in the aforementioned embodiments, is also a matrix of fixed size.

In some embodiments, the preset gain may be configured according to the noise level. For example, the greater the noise level, the greater the gain; and the lower the noise level, the smaller the gain.

In some embodiments, the noise level may be determined based on noise information, which may be external a priori information, and the specific acquisition method is not limited herein.

Furthermore, according to the feature point response value matrix, the Gaussian matrix and the preset gain, multiple iterative screenings may be performed on at least two feature points.

As an optional implementation, any one of the multiple screening processes includes: determining a neighborhood feature point response value matrix based on the target feature point, the number of rows and columns of the Gaussian matrix, and the feature point response value matrix; the target feature point is the feature point with the largest feature point response value among the feature points currently to be screened, the number of rows and columns of the neighborhood feature point response value matrix is the same as the number of rows and columns of the Gaussian matrix, and the neighborhood feature point response value matrix includes the feature point response values of the neighborhood feature points of the target feature point; subtracting the neighborhood feature point response value matrix from the product matrix of the Gaussian matrix and the preset gain to obtain a residual matrix; determining the target matrix element in the residual matrix, and setting the target matrix element to a preset value to delete the feature point corresponding to the target matrix element from the image to be processed; the target matrix element is less than the preset value.

In this implementation, in any screening process, it is necessary to first determine the target feature point, which is the feature point with the largest feature point response value among the feature points currently to be screened.

Then, according to the number of rows and columns of the Gaussian matrix, the neighborhood feature point response value matrix corresponding to the target feature point is found in the feature point response value matrix. For example, assuming that the Gaussian matrix is a 7*7 matrix, then, with the feature point response value as the center, a 7*7 neighborhood feature point response value matrix is extracted from the feature point response value matrix.

Next, feature points can be screened based on the neighborhood feature point response value matrix. In the embodiment of the present disclosure, direct screening is not adopted, but indirect screening is adopted. Specifically, the neighborhood feature point response value matrix is subtracted from the product matrix of the Gaussian matrix and the preset gain to obtain a residual matrix.

For example, assuming that the Gaussian matrix is A, the neighborhood feature point response value matrix is B, and the preset gain is G, the residual matrix is B-G*A. The residual matrix corresponds to the understanding of the aforementioned feature point response value matrix and can also be called a residual map.

Furthermore, based on the residual matrix, the values of the matrix elements therein should meet the conditions of the preset values. Thus, the target matrix elements whose matrix elements are less than the preset values are first determined, and their values are set to the preset values. And, if the matrix element is the preset value, it means that the feature points corresponding to the matrix element have been deleted.

In some embodiments, the preset value may be 0; or other constraint values, which are not limited here.

Since multiple iterations of screening are required, relevant information may be recorded after each iteration of screening to determine whether to stop the iteration or continue the iteration process based on the relevant information.

Therefore, as an optional implementation, any screening process also includes: in response to detecting that the target feature point does not belong to the feature points in the preset feature point statistical table, recording the target feature point in the preset feature point statistical table, and increasing the number of feature points by 1; wherein, during the first screening, the number of feature points is the preset number, and the preset feature point statistical table is empty; in response to detecting that the number of feature points reaches the preset number of feature points, determining the currently remaining feature points as the feature points finally screened out.

In this embodiment, a feature point statistics table is preset, which can be used to record the target feature points that have been used for screening, so that the number of feature points can be counted based on the feature point statistics table to determine whether the iteration is completed based on the number of feature points.

In the feature point statistics table, the coordinates of each feature point can be recorded. When detecting whether a target feature point belongs to the feature point statistics table, the coordinates of each feature point in the feature point statistics table are compared with the coordinates of the target feature point to achieve detection.

The number of preset feature points can be configured in combination with specific application scenarios, and no specific value is limited here.

Correspondingly, if the number of detected feature points does not reach the preset number of feature points, the next screening process will continue.

In some embodiments, in addition to judging from the number of feature points, judgment may also be made in combination with the number of iterations.

As an optional implementation, the image processing method further includes: in response to detecting that the number of screening times of the feature point is greater than or equal to the preset number of screening times, displaying prompt information indicating that the preset gain is unreasonable.

The preset number of screening times may be a larger value, so that the number of screening times generally does not reach the preset number of screening times, which is used to prevent an infinite iterative loop.

