WO2019019595A1

WO2019019595A1 - Image matching method, electronic device method, apparatus, electronic device and medium

Info

Publication number: WO2019019595A1
Application number: PCT/CN2018/074841
Authority: WO
Inventors: 郑大明
Original assignee: 平安科技（深圳）有限公司
Priority date: 2017-07-27
Filing date: 2018-01-31
Publication date: 2019-01-31
Also published as: CN107688824A

Abstract

The present application is applicable to the field of image processing technology, and provided thereby are an image matching method, an electronic device method, an apparatus, an electronic device, and a medium. The method comprises: acquiring shape feature information and tone feature information of target objects in source images, and generating a feature descriptor that represents the shape feature information and the tone feature information of the target objects, the source images being at least one image; matching each image to be processed comprised in a target folder with the source images according to the feature descriptor; and issuing the images to be processed having a match level that is greater than a threshold as target images. By means of the described image matching method and electronic device, the shape feature information and the tone feature information of the target objects in the source images are acquired so as to match the source images and the images to be processed, thereby being able to reduce the time required for individually selecting all images of the target objects, reducing the operation by a user, and improving the user experience.

Description

Picture matching method and electronic device method, device, electronic device and medium

The present application claims priority to Chinese Patent Application No. PCT Application No. No. No. No. No. No. No. No. No. No. No. No. No.

Technical field

The present application belongs to the field of picture processing technologies, and in particular, to a picture matching method, device, electronic device, and medium.

Background technique

With the advancement of technology and economic development, more and more people are opening online stores on the Internet to conduct commodity transactions. In the display of goods, you need to first publish pictures of all angles of the product to the network. At present, when an image is published for an item, it is usually necessary to select all the pictures of the product one by one, so that all the pictures of the product are released, but the above process wastes the user's time, and the operation is cumbersome.

technical problem

In view of this, the embodiment of the present application provides a picture matching method and an electronic device, so as to solve the problem that the time of the user is wasted and the operation is cumbersome for all the pictures of the product are selected one by one in the prior art.

Technical solution

A first aspect of the embodiment of the present application provides a picture matching method, including:

Obtaining the shape feature information and the tone feature information of the target object in the source image, and generating a feature descriptor that represents the shape feature information and the tone feature information of the target object; the source image is at least one image;

And matching each of the to-be-processed pictures included in the target folder with the source picture according to the feature descriptor;

The to-be-processed image whose matching degree is greater than the threshold is published as the target image.

A second aspect of the embodiments of the present application provides a picture matching apparatus, including:

a feature generating module, configured to acquire shape feature information and tone feature information of the target object in the source image, and generate a feature descriptor that represents the shape feature information and the tone feature information of the target object; the source image is at least one image;

a matching module, configured to match each of the to-be-processed images included in the target folder with the source image according to the feature descriptor;

A publishing module is configured to publish a to-be-processed image with a matching degree greater than a threshold as a target image.

A third aspect of the embodiments of the present application provides a picture matching electronic device, including a memory, a processor, and the computer storing computer readable instructions executable on the processor, the processor executing the The computer readable instructions implement the following steps:

A fourth aspect of the embodiments of the present application provides a computer readable storage medium storing computer readable instructions, wherein the computer readable instructions are implemented by at least one processor The following steps:

Beneficial effect

In the embodiment of the present application, the shape feature information and the tone feature information of the target object in the source image are obtained, and a feature descriptor that represents the shape feature information and the tone feature information of the target object is generated, and the source image and the image are processed according to the feature description. The image is processed for matching, and the to-be-processed image whose matching degree is greater than the threshold is released as the target image, which can reduce the time required for selecting all the images of the target object one by one, and can reduce the operation of the user and improve the user experience.

DRAWINGS

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings used in the embodiments or the prior art description will be briefly described below. Obviously, the drawings in the following description are only the present application. For some embodiments, other drawings may be obtained from those of ordinary skill in the art without departing from the drawings.

1 is a schematic flowchart of a picture matching method according to an embodiment of the present application;

FIG. 2 is a flowchart of an implementation of step S101 in FIG. 1;

FIG. 3 is a flowchart of an implementation of step S102 in FIG. 1;

4 is a schematic diagram of an electronic device according to an embodiment of the present application;

FIG. 5 is a structural block diagram of a picture matching program according to an embodiment of the present application.

Embodiments of the invention

In the following description, for purposes of illustration and description However, it will be apparent to those skilled in the art that the present invention may be practiced in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the application.

In order to explain the technical solutions described in the present application, the following description will be made by way of specific embodiments.

Embodiment 1

FIG. 1 is a flowchart showing an implementation process of a picture matching method according to Embodiment 1 of the present application, which is described in detail as follows:

Step S101: Acquire shape characteristic information and tone feature information of the target object in the source image, and generate a feature descriptor that represents the shape feature information and the tone feature information of the target object; the source picture is at least one picture.

For example, the source image may be a specified image in a certain folder, and the specified image may be a picture of any angle of the target object, which is not limited thereto. The target object may be any object, animal, etc., and is not limited thereto. For example, the target object is a product such as a table or chair.

Referring to FIG. 2, in one embodiment, generating a feature descriptor that represents the shape feature information and the tone feature information of the target object in step S101 can be implemented by the following process:

Step S201, constructing a scale space image of the shape feature information and the hue feature information of the target object in the source picture, and detecting the feature points in the scale space image.

Among them, you can pass sift (Scale-invariant The feature transform (scale invariant feature transform) algorithm acquires the shape feature information and the hue feature information of the target object in the source image, and identifies the shape feature and the hue feature of the target object, thereby identifying the target object.

After constructing the shape feature information of the target object and the scale space image of the hue feature information in the source picture, the feature points in the scale space image are detected. In this embodiment, the feature points may be extreme points, but are not limited thereto.

In this step, the scale space image may be generated by convolving the source picture with a variable-scale Gaussian function. Then, a Gaussian difference function is used to convolve with the original image to generate a Gaussian difference image sequence. In the Gaussian difference image sequence, the maximum and minimum values of a plurality of pixel points such as the current scale and the adjacent scale of each current pixel and the neighborhood are compared, thereby obtaining an extreme point.

Step S202: Filter and locate each feature point in the scale space, and obtain a stable feature point that satisfies a preset condition.

The filtering condition may be set to filter and locate each feature point in the scale space, so that the feature points that do not meet the preset condition among the feature points detected in step S201 are removed, and the stable feature points are obtained.

Specifically, each feature point may be positioned to detect whether the feature point is an edge point. If the feature point is an edge point, the feature point is filtered out, otherwise the feature point is retained.

Step S203, setting a direction for each of the stable feature points, and generating a feature descriptor that represents the shape feature information and the tone feature information of the target object.

Specifically, a direction of each of the stable feature points may be set by using a gradient direction distribution characteristic of the neighboring pixels of the stable feature point, so that the stable feature point has rotation invariance.

The method for setting a direction for each of the stable feature points, and generating a feature descriptor for characterizing the shape feature information and the tone feature information of the target object is specifically: taking a predetermined size adjacent to each of the stable feature points The domain is used as a sampling window, and the relative orientation of the sampling point and the corresponding stable feature point is Gaussian weighted and then classified into the direction histogram to obtain the feature descriptor.

For example, a neighborhood of 16*16 is taken as a sampling window centering on each of the stable feature points, and the relative direction of the sampling point and the corresponding stable feature point is Gaussian weighted and then classified into a direction histogram of 8 bins to obtain 4 *4*8 128-dimensional feature descriptor.

Step S102: Match each of the to-be-processed pictures included in the target folder with the source picture according to the feature descriptor.

Referring to FIG. 3, in an embodiment, step S102 can be specifically implemented by the following process:

Step S301: Acquire feature descriptors of objects to be matched in each of the to-be-processed pictures.

In this step, the specific process of obtaining the feature descriptors of the objects to be matched in each picture to be processed is as follows:

Constructing a scale space image of the shape feature information and the tone feature information of the object to be matched in the to-be-processed image, and detecting feature points in the scale space image;

Filtering and locating each feature point in the scale space to obtain a stable feature point that satisfies a preset condition;

A direction is set for each of the stable feature points, and a feature descriptor that represents the shape feature information and the tone feature information of the object to be matched is generated.