In this embodiment, after each iteration, the number of screenings may be increased by 1; and it is determined whether the number of screenings of the current feature point is greater than or equal to the preset number of screenings; if so, a prompt message indicating that the preset gain is unreasonable is displayed; if not, the iteration is continued.

In some embodiments, the prompt message may be, for example: "The preset gain is unreasonable and needs to be adjusted."

The iterative screening method of feature points provided in the embodiment of the present disclosure does not directly screen feature points in a large range according to the target feature screening scale, but uses matrix operations to perform indirect screening in a small range, thereby retaining the distribution of feature points as uniform as possible; feature points with sufficiently large responses and slightly close distances can also be retained to avoid being roughly screened out, thereby reasonably screening out high-quality feature points with uniform spatial distribution.

Further reference is made to FIG. 2 and FIG. 3 , which are optional processes for the aforementioned feature screening scheme in practical application.

In Figure 2, for the input image, the feature point extraction algorithm is first used to realize feature extraction; then the feature point response value is calculated based on the extracted feature points. Next, the feature points whose feature point response values are greater than the preset response value are retained to realize the preliminary screening of the feature points. Next, the number of feature points is counted, and the feature point screening scale is determined by combining the number of feature points, the size information of the input image, the preset upper limit of the feature points, and the optical parameters of the image acquisition device. Then, the feature point screening scale is used as the standard deviation to generate a two-dimensional Gaussian function, and the maximum value normalization is performed. Furthermore, the two-dimensional Gaussian function is used to realize feature screening, and the image and the screened feature points are output.

FIG3 shows an iterative screening process of feature points. The input of the iterative screening process is: a feature screening prior matrix, i.e., a Gaussian matrix; a feature point response value matrix, including coordinates and response values; a preset processing gain and a maximum number of iterations K.

First, based on the feature point response value matrix and the preset processing gain, the feature point with the largest current feature point response value is found, and the residual map is calculated based on the feature point and the feature screening prior matrix, that is, feature point screening is performed.

Next, the number of iterations K is incremented by 1, and a determination is made as to whether the number of iterations is less than the maximum number of iterations Km. If so, the feature point statistics table is updated; if not, it is indicated that the preset processing gain is unreasonable and needs to be adjusted.

Furthermore, when updating the feature point statistics table, it is determined whether the feature point coordinates already exist in the feature point statistics table. If so, the next iteration is continued; if not, the number of feature points is increased by 1, and it is determined whether the number of feature points is less than the preset number of feature points. If so, the next iteration is continued; if not, the filtered feature point coordinates are output.

Please refer to FIG4 , which is a flowchart of another implementation of an image processing method provided in an embodiment of the present disclosure. The image processing method includes:

Step 401: Determine at least two feature points and a feature screening scale corresponding to the image to be processed.

The feature screening scale may be determined with reference to the process shown in FIG. 1 ; it may also be determined using other implementations, for example, based on noise information, using a preset feature screening scale, etc., which is not limited here.

The process of determining at least two feature points has been described in the foregoing embodiments and will not be repeated here.

Step 402, using the feature screening scale as the standard deviation to generate a Gaussian function.

Step 403: Screen at least two feature points multiple times according to the Gaussian function.

In the embodiments of the present disclosure, the screening of feature points can be improved from two aspects: one is the improvement of the calculation of the feature screening scale, and the other is the improvement of the iterative screening process of the feature points. Moreover, these two aspects can be used in combination or separately. Therefore, the embodiments of the present disclosure can protect three schemes: one is a feature point screening scheme based on the calculation method of the improved feature point screening scale, the second is a feature point screening scheme based on the improved feature point iterative screening process, and the third is a feature point screening scheme based on the calculation method of the improved feature point screening scale and the improved feature point iterative screening process.

Therefore, the implementation of step 402 and step 403 may refer to the introduction in the aforementioned implementation.

Thus, step 403 may include: obtaining a feature point response value matrix corresponding to at least two feature points, the feature point response value matrix including feature point response values of each feature point; determining a Gaussian matrix according to a Gaussian function; and screening at least two feature points multiple times according to the feature point response value matrix, the Gaussian matrix and a preset gain.

And, any screening process in multiple screenings includes: determining a neighborhood feature point response value matrix based on the target feature point, the number of rows and columns of the Gaussian matrix and the feature point response value matrix; the target feature point is the feature point with the largest feature point response value among the feature points to be screened, the number of rows and columns of the neighborhood feature point response value matrix is the same as the number of rows and columns of the Gaussian matrix, and the neighborhood feature point response value matrix includes the feature point response values of the neighborhood feature points of the target feature point; subtracting the neighborhood feature point response value matrix from the product matrix of the Gaussian matrix and the preset gain to obtain a residual matrix; determining a target matrix element in the residual matrix, and setting the target matrix element to a preset value to delete the feature point corresponding to the target matrix element from the image to be processed; the target matrix element is less than the preset value.