The detailed process of each step above is not described herein again, and the processes of steps S201 to S203 can be referred to.

Step S302: Obtain two feature descriptors in the feature descriptors of the to-be-processed picture that are closest to the Euclidean distance of the feature descriptor of the source picture.

Each feature descriptor of the to-be-processed picture is a feature descriptor of the object to be matched in the to-be-processed picture. The feature descriptor of the source picture is a feature descriptor of the target object in the source picture. For each feature descriptor of a picture to be processed, the Euclidean distance between the feature descriptors of the target object in the source picture can be calculated according to the direction and position of the feature descriptor. Calculating the Euclidean distance between each feature descriptor of the object to be matched in the image to be processed to the feature descriptor of the target object in the source image, and then obtaining the feature descriptor between the image to be processed and the source image The two feature descriptors with the smallest Euclidean distance.

Step S303, determining, according to the acquired Euclidean distance relationship between the two feature descriptors of the to-be-processed picture and the feature descriptor of the source picture, determining the object to be matched and the source picture in each picture to be processed. The matching relationship between the target objects.

The two feature descriptors of the to-be-processed picture are a first feature descriptor and a second feature descriptor, respectively, and the first feature descriptor is first European between the feature descriptors of the source image. The distance is greater than a second Euclidean distance between the second feature descriptor and a feature descriptor of the source picture. The specified feature descriptor is a feature descriptor of the target object in the source picture. The step S303 may be specifically: when the ratio of the second Euclidean distance to the first Euclidean distance is less than a preset value, determining that the to-be-processed picture matches the source picture. In this embodiment, the preset value may be a value greater than 0.6 and less than 0.9. More specifically, the preset value may be a value greater than 0.6 and less than 0.7.

It can be understood that the preset value becomes smaller, and the number of matched feature points becomes smaller, but the stability becomes higher.

Step S103: Publish the to-be-processed picture whose matching degree is greater than the threshold as the target picture.

Specifically, the to-be-processed picture with the matching degree of the source picture that is greater than the threshold may be put into the target picture set, and after each picture to be processed is matched with the source picture, each target in the target picture set is obtained. The image is published to generate image information of the target object. The target picture is a to-be-processed picture whose matching degree with the source picture is greater than a threshold.

The image matching method obtains the shape feature information and the tone feature information of the target object in the source image, and generates a feature descriptor that represents the shape feature information and the tone feature information of the target object, and according to the feature descriptor, the target file Each of the to-be-processed pictures included in the folder is matched with the source picture, and finally the to-be-processed picture whose matching degree is greater than the threshold is released as the target picture, thereby reducing the time required for selecting all the pictures of the target object one by one. At the same time, it can reduce the user's operation and improve the user's experience.

It should be understood that the size of the sequence of the steps in the above embodiments does not mean that the order of execution is performed. The order of execution of each process should be determined by its function and internal logic, and should not be construed as limiting the implementation process of the embodiments of the present application.

Embodiment 2

Corresponding to the picture matching method described in the foregoing embodiment, FIG. 4 is a schematic diagram showing an operating environment of the picture matching program provided by the embodiment of the present application. For the convenience of explanation, only the parts related to the present embodiment are shown.

In the embodiment, the picture matching program 400 is installed and runs in the electronic device 40. The electronic device 40 can be a mobile terminal, a palmtop computer, a server, or the like. The electronic device 40 can include, but is not limited to, a memory 401 and a processor 402. FIG. 4 shows only electronic device 40 having components 401-402, but it should be understood that not all illustrated components may be implemented and that more or fewer components may be implemented instead.

The memory 401 may be an internal storage unit of the electronic device 40, such as a hard disk or memory of the electronic device 40, in some embodiments. The memory 401 may also be an external storage device of the electronic device 40 in other embodiments, such as a plug-in hard disk equipped on the electronic device 40, a smart memory card (SMC), and a secure digital device. (Secure Digital, SD) card, flash card, etc. Further, the memory 401 may also include both an internal storage unit of the electronic device 40 and an external storage device. The memory 401 is configured to store application software and various types of data installed in the electronic device 40, such as program codes of the picture matching program 400. The memory 401 can also be used to temporarily store data that has been output or is about to be output.

The processor 402, in some embodiments, can be a central processor (Central) A processing unit (CPU), a microprocessor or other data processing chip for running program code or processing data stored in the memory 401, such as executing the picture matching program 400 and the like.

The electronic device 40 can also include a display. The display may be an LED display, a liquid crystal display, a touch liquid crystal display, and an OLED (Organic) in some embodiments. Light-Emitting Diode, organic light-emitting diodes, etc. The display is for displaying information processed in the electronic device 40 and a user interface for displaying visualizations, such as an application menu interface, an application icon interface, and the like. The components 401-42 of the electronic device 40 communicate with one another via a system bus.

Please refer to FIG. 5 , which is a functional block diagram of a picture matching program 400 provided by an embodiment of the present application. In this embodiment, the picture matching program 400 may be divided into one or more modules, and the one or more modules are stored in the memory 401 and executed by one or more processors (this implementation) For example, the processor 402) is executed to complete the application. For example, in FIG. 5, the picture matching program 400 can be divided into an information acquisition module 501, a matching module 502, and a processing module 503. A module as referred to herein refers to a series of computer readable instruction instructions that are capable of performing a particular function, and are more suitable than the program to describe the execution of the picture matching program 400 in the electronic device 40. The following description will specifically describe the functions of the modules 501-503.

The information acquiring module 501 is configured to acquire the shape feature information and the tone feature information of the source image, and generate a feature descriptor that represents the shape feature information and the tone feature information of the target object; the source image is at least one image. The matching module 502 is configured to match each of the to-be-processed pictures included in the target folder with the source picture according to the feature descriptor. The processing module 503 is configured to publish the to-be-processed picture whose matching degree is greater than the threshold as the target picture.

Optionally, the information acquiring module 501 may be divided into a building unit 601, a processing unit 602, and a feature descriptor generating unit 603.

The constructing unit 601 is configured to construct the scale feature image of the source image and the scale space image of the tone feature information, and detect feature points in the scale space image. The processing unit 602 is configured to filter and locate each feature point in the scale space, and obtain a stable feature point that meets a preset condition. The feature descriptor generation unit 603 is configured to set a direction for each of the stable feature points, and generate a feature descriptor that represents the shape feature information and the tone feature information.

As an implementation manner, the feature description sub-generating unit 603 is specifically configured to: take a neighborhood of a preset size centering on each of the stable feature points as a sampling window, and compare the sampling point with the corresponding stable feature point. The direction is Gaussian weighted and then classified into the direction histogram to obtain the feature descriptor.

Optionally, the matching module 502 can be divided into an obtaining unit 701 and a matching determining unit 702.

The obtaining unit 701 is configured to acquire a feature descriptor of the object to be matched in each of the to-be-processed images, and obtain a feature identifier of the image to be processed that is closest to the feature descriptor of the source image. Two feature descriptors. The matching determining unit 702 is configured to determine, according to the acquired Euclidean distance relationship between the two feature descriptors of the to-be-processed picture and the feature descriptor of the source picture, each picture to be processed and the source picture Matching relationship between.

As an implementation manner, the two feature descriptors of the to-be-processed picture are a first feature descriptor and a second feature descriptor, respectively, and the first feature descriptor is to a feature descriptor of the source image. The first Euclidean distance between the two is greater than the second Euclidean distance between the second feature descriptor and the feature descriptor of the source picture.

The matching determining unit 702 is specifically configured to: when the ratio of the second Euclidean distance to the first Euclidean distance is less than a preset value, determine that the to-be-processed picture matches the source picture.