Any screening process also includes: in response to detecting that the target feature point does not belong to the feature points in the preset feature point statistical table, recording the target feature point in the preset feature point statistical table, and increasing the number of feature points by 1; wherein, during the first screening, the number of feature points is the preset number, and the preset feature point statistical table is empty; in response to detecting that the number of feature points reaches the preset number of feature points, determining the currently remaining feature points as the feature points finally screened out.

The image processing method further includes: in response to detecting that the number of screening times of the feature point is greater than or equal to the preset number of screening times, displaying prompt information for indicating that the preset gain is unreasonable.

For the sake of brevity, the detailed implementation of each step of the process shown in FIG. 4 will not be repeated here.

Further referring to FIG. 5 , as an implementation of the method shown in FIG. 1 , the present disclosure provides an embodiment of an image processing device. The device embodiment corresponds to the image processing method embodiment shown in FIG. 1 , and the device can be specifically applied to various electronic devices.

As shown in FIG5 , the image processing device of this embodiment includes:

The first determination unit 501 is used to determine at least two feature points corresponding to the image to be processed; the first determination unit 501 is also used to determine the target feature screening scale according to the first feature screening scale and/or the second feature screening scale; wherein the first feature screening scale is related to the image information to be processed and the feature point information, and the second feature screening scale is related to the optical parameters of the image acquisition device; the first screening unit 502 is used to screen the at least two feature points according to the target feature screening scale to obtain the screened feature points.

In some embodiments, the image information to be processed includes: the size of the image to be processed; the feature point information includes: the number of preset feature points and the number of the at least two feature points; the first determination unit 501 is also used to determine the first feature screening scale based on the size of the image to be processed, the number of preset feature points and the number of the at least two feature points.

In some embodiments, the optical parameters include: a point spread function and a half-height width; the image processing method further includes: a first determination unit 501, which is also used to determine the second feature screening scale according to the point spread function and the half-height width.

In some embodiments, the first determining unit 501 is further configured to determine a maximum feature screening scale between the first feature screening scale and the second feature screening scale; and determine the target feature screening scale according to the maximum feature screening scale.

In some embodiments, the first screening unit 502 is further configured to use the target feature screening scale as a standard deviation to generate a Gaussian function; and perform multiple screening on the at least two feature points according to the Gaussian function.

In some embodiments, the first screening unit 502 is further used to: obtain a feature point response value matrix corresponding to the at least two feature points, the feature point response value matrix including feature point response values of each feature point; determine a Gaussian matrix according to the Gaussian function; and perform multiple screening on the at least two feature points according to the feature point response value matrix, the Gaussian matrix and a preset gain.

In some embodiments, any one of the multiple screening processes includes: determining a neighborhood feature point response value matrix based on the target feature point, the number of rows and columns of the Gaussian matrix, and the feature point response value matrix; the target feature point is the feature point with the largest feature point response value among the feature points currently to be screened, the number of rows and columns of the neighborhood feature point response value matrix is the same as the number of rows and columns of the Gaussian matrix, and the neighborhood feature point response value matrix includes feature point response values of neighborhood feature points of the target feature point; subtracting the neighborhood feature point response value matrix from the product matrix of the Gaussian matrix and the preset gain to obtain a residual matrix; determining a target matrix element in the residual matrix, and setting the target matrix element to a preset value to delete the feature point corresponding to the target matrix element from the image to be processed; the target matrix element is less than the preset value.

In some embodiments, any screening process also includes: in response to detecting that the target feature point does not belong to the feature points in the preset feature point statistical table, recording the target feature point in the preset feature point statistical table, and increasing the number of feature points by 1; wherein, during the first screening, the number of feature points is the preset number, and the preset feature point statistical table is empty; in response to detecting that the number of feature points reaches the preset number of feature points, determining the currently remaining feature points as the feature points finally screened out.

In some embodiments, the first screening unit 502 is further used for: in response to detecting that the number of screening times of the feature point is greater than or equal to the preset number of screening times, displaying prompt information indicating that the preset gain is unreasonable.