It will be clearly understood by those skilled in the art that, for convenience and brevity of description, only the division of each functional unit and module described above is exemplified. In practical applications, the above functions may be assigned to different functional units according to needs. The module is completed by dividing the internal structure of the device into different functional units or modules to perform all or part of the functions described above. Each functional unit and module in the embodiment may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit, and the integrated unit may be implemented by hardware. Formal implementation can also be implemented in the form of software functional units. In addition, the specific names of the respective functional units and modules are only for the purpose of facilitating mutual differentiation, and are not intended to limit the scope of protection of the present application. For the specific working process of the unit and the module in the foregoing system, reference may be made to the corresponding process in the foregoing method embodiment, and details are not described herein again.

In the above embodiments, the descriptions of the various embodiments are different, and the parts that are not detailed or described in the specific embodiments may be referred to the related descriptions of other embodiments.

Those of ordinary skill in the art will appreciate that the elements and algorithm steps of the various examples described in connection with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the solution. A person skilled in the art can use different methods to implement the described functions for each particular application, but such implementation should not be considered to be beyond the scope of the present application.

In the embodiments provided by the present application, it should be understood that the disclosed apparatus/electronic device and method may be implemented in other manners. For example, the device/electronic device embodiment described above is merely illustrative. For example, the division of the module or unit is only a logical function division. In actual implementation, there may be another division manner, for example, multiple units. Or components may be combined or integrated into another system, or some features may be omitted or not performed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in electrical, mechanical or other form.

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

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

The integrated modules/units, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium. Based on such understanding, the present application implements all or part of the processes in the foregoing embodiments, and may also be implemented by computer readable instructions, which may be stored in a computer readable storage medium. The computer readable instructions, when executed by a processor, may implement the steps of the various method embodiments described above. Wherein, the computer readable instructions comprise computer readable instruction code, which may be in the form of source code, an object code form, an executable file or some intermediate form or the like. The computer readable medium can include any entity or device capable of carrying the computer readable instruction code, a recording medium, a USB flash drive, a removable hard drive, a magnetic disk, an optical disk, a computer memory, a read only memory (ROM, Read-Only) Memory), random access memory (RAM, Random) Access Memory), electrical carrier signals, telecommunications signals, and software distribution media. It should be noted that the content contained in the computer readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in a jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, computer readable media Does not include electrical carrier signals and telecommunication signals.

The above-mentioned embodiments are only used to explain the technical solutions of the present application, and are not limited thereto; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still implement the foregoing embodiments. The technical solutions described in the examples are modified or equivalently replaced with some of the technical features; and the modifications or substitutions do not deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in Within the scope of protection of this application.

Claims

A picture matching method, which is characterized by comprising:

Obtaining the shape feature information and the tone feature information of the target object in the source image, and generating a feature descriptor that represents the shape feature information and the tone feature information of the target object; the source image is at least one image;

And matching each of the to-be-processed pictures included in the target folder with the source picture according to the feature descriptor;

The to-be-processed image whose matching degree is greater than the threshold is published as the target image.
The picture matching method according to claim 1, wherein the generating a feature descriptor that represents the shape feature information and the tone feature information of the target object comprises:

Constructing a shape space information of the target object in the source picture and a scale space image of the tone feature information, and detecting feature points in the scale space image;

Filtering and locating each feature point in the scale space to obtain a stable feature point that satisfies a preset condition;

A direction is set for each of the stable feature points, and a feature descriptor that characterizes the shape feature information and the tone feature information of the target object is generated.
The picture matching method according to claim 2, wherein the setting a direction for each of the stable feature points, and generating a feature descriptor that represents the shape feature information and the tone feature information of the target object is specifically:

A neighborhood of a predetermined size is taken as a sampling window around each of the stable feature points, and a relative direction of the sampled point and the corresponding stable feature point is Gaussian weighted and then classified into a direction histogram to obtain the feature descriptor. .
The picture matching method according to claim 2, wherein the matching between the to-be-processed pictures included in the target folder and the source picture according to the feature descriptor includes:

Obtaining feature descriptors of the objects to be matched in each of the to-be-processed pictures;

Obtaining, in the plurality of feature descriptors of the to-be-processed picture, two feature descriptors that are closest to the feature description sub-European distance of the source picture;