Please refer to FIG6 . As an implementation of the method shown in FIG4 , the present disclosure provides an embodiment of an image processing device. The device embodiment corresponds to the image processing method embodiment shown in FIG4 . The device can be specifically applied to various electronic devices.

As shown in Figure 6, the image processing device of this embodiment includes: a second determination unit 601, used to determine at least two feature points and a feature screening scale corresponding to the image to be processed; a generation unit 602, used to generate a Gaussian function using the feature screening scale as a standard deviation; and a second screening unit 603, used to perform multiple screening on the at least two feature points according to the Gaussian function.

In some embodiments, the second screening unit 603 is further used to: obtain a feature point response value matrix corresponding to the at least two feature points, the feature point response value matrix including feature point response values of each feature point; determine a Gaussian matrix according to the Gaussian function; and perform multiple screening on the at least two feature points according to the feature point response value matrix, the Gaussian matrix and a preset gain.

In some embodiments, any one of the multiple screening processes includes: determining a neighborhood feature point response value matrix based on the target feature point, the number of rows and columns of the Gaussian matrix, and the feature point response value matrix; the target feature point is the feature point with the largest feature point response value among the feature points currently to be screened, the number of rows and columns of the neighborhood feature point response value matrix is the same as the number of rows and columns of the Gaussian matrix, and the neighborhood feature point response value matrix includes feature point response values of neighborhood feature points of the target feature point; subtracting the neighborhood feature point response value matrix from the product matrix of the Gaussian matrix and the preset gain to obtain a residual matrix; determining a target matrix element in the residual matrix, and setting the target matrix element to a preset value to delete the feature point corresponding to the target matrix element from the image to be processed; the target matrix element is less than the preset value.

In some embodiments, any screening process also includes: in response to detecting that the target feature point does not belong to the feature points in the preset feature point statistical table, recording the target feature point in the preset feature point statistical table, and increasing the number of feature points by 1; wherein, during the first screening, the number of feature points is the preset number, and the preset feature point statistical table is empty; in response to detecting that the number of feature points reaches the preset number of feature points, determining the currently remaining feature points as the feature points finally screened out.

In some embodiments, the second screening unit 603 is further used for: in response to detecting that the number of screening times of the feature point is greater than or equal to the preset number of screening times, displaying prompt information indicating that the preset gain is unreasonable.

Please refer to FIG. 7 , which shows an exemplary system architecture in which the image processing method according to an embodiment of the present disclosure can be applied.

As shown in Fig. 7, the system architecture may include terminal devices 701, 702, 703, a network 704, and a server 707. The network 704 may be used to provide a medium for a communication link between the terminal devices 701, 702, 703 and the server 707. The network 704 may include various connection types, such as wired, wireless communication links, or optical fiber cables.

The terminal devices 701, 702, and 703 can interact with the server 707 through the network 704 to receive or send messages, etc. Various client applications, such as web browser applications, search applications, and news information applications, can be installed on the terminal devices 701, 702, and 703. The client applications in the terminal devices 701, 702, and 703 can receive user instructions and perform corresponding functions according to the user instructions, such as adding corresponding information to the information according to the user instructions.

Terminal devices 701, 702, and 703 can be hardware or software. When terminal devices 701, 702, and 703 are hardware, they can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III, Moving Picture Experts Group Audio Layer 3), MP4 (Moving Picture Experts Group Audio Layer IV, Moving Picture Experts Group Audio Layer 4) players, laptop computers, and desktop computers, etc. When terminal devices 701, 702, and 703 are software, they can be installed in the electronic devices listed above. It can be implemented as multiple software or software modules (for example, software or software modules used to provide distributed services), or it can be implemented as a single software or software module. No specific limitation is made here.

The server 707 may be a server that provides various services, such as receiving information acquisition requests sent by the terminal devices 701, 702, and 703, acquiring display information corresponding to the information acquisition requests in various ways according to the information acquisition requests, and sending the relevant data of the display information to the terminal devices 701, 702, and 703.

It should be noted that the image processing method provided in the embodiment of the present disclosure can be executed by a terminal device, and accordingly, the image processing apparatus can be set in the terminal devices 701, 702, and 703. In addition, the image processing method provided in the embodiment of the present disclosure can also be executed by a server 707, and accordingly, the image processing apparatus can be set in the server 707.