Determining a match between the object to be matched and the target object in the source image in each picture to be processed according to the Euclidean distance relationship between the obtained two feature descriptors and the feature descriptor of the target object in the source picture relationship.
The picture matching method according to claim 4, wherein the two feature descriptors of the to-be-processed picture are a first feature descriptor and a second feature descriptor, respectively, and the first feature descriptor is The first Euclidean distance between the feature descriptors of the source picture is greater than the second Euclidean distance between the second feature descriptor and the feature descriptor of the source picture;

Determining, according to the acquired Euclidean distance relationship between the two feature descriptors of the to-be-processed picture and the feature descriptor of the source picture, determining a matching relationship between each to-be-processed picture and the source picture. for:

When the ratio of the second Euclidean distance to the first Euclidean distance is less than a preset value, determining that the to-be-processed picture matches the source picture.
A picture matching device, comprising:

a feature generating module, configured to acquire shape feature information and tone feature information of the target object in the source image, and generate a feature descriptor that represents the shape feature information and the tone feature information of the target object; the source image is at least one image;

a matching module, configured to match each of the to-be-processed images included in the target folder with the source image according to the feature descriptor;

A publishing module is configured to publish a to-be-processed image with a matching degree greater than a threshold as a target image.
The picture matching terminal device according to claim 6, wherein the feature generation module comprises:

a feature detecting submodule, configured to construct a scale space image of the shape feature information and the hue feature information of the target object in the source image, and detect feature points in the scale space image;

a feature extraction sub-module, configured to filter and locate each feature point in the scale space, and obtain a stable feature point that meets a preset condition;

And a feature generation sub-module configured to set a direction for each of the stable feature points, and generate a feature descriptor that represents the shape feature information and the tone feature information of the target object.
The picture matching device according to claim 7, wherein the feature generation sub-module comprises:

A neighborhood of a predetermined size is taken as a sampling window around each of the stable feature points, and a relative direction of the sampled point and the corresponding stable feature point is Gaussian weighted and then classified into a direction histogram to obtain the feature descriptor. .
The picture matching terminal device according to claim 7, wherein the matching module comprises:

a first feature acquisition sub-module, configured to acquire feature descriptors of objects to be matched in each of the to-be-processed pictures;

a second feature acquiring sub-module, configured to acquire two feature descriptors in each feature descriptor of the to-be-processed picture that are closest to a feature descriptor of the source picture;

a matching determining module, configured to determine, according to the Euclidean distance relationship between the acquired two feature descriptors and a feature descriptor of the target object in the source image, the object to be matched in each to-be-processed image and the source image The matching relationship between the target objects.
The picture matching terminal device according to claim 9, wherein the two feature descriptors of the to-be-processed picture are a first feature descriptor and a second feature descriptor, respectively, and the first feature descriptor is The first Euclidean distance between the feature descriptors of the source picture is greater than the second Euclidean distance between the second feature descriptor and the feature descriptor of the source picture;

The matching determination module includes:

When the ratio of the second Euclidean distance to the first Euclidean distance is less than a preset value, determining that the to-be-processed picture matches the source picture.
A picture matching electronic device, characterized in that the identification processing electronic device of the potential customer comprises a memory, a processor having stored thereon computer readable instructions executable on the processor, the processor executing the The following steps are implemented when the computer readable instructions are described:

Obtaining the shape feature information and the tone feature information of the target object in the source image, and generating a feature descriptor that represents the shape feature information and the tone feature information of the target object; the source image is at least one image;

And matching each of the to-be-processed pictures included in the target folder with the source picture according to the feature descriptor;

The to-be-processed image whose matching degree is greater than the threshold is published as the target image.
The picture matching electronic device according to claim 11, wherein the generating a feature descriptor that represents the shape feature information and the tone feature information of the target object comprises:

Constructing a shape space information of the target object in the source picture and a scale space image of the tone feature information, and detecting feature points in the scale space image;

Filtering and locating each feature point in the scale space to obtain a stable feature point that satisfies a preset condition;

A direction is set for each of the stable feature points, and a feature descriptor that represents the shape feature information and the tone feature information of the target object is generated.
The picture matching electronic device according to claim 12, wherein the setting a direction for each of the stable feature points, and generating a feature descriptor that represents the shape feature information and the tone feature information of the target object is specifically:

A neighborhood of a predetermined size is taken as a sampling window around each of the stable feature points, and a relative direction of the sampled point and the corresponding stable feature point is Gaussian weighted and then classified into a direction histogram to obtain the feature descriptor. .
The picture matching electronic device according to claim 12, wherein the matching the to-be-processed pictures included in the target folder with the source picture according to the feature descriptor includes:

Obtaining feature descriptors of the objects to be matched in each of the to-be-processed pictures;

Obtaining two feature descriptors in each feature descriptor of the to-be-processed picture that are closest to the feature descriptor of the source picture;

Determining a match between the object to be matched and the target object in the source image in each picture to be processed according to the Euclidean distance relationship between the obtained two feature descriptors and the feature descriptor of the target object in the source picture relationship.
The picture matching electronic device according to claim 14, wherein the two feature descriptors of the to-be-processed picture are a first feature descriptor and a second feature descriptor, respectively, and the first feature descriptor is The first Euclidean distance between the feature descriptors of the source picture is greater than the second Euclidean distance between the second feature descriptor and the feature descriptor of the source picture;

Determining, according to the acquired Euclidean distance relationship between the two feature descriptors of the to-be-processed picture and the feature descriptor of the source picture, determining a matching relationship between each to-be-processed picture and the source picture. for:

When the ratio of the second Euclidean distance to the first Euclidean distance is less than a preset value, determining that the to-be-processed picture matches the source picture.
A computer readable storage medium storing computer readable instructions, wherein the computer readable instructions, when executed by at least one processor, implement the following steps:

Obtaining the shape feature information and the tone feature information of the target object in the source image, and generating a feature descriptor that represents the shape feature information and the tone feature information of the target object; the source image is at least one image;

And matching each of the to-be-processed pictures included in the target folder with the source picture according to the feature descriptor;

The to-be-processed image whose matching degree is greater than the threshold is published as the target image.
The computer readable storage medium of claim 16, wherein the generating a feature descriptor that represents the shape feature information and the tone feature information of the target object comprises:

Constructing a shape space information of the target object in the source picture and a scale space image of the tone feature information, and detecting feature points in the scale space image;

Filtering and locating each feature point in the scale space to obtain a stable feature point that satisfies a preset condition;

A direction is set for each of the stable feature points, and a feature descriptor that represents the shape feature information and the tone feature information of the target object is generated.
The computer readable storage medium according to claim 17, wherein the setting a direction for each of the stable feature points, and generating a feature descriptor that represents the shape feature information and the tone feature information of the target object is specifically:

A neighborhood of a predetermined size is taken as a sampling window around each of the stable feature points, and a relative direction of the sampled point and the corresponding stable feature point is Gaussian weighted and then classified into a direction histogram to obtain the feature descriptor. .
The computer readable storage medium according to claim 17, wherein the matching the to-be-processed pictures included in the target folder with the source picture according to the feature descriptor comprises:

Obtaining feature descriptors of the objects to be matched in each of the to-be-processed pictures;

Obtaining two feature descriptors in each feature descriptor of the to-be-processed picture that are closest to the feature descriptor of the source picture;

Determining a match between the object to be matched and the target object in the source image in each picture to be processed according to the Euclidean distance relationship between the obtained two feature descriptors and the feature descriptor of the target object in the source picture relationship.
The computer readable storage medium according to claim 19, wherein the two feature descriptors of the to-be-processed picture are a first feature descriptor and a second feature descriptor, respectively, and the first feature descriptor The first Euclidean distance between the feature descriptors of the source picture is greater than the second Euclidean distance between the second feature descriptor and the feature descriptor of the source picture;

Determining, according to the acquired Euclidean distance relationship between the two feature descriptors of the to-be-processed picture and the feature descriptor of the source picture, determining a matching relationship between each to-be-processed picture and the source picture. for:

When the ratio of the second Euclidean distance to the first Euclidean distance is less than a preset value, determining that the to-be-processed picture matches the source picture.