It should be understood that the number of terminal devices, networks and servers in Figure 7 is only illustrative. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

Referring to FIG8 below, it shows a schematic diagram of the structure of an electronic device (such as the terminal device or server in FIG7) suitable for implementing the embodiment of the present disclosure. The terminal device in the embodiment of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. The electronic device shown in FIG8 is only an example and should not bring any limitation to the functions and scope of use of the embodiment of the present disclosure.

As shown in FIG8 , the electronic device may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 801, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 802 or a program loaded from a storage device 708 into a random access memory (RAM) 803. In RAM 803, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 801, ROM 802, and RAM 803 are connected to each other via a bus 804. An input/output (I/O) interface 805 is also connected to the bus 804.

Typically, the following devices may be connected to the I/O interface 805: input devices 806 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 807 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 808 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 809. The communication device 809 may allow the electronic device to communicate wirelessly or wired with other devices to exchange data. Although FIG. 8 shows an electronic device with various devices, it should be understood that it is not required to implement or have all the devices shown. More or fewer devices may be implemented or have alternatively.

In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 809, or installed from the storage device 808, or installed from the ROM 802. When the computer program is executed by the processing device 801, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, 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 disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

The above-mentioned computer-readable medium carries one or more programs. When the above-mentioned one or more programs are executed by the electronic device, the electronic device: determines at least two feature points corresponding to the image to be processed; determines the target feature screening scale according to the first feature screening scale and/or the second feature screening scale; wherein the first feature screening scale is related to the image information to be processed and the feature point information, and the second feature screening scale is related to the optical parameters of the image acquisition device; and screens the at least two feature points according to the target feature screening scale to obtain the screened feature points.

Computer program code for performing the operations of the present disclosure may be written in one or more programming languages or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and/or flow chart, and the combination of the square boxes in the block diagram and/or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

The units involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of a unit does not limit the unit itself in some cases. For example, the first determination unit 501 may also be described as a "unit for determining at least two feature points corresponding to the image to be processed".

The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip (SOCs), complex programmable logic devices (CPLDs), and the like.

In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, 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 disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other.

In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

Although the subject matter has been described in language specific to structural features and/or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims.

Claims (18)

1. An image processing method, comprising:

Determining at least two feature points corresponding to an image to be processed;

Determining a target feature screening scale according to the first feature screening scale and/or the second feature screening scale; the first feature screening scale is related to the image information to be processed and the feature point information, and the second feature screening scale is related to the optical parameters of the image acquisition equipment;

And screening the at least two feature points according to the target feature screening scale to obtain screened feature points.

2. The image processing method according to claim 1, wherein the image information to be processed includes: the size of the image to be processed; the feature point information includes: presetting the number of feature points and the number of the at least two feature points; the image processing method further includes:

And determining the first feature screening scale according to the size of the image to be processed, the number of preset feature points and the number of the at least two feature points.

3. The image processing method according to claim 1, wherein the optical parameters include: a point spread function and a half-width; the image processing method further includes:

And determining the second feature screening scale according to the point spread function and the half-width.

4. The image processing method according to claim 1, wherein the determining a target feature screening scale according to the first feature screening scale and/or the second feature screening scale comprises:

determining a largest feature screening scale of the first feature screening scale and the second feature screening scale;

And determining the target feature screening scale according to the maximum feature screening scale.

5. The image processing method according to claim 1, wherein the screening the at least two feature points according to the target feature screening scale includes:

Taking the target feature screening scale as a standard deviation to generate a Gaussian function;

and screening the at least two characteristic points for multiple times according to the Gaussian function.

6. The image processing method according to claim 5, wherein the performing a plurality of filtering on the at least two feature points according to the gaussian function includes:

Acquiring a characteristic point response value matrix corresponding to the at least two characteristic points, wherein the characteristic point response value matrix comprises characteristic point response values of all the characteristic points;

determining a Gaussian matrix according to the Gaussian function;

And screening the at least two characteristic points for multiple times according to the characteristic point response value matrix, the Gaussian matrix and the preset gain.

7. The image processing method according to claim 6, wherein any one screening process among the plurality of screens includes:

Determining a neighborhood characteristic point response value matrix based on the target characteristic points, the row and column numbers of the Gaussian matrix and the characteristic point response value matrix; the target feature points are feature points with the largest feature point response values in the feature points to be screened currently, the number of rows and columns of the neighborhood feature point response value matrix is the same as that of the Gaussian matrix, and the neighborhood feature point response value matrix comprises feature point response values of neighborhood feature points of the target feature points;

The neighborhood characteristic point response value matrix is subjected to difference with a product matrix of the Gaussian matrix and the preset gain, and a residual error matrix is obtained;

determining target matrix elements in the residual matrix, and setting the target matrix elements as preset values so as to delete characteristic points corresponding to the target matrix elements from the image to be processed; the target matrix element is smaller than the preset value.

8. The image processing method according to claim 7, wherein the arbitrary one-time filtering process further includes:

in response to detecting that a target feature point does not belong to a feature point in a preset feature point statistics table, recording the target feature point into the preset feature point statistics table, and adding 1 to the number of feature points; the feature point statistical table is used for selecting feature points according to the number of the feature points, wherein the number of the feature points is preset in the first screening process, and the preset feature point statistical table is empty;

And determining the current residual characteristic points as the finally screened characteristic points in response to the fact that the number of the characteristic points reaches the preset number of the characteristic points.

9. The image processing method according to claim 6, characterized in that the image processing method further comprises:

And displaying prompt information for indicating that the preset gain is unreasonable in response to the detection that the screening times of the feature points are larger than or equal to the preset screening times.

10. An image processing method, comprising:

determining at least two feature points and feature screening scales corresponding to the image to be processed;

taking the feature screening scale as a standard deviation to generate a Gaussian function;

and screening the at least two characteristic points for multiple times according to the Gaussian function.

11. The image processing method according to claim 10, wherein the performing a plurality of filtering on the at least two feature points according to the gaussian function includes:

Acquiring a characteristic point response value matrix corresponding to the at least two characteristic points, wherein the characteristic point response value matrix comprises characteristic point response values of all the characteristic points;

determining a Gaussian matrix according to the Gaussian function;

And screening the at least two characteristic points for multiple times according to the characteristic point response value matrix, the Gaussian matrix and the preset gain.

12. The image processing method according to claim 11, wherein any one screening process of the plurality of screens includes:

Determining a neighborhood characteristic point response value matrix based on the target characteristic points, the row and column numbers of the Gaussian matrix and the characteristic point response value matrix; the target feature points are feature points with the largest feature point response values in the feature points to be screened currently, the number of rows and columns of the neighborhood feature point response value matrix is the same as that of the Gaussian matrix, and the neighborhood feature point response value matrix comprises feature point response values of neighborhood feature points of the target feature points;

The neighborhood characteristic point response value matrix is subjected to difference with a product matrix of the Gaussian matrix and the preset gain, and a residual error matrix is obtained;

determining target matrix elements in the residual matrix, and setting the target matrix elements as preset values so as to delete characteristic points corresponding to the target matrix elements from the image to be processed; the target matrix element is smaller than the preset value.

13. The image processing method according to claim 12, wherein the arbitrary one-time filtering process further includes:

in response to detecting that a target feature point does not belong to a feature point in a preset feature point statistics table, recording the target feature point into the preset feature point statistics table, and adding 1 to the number of feature points; the feature point statistical table is used for selecting feature points according to the number of the feature points, wherein the number of the feature points is preset in the first screening process, and the preset feature point statistical table is empty;

And determining the current residual characteristic points as the finally screened characteristic points in response to the fact that the number of the characteristic points reaches the preset number of the characteristic points.

14. The image processing method according to claim 10, characterized in that the image processing method further comprises:

And displaying prompt information for indicating that the preset gain is unreasonable in response to the detection that the screening times of the feature points are larger than or equal to the preset screening times.

15. An image processing apparatus, comprising:

the determining unit is used for determining at least two characteristic points corresponding to the image to be processed;

The determining unit is further used for determining a target feature screening scale according to the first feature screening scale and/or the second feature screening scale; the first feature screening scale is related to the image information to be processed and the feature point information, and the second feature screening scale is related to the optical parameters of the image acquisition equipment;

And the screening unit is used for screening the at least two characteristic points according to the target characteristic screening scale to obtain screened characteristic points.

16. An image processing apparatus, comprising:

the determining unit is used for determining at least two characteristic points and characteristic screening scales corresponding to the image to be processed;

the generation unit is used for generating a Gaussian function by taking the characteristic screening scale as a standard deviation;

And the screening unit is used for carrying out multiple screening on the at least two characteristic points according to the Gaussian function.

17. An electronic device, comprising:

one or more processors;

Storage means for storing one or more programs,

When executed by the one or more processors, causes the one or more processors to implement the method of any of claims 1-14.

18. A computer readable medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements the method according to any one of claims 1-14.